389
Views
0
CrossRef citations to date
0
Altmetric
Abstracts

Abstracts from CAS-H 2004, 2nd International symposium on computer aided surgery around the head, Bern, Switzerland september 17–19, 2004

Pages 181-196 | Published online: 06 Jan 2010

Session I: Computer assistance for Cranio-Maxillofacial surgery (CMF)

Computer-assisted skull base surgery

A. Schramm, N.-C. Gellrich, J. Schipper, J. Zenter, R. Laszig and R. Schmelzeisen

University Hospital Freiburg, Germany

The therapy of skull base lesions needs exact planning to show extension of the neoplasm and point out vital structures. Reconstruction following tumor resection needs reliable information also to choose correct type and volume of grafts and to predict the outcome. An optical navigation system was used for planning, navigation of tumor resection and reconstruction. The advantage concerning postsurgical outcome was investigated in the treatment of patients with tumors and ankylosis. Resection and reconstruction were preoperatively planned. Intraoperatively safety margins and contours of transplants were navigated. The postsurgical outcome was validated measuring orbital dimensions. Computer-assisted treatment was performed in 48 cases. Preoperatively outlined safety margins could be controlled during tumor resection. Reconstruction was designed and bone could be molded to the demands to cover and augment defects. Postoperative orbital dimensions showed significant decrease in orbital volumes and symmetrical orbital values. Computer-assisted treatment improves preoperative planning by combining CT and MRI data. Tumor volume before chemotherapy can be assessed and transferred to the data set after chemotherapy to perform radical resection inside the original margins. Intraoperative navigation makes tumor surgery more reliable by going for the safety margins, saving vital structures and leading the reconstruction to preplanned results.

Intraoperative navigation-guided surgical reconstruction in case of orbital bone defects and deformities

J. Hoffmann1, C. Westendorff1, J. Kaminsky2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery, University Hospital, Tübingen, Germany; and 2Department of Neurosurgery, University Hospital, Tübingen, Germany

Introduction: Secondary orbital or periorbital skeletal reconstruction remains a major surgical challenge. Recent use of high-resolution computed tomography data for preoperative planning using image guided surgery techniques offers an approach to an improved outcome of orbital reconstructive surgery.

Materials and Methods: Non-comparative interventional series of 10 consecutive orbital and periorbital procedures were scheduled for image-guided surgery by use of a wireless passive infrared light surgical navigation system (VectorVision™, BrainLAB).

Results: The series consisted of 6 cases of orbital wall fractures, 3 cases of orbital rim reconstruction, and 1 case of tumor resection. The image-guided technique was useful in providing preoperative planning and intraoperative surgical guidance in all cases. No complications were observed.

Conclusion: Image-guidance is a useful adjunctive tool in providing three-dimensional intraoperative anatomical information in a variety of orbital procedures. Preoperative planning facilities, virtual orbital wall reconstruction and transfer into the operating room provide very useful additional data for a well controlled treatment.

Automatic method for mid-facial symmetry plane extraction in computer-assisted CMF surgery

E. De Momi1, I.P.I. Pappas2, J. Chapuis2, A. Schramm3, W. Hallermann4 and M. Caversaccio5

1Biomedical Engineering Department, Politecnico di Milano, Italy; 2MEM Research Center, Bern, Switzerland; 3Department of Oral and Maxillofacial Surgery, University of Freiburg, Germany; 4CMF-Department, Inselspital, Bern, Switzerland; and 5ENT-Department, Inselspital, Bern, Switzerland

Extraction of the mid-facial plane is very important in the area of computer-aided cranio-maxillofacial surgery for the assessment of deformities and planning of reconstructive procedures. A possible way to define the mid-facial symmetry plane is by selecting at least three anatomical landmarks in the axial and coronal views of a CT dataset, but it is time consuming and the results are user-dependent. Moreover, in pathological cases key-landmarks may be missing or difficult to identify. We developed an automatic method for mid-facial plane extraction based on a 3D surface model of the segmented skull obtained from CT datasets. Homologous areas on the facial skeleton are selected, mirrored and matched using a surface registration algorithm. The resulting transformation matrix is used to compute the mid-facial plane. In this presentation, we compare the surface-based method with the landmark-based method on 12 different patient data and four different users. Results show that the method is reliable, fast and promising in particular in cases of severe malformations. As the human face is not perfectly symmetric there is no perfect symmetry plane. An advantage of the surface-based method is that it allows the user to specify on which areas the algorithm should focus on. The described algorithm is embedded in a computer-aided planning and navigation environment for CMF surgery.

Planning mandibular reconstructions by combined use of computer-assisted design tools and rapid prototype models

W. Hallermann1, S. Olsen2, T. Bardyn2, K. Smolka1 and T. Iizuka1

1CMF-Department, Inselspital, Bern, Switzerland; and 2MEM Research Center, Bern, Switzerland

Reconstruction using bone grafts are particularly challenging in cranio-maxillofacial surgery where both functional and aesthetic outcome determine the success of surgery. Not only is preoperative planning of great importance, but also ensuring its accurate execution during surgery is vital. Complex mandibular reconstruction requires several graft pieces to be arranged accurately to fill the curved defect.

A new methodology has been developed by the authors where a combination of computer methods and physical models, as a well traditional planning is fused to improve the surgical outcome. A 3D model of the patient's CT data is manipulated virtually by resecting and placing grafts in the desired correct location. A rapid prototype model (SLS) is then made from the ideal computer-aided plan. Next, mandibular plates are pre-bend to fit the model. Finally, the pre-bent plate is used for reconstruction to ensure a rapid and accurate implementation of the plan. A case study is presented to illustrate this new technique.

Robotic-assisted microsurgery

N. Singh and R. Katz

Johns Hopkins University School of Medicine, Baltimore, MD, USA

Craniofacial reconstruction often requires the use of free tissue transfer with microvascular anastomoses. These flaps may be combinations of bone, muscle, fascia, and skin in composite or chimerical configurations to restore and stabilize 3-dimensionally complex defects. The techniques of microvascular reconstruction are challenged by requirements of precision in anastomotic repair and are traditionally very invasive.

A porcine animal model was developed to test and advance microsurgery for vessel anastomoses of 1.5–2.0 mm diameter using the Da Vinci robot (Intuitive Surgical, Inc., Sunnyvale, CA). This technique has previously not been demonstrated. A randomized prospective trial was conducted, and demonstrated set-up time for operating microscope vs. Da Vinci robot were comparable (p < 0.05). Time for anastomosis completion were favorable with the robot than with traditional microvascular technique (p < 0.05).

All anastomoses remained patent (p < 0.02), with either technique, however, the advantage of improved dexterity, greater precision, and elimination of tremor in student surgeons was noted. Instruments can be rotated through 360° with 6 degrees of freedom, permitting anastomoses in deep operative fields and around external fixators. The robotic method opens the portal even further for minimally invasive microsurgery via smaller incisions.

Session II: Smart devices – part I

The Use of piezosurgery in maxillofacial surgery

A. Schramm, K.-H. Borman, N.-C. Gellrich and R. Schmelzeisen

University hospital Freiburg, Germany

The use of ultrasound based osteotomy in oral and maxillofacial surgery is rarely described. The advantage of this technique is the gentle removal of bone, without the danger of injuring the soft tissue. Especially, in the field of dental implantology there is a wide spread of indications, e.g. elevation of the maxillary sinus floor, apicectomies and bone-split operations. In our clinic we extended the range of this application to nerve lateralization in the posterior mandible, minimal-invasive-bone grafting from the chin or the retromolar area and sagittal osteotomy of the maxilla in orthognathic surgery. The aim of this presentation is to point out the advantages of ultrasound based osteotomy based on clinical cases and to analyze the histological findings in an animal study. Possibilities and indications of this innovative method will be presented.

Computer-assisted osteotomy with a pulsed Co2 laser

M. Ivanenko1, A. Afilal1, M. Klasing1, M. Werner1 and P. Hering1,2

1center of advanced european studies and research (caesar), Bonn, Germany; and 2University of Düsseldorf Germany

A new laser technique will be introduced, which allows non-contact cutting and 3D removal of bone and cartilage tissue without the usual thermal damage and with arbitrary sophisticated cut geometry. A fast and precise PC-controlled beam deflector guides the laser beam over the tissue, so that the procedure can be straightforwardly combined with a computer assisted navigation system. An additional advantage of the laser treatment is that it does not produce traumatic vibration nor any bone dust or metal abrasion. The absence of thermal side effects after the laser osteotomy has been confirmed in detailed histological studies.

A mobile prototype of the laser osteotome has been used in three series of animal trials. It makes very fine cuts (100–200 µm width) with depth up to 7 mm in compact tissue possible. A special irradiation technique allows much deeper incisions in multi-layer bones such as jawbone at the expense of a broader incision (0.8–1 mm). A typical time to cut out a trapeze shaped bloc of 20 mm length and 9 mm width from a 19 mm thick pig jaw amounts to 7 min.

Different possibilities to control the cut depth during the laser application will be also shortly discussed.

Surgical robot for craniotomies

S. Hassfeld1, R. Marmulla1, W. Korb1, J. Raczkowsky2, H. Wörn2 and G. Eggers1

1Department for Oral- and Craniomaxillofacial Surgery, University of Heidelberg, Heidelberg, Germany; and 2University of Karlsruhe, Institute for Process Control and Robotics, Karlsruhe, Germany

Aim: Within the framework of the Collaborative Research Centre (SFB 414) – Information Technology in Medicine: Computer and Sensor Aided Surgery – the surgical robot system RobaCKa was developed. It is dedicated for maxillofacial surgery, i.e. milling of the bony skull. In April 2003, the first patient was treated with the system.

Materials and Methods: The system consists of a ‘Staubli RX 90’ robot, the robot control system, an infrared-navigation-system (Polaris), a force-torque-sensor and an overload-protection. As surgical robots are complex mechatronic systems, it is important to apply systematic approaches for fault-free design, error detection and quality assurance. This includes besides risk analysis and proper design methods also the tests on phantoms to evaluate the accuracy of the system.

Results: In the first clinical trial mainly the workflow and performance of the system were tested. The system works safe and can be used intuitively within surgical environment. The accuracy can be given as an error of 0.66 ± 0.2 mm (maximum deviation is 1.06 mm).

Conclusion: It is necessary to implement risk management also in clinical research. The correctness of the concept and the final system setup was proofed by performing a verification protocol and phantom tests. Finally, the system was used for a patient trial in the department of maxillofacial surgery in Heidelberg. This trial was the first time, when a robot was used for milling complex trajectories, i.e. with relevant position and angulation changes, autonomously.

Session III: Computer assistance for dental surgery – part I

Pain experienced in implant placement with flapless surgery using an image-guided system

T. Fortin1,2, J.L. Bosson2, M. Isidori3 and E. Blanchet3

1Department of Oral Surgery, Dental University of Lyon, France; 2TIMC, Medical University of Grenoble, France; and 3Hospices Civils de Lyon, France

Purpose: To compare the pain experienced after implant insertion with a flapless surgical procedure using an Image Guided System based on a template and an open-flap procedure.

Materials and Methods: Thirty patients were referred for the placement of 80 implants and treated with a flapless procedure. Other group of 30 patients was referred for the placement of 72 implants with a conventional procedure. Patients were selected randomly. They were requested to fill out a questionnaire using a visual analogue scale (VAS) to assess the pain experienced and to indicate the number of analgesic tablets taken every postoperative day from the day of the surgery (D0) to 6 days after surgery (D6).

Results: The results showed a significant difference in pain measurements, with higher scores on the VAS with open-flap surgery (p < 0.01). Pain decreased faster with the flapless procedure (p = 0.05). The number of patients who felt no pain (VAS = 0) was higher with the flapless procedure (43% at D0 versus 20%). With the flapless procedure, patients took fewer pain tablets (p = 0.03) and the number of tablets taken decreased faster (p = 0.04).

Conclusion: With the flapless procedure, patients experienced less pain and for a shorter periods of time.

A cost- and effort-reduced concept of computer-aided implant dentistry

G. Widmann1, R. Widmann2, E. Widmann3, W. Jaschke1 and R. Bale1

1Clinical Division of Diagnostic Radiology I, Innsbruck, Austria; 2Dental Laboratory Czech, Absam, Austria; and 3Dental Practice Dr. Widmann, Zell/Ziller, Austria

Introduction: Although computer-aided implant dentistry (CAID) shows clear advantages over the conventional method, there is no widespread use. Costs and effort are still too high to establish CAID as a standard procedure. In order to overcome these drawbacks, we developed a novel concept of image guided template production (IGTP).

Materials and Methods: By virtue of a vacuum mouthpiece – and external reference frame based registration method the CT-scan of the patient is registered to the corresponding dental stone cast. Implants are virtually planned on the basis of the wax-up of the prosthetic restoration. Guided through a stereotactic aiming device, burr tubes are positioned and glued into prefabricated templates. A clinical case with free-end situation in Regio 44, 45 and 46 is presented.

Results: All implants could be positioned correctly without peri- and postoperative complications. The use of a consecutive second template enhanced the surgical procedure.

Conclusions: The existing equipment of a University hospital could be used with only minimal additional costs. There is no need for radiographic templates, reference templates and modifications of the standard casts. This concept allows for a wide spread use addressing to the private practice due to minimal effort at affordable costs.

Applying electromagnetic position sensing for navigation in dental implantology – a first analysis

D. Mucha1, B. Kosmecki1 and T.C. Lüth1,2

1Fraunhofer-Institute for Production Systems and Design Technology, Berlin, Germany; and 2Maxillofacial Surgery – Clinical Navigation and Robotics, Berlin, Germany

Introduction: This article presents a first analysis of applying electromagnetic position sensors to navigation in dental implantology. Electromagnetic sensors overcome some known limitations of optical sensors (i.e. visibility) but accuracy interferes with metallic items such as surgical instruments.

Materials and Methods: The setup consists of a dental handpiece and a reference plate both attached with electromagnetic trackers. 52 points were defined on the reference plate within a working volume of 10 × 10 × 10 cm3. Initially, the reference plate is registered with the working volume using a simple 3-point registration method. The navigated handpiece is then calibrated. During the course of the experiment every defined point was approached with the drill. The position recorded by the electromagnetic position sensor is compared with the actual position within the working volumes coordinate system.

Results: Positioning accuracy strongly depends from (1) the position and (2) the orientation of the navigated handpiece within the defined working volume. The applied method is not yet sufficient to achieve high accuracy at all positions within the working volume. Thus, applying electromagnetic position sensing with metallic instruments requires suitable calibration and registration methods to cope with the intrinsic properties of this interesting method of electromagnetic position sensing.

Session IV: Computer assistance for dental surgery – part II

Assessment of accuracy of navigation-guided dental implant socket drilling: An experimental study

C. Westendorff1, J. Hoffmann1, D. Troitzsch1, G. Gomez-Roman2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery, University Hospital, Tübingen, Germany; and 2Department of Prostodontics, University Hospital, Tübingen, Germany

Introduction: A correct placement of dental implants represents a limiting factor to both, biomechanical stability and sufficient functional and esthetic prosthodontic rehabilitation. CT-guided navigation technology is applied in an ever-increasing number of cases, but little is known about accuracy and outcome of image-guided implant dentistry.

Materials and Methods: In this study, 3D-accuracy of navigational-assisted implant socket drilling was assessed in an edentulous lower jaw model. The outcome was compared with the outcome of conventionally drilled sockets.

Results: A total of 224 drillings was assessed. Mean angular deviation in relation to the occlusal plane was reduced from 7.7° ± 5.2° in conventional approaches to 2.6° ± 2.2° with image-guidance (p < 0.01). Three versus zero perforations of the canal roof were seen in the conventionally drilled sockets.

Conclusion: The results strongly indicate a benefit from image-data-based navigation in the field of implant dentistry.

Accuracy in computer-aided implant dentistry: precision of a novel concept

G. Widmann1, R. Widmann2, E. Widmann3, W. Jaschke1 and R. Bale1

1Clinical Division of Diagnostic Radiology I, Innsbruck, Austria; 2Dental Laboratory Czech, Absam, Austria; and 3Dental Practice Dr. Widmann, Zell/Ziller, Austria

Introduction: Aimed at cost- and effort reduction in computer-aided implant surgery, a novel image guided template production (IGTP) technique was developed. We present the in-vitro accuracy of the obtained surgical templates.

Materials and Methods: Using an optical navigation system, surgical paths were planned on the CT-data of target pellet supported dental stone casts. By virtue of a vacuum mouthpiece – an external reference frame-based registration method, the plan was transferred into IGTP-templates under guidance of a stereotactic aiming device. Template guided drillings were evaluated on the postoperative CT-scan.

Results: The accuracy [xy] of the IGTP-templates was 0.48 ± 0.35 mm (maximum 1.2 mm) and [z] 0.25 ± 0.12 mm (maximum 0.6 mm).

Conclusions: The presented IGTP-technique is a precise means to plan and guide implant surgery. The accuracy is similar to published results of burr tracking and other IGTP-methods. For the secure use on patients, a safety distance of 1.5 mm is recommended to reliably avoid damaging vital anatomical structures. The fabrication of consecutive templates with rising burr tube diameter may enhance the clinical outcome.

Structural analysis applied to computer-assisted oral implantology

T. Bardyn1, W. Hallermann2 and S. Olsen1

1MEM Research Center, Bern, Switzerland; and 2CMF-Department, Inselspital, Bern, Switzerland

Osseointegrated implants are now widely used for treating partial or total endentulousness. Long-term clinical success requires functional connection between the titanium surface of implants and host bone to be developed and maintained. The gradual process of osseointegration is regulated in part by mechanical factors which influence the tissue differentiation process. Consequently, knowing in advance the expected primary implant stability and concomitant mechano-biological response of the bone-implant interface may predict long-term clinical outcome.

In the proposed method, finite element analysis of the peri-implant region is integrated with a planning tool for implant surgery. The purpose-made numerical solver allows information about the expected implant behavior to be presented to the surgeon in real-time during the planning process. The solver has been proven fast enough for interactive use and will form a solid and repeatable platform for clinical studies.

Preliminary in vitro experiments were performed on porcine mandibles and compared with results of simple axial analysis. This study showed the feasibility of this methodology. Further laboratory experiments are being conducted in order to accurately calibrate the structural analysis to ensure good accuracy of the predicted micromotion. Next, the ability of the analysis methodology to predict long-term clinical results will be assessed by prospective studies. This research is the first step towards a planning tool which would be able to predict the outcome of surgery.

Computer-assisted surgical management and incision-less procedures: Technical report

M. Isidori1 and T. Fortin2

1Implant Dentistry, Hospices Civils de Lyon, France; and 2Department of Oral Surgery, Dental University of Lyon, France

The final outcome of an oral rehabilitation with implants may be improved by the congruence between the location and direction of implants with the planned prosthesis (Belser et al., 2000; Mericske et al., 2000). A specially designed mechanical tool has been elaborated to transfer the preoperative implant axis planned an a 3-dimensional software (Fortin et al., PRAXIM®) into a surgical template by a numerically controlled drilling machine and to allow a very precise implant placement in relation to the final prosthetic and anatomical requirements. So using such a surgical template reduces the surgical invasiveness: flapless surgery or incision-less surgery (LaMI). A rigorous protocol allows preparing temporary or final prosthesis prior to surgery for possible immediate loading. The technology's benefits; (1) decrease of surgical outcome, bleedings, pains and tumefactions; (2) maintenance of high osseous vascularization, particularly important for high osseous density and in the immediate loading; (3) maintenance of the gingiva and the top of the osseous ridge; it is very important in esthetic because we can very precisely place the implant compared with the gingival form (depth); (4) incision-less tissues, keeping the tissue potential; (5) less time consuming, very easy surgery and prevention of the flap rip, as well as the requirements will be described. The CAD-Implant® protocol may promote the use of a flapless surgical technique for a very accurate implant placement. Several clinical cases are presented, particularly implant placement in the mandible with immediate temporary prosthesis, one case of early immediate loading (3 days) and certainly the most benefit for our patient, two cases of immediate loading of definitive prosthesis.

Session V: smart devices – part II

Computer- and robot-assisted surgery at the lateral skull base can be improved by sensory feedback

M. Maasen1, F. Dammann2, E. Schwaderer2, J. Stallkamp3 and D. Malthan3

1Department of Otolaryngology, University of Tübingen, Tübingen, Germany; 2Department of Diagnostic Radiology, University of Tübingen, Germany; and 3Fraunhofer-Institute for Manufacturing Engineering and Automation IPA, Stuttgart, Germany

Computer-assisted surgical systems are expected to provide an essential improvement of both efficiency and precision for interventions at the lateral skull base in the nearer future. We developed a computer- and robot-assisted surgical system (ROBIN) that integrates sensory feedback. The development efforts are based upon a PC-based platform, utilizing the available libraries ‘Visualization Toolkit’ (VTK), ‘Insight Segmentation and Registration Toolkit’ (ITK) and ‘Apache XML-Parser’ (Xerces). In order to realize a both flexible and dependable CAS/RAS-system, a modular and distributed control and communication architecture was realized. The system employs different sensor systems, like force sensors, laser triangulation sensors and auto fluorescence sensors for tissue differentiation, in order to realize quality assurance and documentation steps. The integration of sensory systems feeding back in-situ information allows for an employment of the robot system in a so-called closed loop control architecture. Results will be presented that the feedback of sensory information improves the safety and the precision of robot-assisted surgical procedures in a surgical utility laboratory.

 This work was supported by a grant of the German Research Foundation (DFG) to M.M.M. (No. MA 1458/2).

A motorized injection device with haptic feedback for advanced cement augmentation

M. Löffel1, I.P.I. Pappas1, P. Heini2, T. Steffen1,3 and L.-P. Nolte1

1MEM Research Center, Bern, Switzerland; 2Orthopaedics Department, Inselspital, Bern, Switzerland; and 3Department of Biomedical Engineering, McGill University, Montreal, Canada

In certain surgical interventions on the spine, such as Vertebroplasty and spinal disc nucleus replacement, it is required to inject highly viscous fluids into bony or cartilaginous structures. The higher the viscosity of the injected cement is, the lower is the risk of cement leakage. Currently, the injection is performed either with traditional syringes, which suffer from low achievable pressure, or by using screw-type systems, which eliminate tactile information.

 To overcome these shortcomings, a hand held, computer-controlled, motorized injection device for viscous fluids, in particular bone cement, was developed. The re-usable and autoclavable device can generate pressures of up to 3 MPa in a standard 20 ml syringe and provides tactile feedback based on the applied force. The device offers the flexibility to implement various injection protocols such as constant or variable volume flow or pressure, including the capability of ensuring an immediate flow stop.

This setup provides the technological means to reduce cement leakage and it is a useful tool for standardizing injection procedures. First experimental results are encouraging and show that this device is able to render cement injection procedures more controllable and repeatable.

Navigated control evaluation of the surface accuracy of a navigated and power-controlled high-speed drilling system

M. Kneissler1, M. Strauss1 and T.C. Lüth1,2

1Fraunhofer-Institute for Production Systems and Design Technology; and 2Maxillofacial Surgery – Clinical Navigation and Robotics

Purpose: Drilling of the skull base bone without damaging the important inside structures is very difficult even with the help of the anatomical landmarks. However, the accuracy is based on the surgeon's experience and knowledge of the anatomy. The new concept is to combine a position measurement system (navigation system) with a high speed drill whose power can be controlled depending on the relative distance between drill and anatomy. The surgeon holds and guides the instrument i.e. drill manually as usual.

Materials and Methods: To investigate the accuracy experiments were performed. (1) Different geometries (plane, cylinders, and free form surfaces) were planned within a test phantom. (2) The geometries were exported to 3D surface models. (3) The target geometries were shaped in convex and concave manner using the navigated control drill. (4) The accuracy of the resulting shapes were measured using a mechanical position sensor (Faro Arm, Bronze).

Results: The accuracy of the drilling with regards to the planned shapes was measured below 1 mm. This approach can be used in different clinical applications to accomplish difficult drilling tasks very precisely and in a shorter period of time.

Session VI: Neurosurgery

Diffusion tensor imaging-based neuro-navigation

C. Nimsky1, O. Ganslandt1, P. Grummich1, A.G. Sorensen2 and R. Fahlbusch1

1Department of Neurosurgery, University of Erlangen, Erlangen, Germany; and 2Department of Radiology, Massachusetts General Hospital, Boston, MA, USA

Objective: To integrate diffusion tensor imaging (DTI) into navigational datasets for intraoperative visualization of major white matter tracts.

Methods: A single-shot spin-echo diffusion weighted echo planar imaging sequence on a 1.5 T MR scanner was used for DTI. Color-encoded fractional anisotropy (FA) maps of the principal eigenvector rendered as a boxoid within each voxel were used for segmentation of the pyramidal tract. The segmented images were rigidly registered with a T1-weighted gradient echo 3D dataset for navigation in 16 patients with gliomas.

Results: The whole DTI processing lasted about 25–30 min. In all cases, DTI could be integrated into the navigational dataset. Accuracy measured as target registration error was 1.2 ± 0.46 mm. Registration of FAmaps with the 3D navigational dataset was possible with an error of less than 2 mm. Co-registration with fMRI was consistent with DTI data. We observed a neurological deterioration only in one patient (6.3%).

Conclusions: DTI can be reliably integrated into navigational datasets. Microscope-based neuronavigation can be used for an intraoperative visualization of major white matter tracts. However, a possible shifting of these tracts during surgery has to be taken into account after major tumor parts are removed.

A new method for fluoroscopic-based neuro-navigation

B. Kühn1, O. Süss1, S. Schönherr2 and M. Brock1

1Department of Neurosurgery, Charité Universitätsmedizin Berlin, Germany; and 2Department of Informatics, Free University Berlin, Germany

For the optimization of anatomical orientation and improvement of surgical accuracy, neuronavigation is a useful tool in spinal surgery, e.g. for the exact positioning of pedicle screws for dorsal fixator implantation.

 In this research project a fluoroscopic-based neuronavigation system was developed using new minimized reference sensors fixed on a vertebral body. A phantom with a three-dimensional pattern of steel balls is mounted on the reference sensors to detect the sensor position on the fluoroscopic images by a tracking system. The special pattern on the phantom and a new software algorithm allows the navigation system to calculate the position of the sensor. After the acquisition of the fluoroscopic images the phantom can be removed from the sensor. Then the position of a pointer instrument is continuously measured by the tracking system and displayed in relation to anatomical structures with the aid of the reference sensor.

 Preclinical tests on the accuracy of the system were employed and first clinical experiences were gained during several surgical procedures.

 The fluoroscopic-based neuronavigation system with implanted reference sensors enables computer-animated, real-time visualization of the position of the pointer instrument and allows the continuous orientation on intraoperatively acquired X-ray images.

The use of navigation to improve the handling of a robot system

T. Dannemann1, K. Bumm2, J. Wurm2, H. Iro2 and W.A. Kalender1

1Institute of Medical Physics, University of Erlangen-Nuremberg, Germany; and 2Department of Otorhinolaryngology, University of Erlangen-Nuremberg, Germany

Introduction: Navigation systems are already part of the clinical routine in ENT and neurosurgery. Active robot systems, however, are still in an early stage of development and have been used only during a few head operations although they have special advantages regarding accuracy and stability. The main reasons are, among others, the cumbersome handling and the enormous safety requirements. To reduce these disadvantages we developed a system which combines navigation and robotics. The concept of our system and first results of anatomical experiments will be presented.

Materials and Methods: A navigation system (POLARIS, Northern Digital Inc.) has been integrated into a robot system (RV-1A, Mitsubishi Electric) to perform an automatic patient registration, instrument calibration, patient motion compensation and an intraoperative navigated re-planning of the robot trajectory as well as a monitoring of the robot motion.

Results and Conclusions: First phantom and anatomical experiments confirmed the improved handling of the robot system due to the integration of the navigation. Especially, the automatic patient registration proved to be very helpful and time-saving. Intraoperative navigated re-planning opens numerous possibilities for the robot system. However, the range of applications for the re-planning capabilities and the overall accuracy still have to be evaluated.

Guided surgery in reconstruction procedures of the skull base with cad–cam or titan mesh

J. Schipper1, N.-C. Gellrich2, A. Schramm2, I. Arapakis1 and W. Maier1

1Department of ORL & HN Surgery, Freiburg University, Freiburg, Germany; 2Department of OMF Surgery, Freiburg University, Freiburg, Germany

Purpose: CAD–CAM as well as titan mesh are suitable for reconstruction procedures in anterior skull base surgery. With the help of CAS it is possible to design presurgically hard tissue substitutes by simulation. The presurgical simulation and design procedure for the hard tissue substitute in the anterior skull base saves time and protects from unexpected situations.

Methods: We compare both reconstruction procedures in anterior skull base surgery with regard to the accuracy, reliability and practicability. A retrospective analysis in several cases of patients who underwent extended anterior skull base procedures was done.

Results: We found, that both reconstruction procedures are accurate, reliable and practicable by use of CAS. Of course the costs of CAD–CAM reconstruction procedure are much higher than all other reconstruction strategies in skull base surgery. Both systems are available for a one or two step reconstruction procedure. For a well planed one step reconstruction procedure a presurgical simulation of the approach and a guided surgical removal is useful.

Conclusions: Despite the high costs of the CAD–CAM reconstruction procedures in anterior skull base surgery there are some indications for this. In contrast to the much cheaper titan mesh system, the CAD–CAM implant is more stable. The impact of the stability and the possibility of a less distortion may be of interest more at the anterior than of the lateral skull base in regard to traumas in future. For both reconstruction procedures the use of CAS is mandatory.

The first clinical experiences with electromagnetically tracked image-guided brain surgery

T. Beems

University Medical Centre, St Radboud, Nijmegen, The Netherlands

Image Guidance is more and more incorporated in brain surgery to enhance accuracy in all kind of procedures.

 The Image Guidance Systems have been improved in the last years but in the clinical setting neurosurgeons still encounter difficulties. One problem is the need for skull fixation with a skull clamp, because the reference frame for the optical tracked systems cannot be fixed to the skull itself and the skull clamp is needed for a tight connection between patient and reference frame. This especially creates difficulties in pediatric neurosurgery because skull clamps cannot be used in this population. Another problem is the line of sight problem using optical tracking. It can be difficult to avoid interpositioning during an operation. Both of these problems can be avoided using electromagnetic tracking instead of optical tracking. The author presents his first, very satisfying, experiences with an electromagnetic tracked Image Guided System (Treon Axiem, Medtronic SNT) in different types of brain operations. Using this modality also makes it easier to use advanced contour registration as the modality for registration.

Session VII: Virtual reality and surgical simulation

Multi-user interaction with medical imagery in a real 3d-environment

F. Kral1, A. Mehrle2, G. Özgür3, R. Kikinis4 and W. Freysinger1

1Medical University Innsbruck, Innsbruck, Austria; 2Johannes Kepler University, Linz, Austria; 3University of Innsbruck, Innsbruck, Austria; and 4Harvard Medical School, Boston, MA, USA

Digital radiological imaging, integration with computer-aided surgery (CAS), and the routine creation of ‘three-dimensional’ models of complicated anatomical findings for surgical planning has become a standard. The appropriate three-dimensional presentation of these data is still missing. To overcome hitherto used stereoscopic visualization, we use the CAVE, a multi-user three-dimensional virtual reality. Simultaneous projections for the left and right eye on three walls plus floor, switching by shutter glasses realizes stereopsis. Cyber gloves and space-mice allow manipulations like zooming, rotating, highlighting etc. in space. Models for clinics (surgical planning and simulation) from digital imaging modalities. For education e.g. petrous bone anatomy a very high resolution data set with isotropic resolution of 160 µm is available. We assume, that the CAVE will train the surgeons' mental eye more efficiently. By interacting with real three-dimensional entities the implementation of CAVE in a surgical environment will flatten the learning curve of visualizing complicated three-dimensional findings from radiological medical imaging data mentally.

The creation of a computerized simulator for interventional endovascular procedures – purpose and results

A. Schäffer1, S. Petra1, N. Fisher2 and I. Mazor2

1PolyDimensions GmbH, Bickenbach, Germany; and 2Simbionix USA Corporation, Cleveland Ohio

Purpose: Carotid endarterectomy (CEA) is one of the most frequently performed vascular intervention in the United States, carotid angioplasty with stenting (CAS) the less invasive approach to solve the problem of carotid stenosis. They are highly demanding procedures requiring the mastering of special skills. The current system for education and training lacks skilled instructors, standardized training programs, and faces problems as time efficiency, high costs and patient safety.

We describe the creation of a computerized simulator for interventional endovascular procedures, including training programs, practice, and evaluation.

Methods: Anatomical correctness was reached through the creation of the 3D-model sets based on real patient CT/MRI images. They got adapted to certain needs through specific, unusual combination and modification of various software programs. To generate the X-ray images with contrast agent flow in real-time, innovative software was developed, that also shows vital signs, EKG leads, non-invasive blood pressure and aortic blood pressure.

Results: The ANGIO Mentor TM trains angiography, angioplastic procedures. It utilizes a high-fidelity haptic system together with virtual angiographic suite. It enables to practice a full procedure including manipulation of numerous devices like guide wires, catheters, balloons, into the arterial system, manipulation of the C-arm and images. In addition, vital signs, hemodynamic management and virtual neurological examination enhance the decision making process.

Projector-based computer-assisted surgery

R. Marmulla1, H. Hoppe2, J. Mühling1 and S. Hassfeld1

1University of Heidelberg, Heidelberg, Germany; and 2University of Karlsruhe, Karlsruhe, Germany

With a new navigation system, tumor margins and osteotomy lines can be projected directly onto the surgical site.

 A scanner is mounted above the operating table, which detects the surface of the surgical site intraoperatively. Osteotomy lines and tumor margins are projected onto this surface using an intense beamer. The proceeding is shown for different patients with tumors that are operated through a coronal approach.

 Surgical data can be registered projected with an accuracy of 1 mm. Pointers are not required to transfer the data on the surgical site. As the information is projected directly into the surgical site, no monitor had been necessary either, and the surgeon did not have to alternatively watch a monitor and the patient.

 Ergonomics are especially improved during coronal approaches.

A haptic guidance tool for ct-directed percutaneous interventions

E. Hagmann, P. Rouiller, P. Helmer, S. Grange and C. Baur

EPFL, Lausanne, Switzerland

We have developed a new navigation approach for computer-assisted interventional radiology. Our system combines a virtual reality display with high-fidelity haptic rendering to provide assistance and guidance of the medical gesture. Specifically, the system is designed to improve the accuracy of blind needle placement within tissues. The proposed technique actively helps the surgeon while keeping him in control of the procedure. We have recently developed an experimental setup for CT-guided biopsy. The setup features a high-precision haptic device connected to the biopsy needle, combined with a graphical interface. The haptic system guides the surgeon's hand to the target tissue based on CT data, whereas a real-time, graphical visualization of the tool trajectory provides navigation information. The setup requires rigid registration of the patient with respect to the haptic interface. Tests have been performed in the presence of radiologists to validate the proposed concept, and early results show that the system is easy to use and requires little training. We are planning to conduct clinical testing in the near future to quantitatively assess system performance.

Combining the advantages of the layer manufacturing technologies (lm) with the medical need for models

R. Schindel and G. Levy

FHS Hochschule für Technik, Wirtschaft und Soziale Arbeit, St. Gallen, Switzerland

In the last 10–15 years more then 30 different Layer Manufacturing Technologies (LM) based on different principles were developed. The most promising ones have been developed further and are used for prototypes, parts or models for presurgical planning, some fit into small market niches and some have just not survived. Two LM with a wide range of use, a big market segment and a high potential in R&D are the Selective Laser Sintering (SLS) – and the 3D Printing Processes (DoD). The main advantage of SLS is its big potential in processing various powder materials to 3D parts. The advantage of the 3D-Printing process is its resolution and its way to produce parts with model and support material. This presentation gives an idea of the potential of the LM Technologies for the production of bone structures and scaffolds. Spongiosa geometries have been 3D-printed in 4∶1, 2∶1 and 1∶1 scales. Out of a big range of different biomaterials RPD made SLS trials with poly [R]-3-hydroxybutyrate-co-3-hydroxyvalerate (PHB/HV). Direct out of Computer Tomography data a customized implant was designed and produced with the SLS technology in a two step procedure.

A new navigated image viewer with an integrated position sensor

S. Weber and T.C. Lüth

Berlin Center for Mechatronical Medical Devices, Charité & Fraunhofer IPK, Berlin, Germany

Introduction: This article presents ongoing work on the navigated image viewer. The system consists of a compact and mobile miniature screen and a position sensor. A new type of medical image viewing became possible by relating the actual displayed medical image data to the spatial relationship between the viewing device and a patients anatomy.

Materials and Methods: In the past, the viewing device applied an optical position sensor (Polaris, NDI) to track both the device and a patient through retro-reflective localizers. The new system has now an integrated miniature stereo video camera to track a patient's position. Tracking of the image viewer is thus obsolete. The system is connected to a control unit (i.e. laptop) to import the image data and render video output. Owing to the physical integration of the position sensor into the viewing system the technical overhead is further reduced.

Results: Integration of the viewing device and position sensor leads to a very compact and easy to apply viewing system that can easily be used in clinical routine. Different viewing modes are applied to cope with certain needs of a specific medical field.

Session VIII: Computer assistance for ent surgery – part I

Resection of large petroclival meningioma using neuro-navigated temporal bone drilling

T. Van Havenbergh

Department of Neurosurgery, University Hospital Antwerp, Antwerp, Belgium

Aims: To present the clinical application of neuronavigated temporal bone drilling in petroclival meningioma.

Materials and Methods: In a patient presenting with hearing loss the diagnosis of a giant petroclival meningioma was made. Resection was performed through a combined subtemporal, retrosigmoid and transpetrous approach with conservation of the inner ear structures. The temporal bone drilling was assisted by neuronavigation using bone fixed fiducials.

Results: A resection of > 90% could be achieved with tumor residue in the cavernous sinus. The patient had a favorable outcome with no additional cranial nerve deficit.

Conclusions: Cranial base surgery can profit from neuronavigation assisted temporal and skull base bone drilling by increasing the safety of the transpetrous and trans skull base approaches.

Evaluating the operative accuracy of the ‘a73’ robot system

K. Bumm1, C. Dohr1, T. Dannemann2, C. Nimsky3, H. Iro1 and J. Wurm1

1University of Erlangen, ENT Department, Erlangen, Germany; 2CAS-Innovations GmbH, Erlangen, Germany; and 3University of Erlangen, Neurosurgery Department, Erlangen, Germany

At the ENT Department in Erlangen, we developed the robotic system ‘A73’ for endoscopic skull base surgery. Here we describe the new and improved robotic setup as well as first tests on its operative accuracy.

 In contrast to the previous ‘all in one’ version, we built a modular three component setup. The robot is placed on a pedestal and can be immobilized with the operating table by two mechanical arms. A camera is placed on a mobile tripod. The control module is connected to both via one cable each. Basic feature of the computer-navigation system is the commercially available CAPPA station for ENT surgery. The system references by automatically detecting a referencing frame (SIP-lab frame) mounted on a non-invasive upper jaw vacuum mouthpiece. Initially the operative accuracy was assessed only by relying on given hardware features and visual control. For better surveillance and increased safety, we decided to evaluate the robots accuracy by means of redundant navigational control on a phantom head for frontal and lateral skull base surgery. Multiple measurements were taken from 46 CT markers, representing the operating field of both frontal and lateral skull base. Results from telemanipulatory as well as automated performances were compared with manually obtained navigational accuracies.

Computer-assisted surgery for the skull base

C. Martin1, M.D. Dubois1, D. Desserée2, S. Schmerber3, J.P. Lavieille3 and J.M. Prades1

1CHU Saint-Etienne, France; 2LIGIM Lyon, France; and 3CHU Grenoble, France

Our aim was to develop a computer-aided surgery software for the skull base and to test it on cadavers. This has been done in the framework of the collaboration between different academic actors. Its characteristics consist of: (i) the ability to navigate on IRM or TDM images or both at the same time. The production of these images is done classically with no markers and no frame. The fusion of images during the pre-operative planning is made automatically; (ii) the ability to realize a double matching, on skin and on bone, is possible.

 The rigid bodies perfectly meet the requirements of the oto-neurosurgery and its particular environment (microscope, size of the instruments). It answers to a set of constraints: to localize the patient's head and to match the IRM and TDM images.

 A pre-operative step consists in preparing the images for the surgery. Surgery starts with the fixation of the rigid body on the patient's head. Calibration of the tool is realized. Matching is achieved by surface digitization (skin, bone). Accuracy is controlled on relevant anatomical structures. The surgeon is able to navigate in real-time on both CT and MRI images. Accuracy is of the sub-millimeter order when we point the noble elements: semi-circular canals, cochlea, and so on.

Session IX: Soft tissue modeling

Advances in 3D osteotomy planning with 3D soft tissue prediction

S. Zachow, B. Erdmann and P. Deuflhard

Zuse-Institute-Berlin (ZIB), Germany

The planning of complex osteotomies with bone relocations on a computerized model of a patient's head, including a reliable three-dimensional prediction of the postsurgical facial appearance is a highly appreciated feature in cranio-maxillofacial surgery. We present the current status of our planning environment, comprising truthful 3D model reconstruction from tomographic data, 3D cephalometric analysis, interactive 3D osteotomy planning on polygonal skull models via draw and cut, exact placement of mobilized bone segments with collision control, fusion of digitized dental plaster casts with CT-data for an optimal assessment of dental occlusion, and a 3D soft tissue prediction on the basis of an individual volumetric tissue model with histomechanical parameters that have been exemplarily evaluated with postoperative CT-data. Our planning environment has been successfully utilized in more than 25 cases, ranging from mandibular and midfacial hypoplasia to complex hemifacial microsomia. A collaborative planning mode enables us to let the surgeons concentrate on osteotomy planning and bone relocation only. Our conclusion is that a thorough planning with 3D soft tissue prediction not only enhances mental preparation, but also patient information, quality assurance, and documentation. An assortment of results can be found under http://www.zib.de/visual/projects/cas.

Investigating facial animation for post-operative assessment

N. Magnenat-Thalmann, S. Garchery and L. Egger

MIRALab–University of Geneva, Geneva, Switzerland

A 3D reconstruction of the patient face gives surgeons the tools to perform a virtual operation and immediately see the results. Currently, the evaluation of surgery outcome is limited to a static snapshot of the result. However, the face is not a static shape and an accurate evaluation should include a dynamic evaluation of the reconstructed face performances, including dynamic facial expressions and speech capabilities.

 MIRALab (University of Geneva) has more than 15 years of experience in facial animation including face modeling and real-time animation on different platforms. The facial animation engine uses MPEG-4 Facial Animation Standard that allows to precisely animate a complex mesh with a minimum number of parameters while still keeping a certain degree of freedom and realism.

 One possible direct application of this method is to visually simulate some regional deficiencies in the face, such as for instance facial hemiplegia, local malfunctions of muscles, and so on. Another application could be to visualize how surgery can influence facial animation and speech capabilities.

 An optical motion capture system like the VICON can be used to record and analyze the influence of a specific facial disability/surgery on a facial animation.

Marker-based automated generation of facial models

G. Eggers1, C. Brazanji2, T.C. Lüth3, R. Marmulla1 and S. Hassfeld1

1University Hospital, Heidelberg, Germany; 2Mannheim University of Applied Sciences, Germany; and 3University Hospital, Berlin, Germany

Standard method of quality control for the outcome of maxillofacial surgery is the comparison of pre- and post-operative photograph series. However, this does not allow an objective comparison of the result as there are only 2D representations of the patient. Laser surface scanners are a fast and accurate means for objective shape assessment in three dimensions. However, multiple scans from different directions are necessary to create accurate models. The assembly of these scans manually or by surface matching is tedious and not easily performed by the surgeon. Based on the automated detection of reflective markers, a new method is proposed to facilitate and automate the creation of face models from multiple laser scans. The paper presents first experimental results that show the suitability of the method for clinical routine.

Topometry of the face using pulsed holography

S. Hirsch1, S. Frey1, A. Thelen1, J. Bongartz2 and P. Hering1

1Stiftung caesar, Bonn, Germany; and 2Rhein-Ahr Campus, Remagen, Germany

We introduce a novel approach to precise facial topometry using pulsed holography. With a single pulse of a Nd:YLF laser we record the facial surface of the patient's head in a 270° range. The short exposure time of 35 ns allows for high resolution 3D facial measurement without movement artifacts. The mobile camera captures holograms on photosensitive material. The holograms are digitized in a second self-contained unit, based on tomographic scans of the optically reconstructed real image of the hologram. Numerical data processing leads to 3D computer models of the face with intrinsic gray scale texture.

Pulsed holography for 3D capturing of soft tissue surfaces

K. Schwenzer1, L. Kovacs2, T. Seewald1, P. Hering3, R. Sader1 and H.-F. Zeilhofer1

1Clinic for Cranio-Maxillofacial Surgery, University Hospital, Basel, Switzerland; 2Division of Plastic Surgery, University of Technology, Munich, Germany; and 3caesar – center of advanced european studies and research, Bonn, Germany

Introduction: The complex maxillofacial area requires adequate three-dimensional data acquisition (analysis/operation planning). Latest developments for 3D-surface-capturing are based on pulsed holography. For optimal data without detrimental application the following is necessary: contact-free acquirement, high resolution (<mm), short input-time.

Materials and Methods: 3D-data (validated test specimens, plaster models, test persons (n = 20) patients) were acquired by holography: digitized, and analyzed in comparison to white-light- and laser-scanners. Holography uses a laser pulse (25 ns) split into two beams: The ‘object beam’, scattered by the patients face, creates an interference pattern with the ‘reference beam’. The ‘real image’ of the hologram was digitized by ‘holographic tomography’.

Results: Concerning input-time holography is superior to all other procedures (white-light > 1 s; laser: 2 s–5 min; holography 25 ns) and optimal for uncooperative patients. The comparison of the investigated systems showed local sub-millimeter resolution for all systems in static objects. For moving objects validity and reliability change in range of millimeters, depending on the acquisition time. Another advantage of the holography is the single shot acquisition avoiding artifacts by changing the patient's position.

Conclusion: Holography, with its short recording time and high-resolution surface data appears to be a promising method, certain deficits still exist concerning data reduction and digitization.

Session X: Computer assistance for ent surgery – Part II

Continuous automated patient tracking for cas with ultrasound

G. Diakov and W. Freysinger

ENT-Clinic of the Medical University of Innsbruck, Austria

To overcome the central source of errors in navigated (ENT) procedures, intraoperative patient-to-image-registration, we present initial experimental results of ultrasound signal acquisition and spatial measurement, on the basis of a precisely repositioned ultrasound A-probe which measures the outer surface of the patient's skull.

 The device has been tested in a laboratory environment and under a variety of conditions. Calibration procedures include spatial object detection and measurements of plastic phantoms and cadavers. Experimental results from measurements of closely positioned or elevated, flat or rounded objects, positioned straight or at an angle to the direction of the ultrasound beam have been analyzed. Data analysis is done with LabView and MatLab.

 The setup has shown to provide excellent signal transition and coupling with a variety of materials and media as coupling surfaces. The theoretical spatial resolution (ΔX = 0.1 mm, ΔY = 0.1 mm and ΔZ = 0.26 mm) and the repeatability have been estimated on the basis of the experimental results and theoretical precision limits (acoustic wave length, construction).

 We review the current status of development and discuss implications for the development of clinical tools to be used in conjunction with intraoperative navigation.

 This work was supported by Grant No. 9318 of the Jubilee Funds of the Austrian National Bank.

Software planning tool for manual segmentation for functional endoscopic sinus surgery

K. Koulechov1, G. Strauss2 and T.C. Lüth1

1Berlin Center for Mechatronical Medical Devices, Charité & Fraunhofer IPK, Berlin, Germany; and 2Innovation Center Computer Assisted Surgery ICCAS, University of Leipzig, Germany

Introduction: This article presents a planning system and a segmentation algorithm for functional endoscopic sinus surgery with navigated control. The goal of the planning is to define a region of interest that includes the areas affected by the chronic rhinosinusitis and excludes sensitive structures that must be protected during the surgical procedure. An intraoperative treatment system, which is not described here in depth, regulates the power of the surgical shaver depending on the shaver's position within of without the segmented region.

Materials and Methods: Owing to the high demand on safety, the segmentation is done manually by the surgeon. The segmentation time is decreased by existing knowledge of the anatomy of the paranasal sinuses. Anatomical changes between continuous CT scan slices are minimal. Thus, defining a segmentation area within the actual slice is initiated by copying a segmentation region from a consecutive slice and gradually changing it using an intuitive shrinking/growing function. This simple and fast region segmentation tool leads to high acceptance by the surgeons.

Results: The time needed by an expert surgeon for segmenting seven CT scans of patients with chronic rhinosinusitis is evaluated. The average time for the segmentation is 4 min/4 s.

Neuro-navigated drilling of the internal auditory canal via retrosigmoid approach: A cadaver study

T. van Havenbergh1 and E. Koekelkoren2

1Department of Neurosurgery and 2Department of ENT, University Hospital Antwerp, Antwerp, Belgium

Aims: To determine the precision and thus safety of strictly neuro-navigated drilling of the posterior wall of the internal auditory canal (IAC).

Materials and Methods: Twenty procedures (10 cadaver heads, bilateral) consisting of a retrosigmoid approach with strictly neuro-navigated drilling of the posterior wall of the IAC were performed. For neuro-navigation spiral CT-scan with bone fixed fiducials was used. Fused pre- and post-operative images were used to determine the amount of posterior wall removed. A removal of > 90% was regarded as sufficient. The post-operative integrity of the semicircular canals, the vestibule and their cortical boundaries were evaluated by independent neuroradiologists, ENT surgeons, and neurosurgeons.

Results: The mean length of IAC opening was 92%. Opening was insufficient (89.9%) in one case. The mean residual posterior wall length was 0.75 mm. In none of the cases there was a post-operative violation of the semicircular canals.

Conclusions: Opening of the IAC via retrosigmoid route always harbors risks to the labyrinthic structures. Image guidance as applied in this study provides an useful and reliable tool adding safety to this procedure.

Session XI: Tracking technology

Evaluation of two newly developed commercial electromagnetic tracking systems

K. Schicho1, M. Donat2, M. Figl3, R. Seemann1, W. Birkfellner3 and R. Ewers1

1Medizinische Universität Wien, Universitätsklinik für Mund-, Kiefer- und Gesichtschirurgie, Vienna, Austria; 2Medizinische Universität Wien, Universitätsklinik für Neurochirurgie, Vienna, Austria; and 3Medizinische Universität Wien, Institut für Biomedizinische Technik und Physik, Vienna, Austria

Distortions of electromagnetic tracking systems by conductive respectively ferromagnetic materials are a well-known source of errors in computer-aided surgery. In consequence of the technical progress and recent developments, digitizers are available now that show a stable performance also in the presence of possible sources of errors such as large metallic surgical instruments. In this presentation, results of an evaluation of the StealthStation™ Treon™-EM (Medtronic Inc.) and the Aurora(R) (Northern Digital Inc.) are discussed as a function of the distance from devices with the potential to affect electromagnetic trackers. The devices actually tested are: a large Langenbeck hook, a dental drill with handpiece, and an ultrasound scan head. These specimens were mounted on a special measurement rack in defined distances to the position sensor within the measurement volume of the digitizer. Furthermore, a check of absolute accuracy was performed. Performance and stability was found to be clearly improved compared to systems at an earlier state of development.

Precision analysis of a stereo camera for application in surgical navigation systems

T. Krüger and T.C. Lüth

Charité/CMF-Surgery – Clinical Navigation and Robotics, Berlin, Germany

Purpose: An analysis of the working volume of stereo video camera for implementation in surgical navigation is presented. The video based Bumblebee camera is clearly smaller and cheaper than other known surgical navigation cameras used today. It has been shown that this camera is less accurate as systems like Polaris Pyramidal and Accedo (both NDI). On the other hand the position error is smaller than the influence of other error sources in surgical navigation systems.

Materials and Methods: A coordinate measuring machine (CMM) is moved in 50 mm increments through the whole working area of the navigation camera. A tracker from the medical navigation system NaviBase/RoboDent is provided with marker pads for detection by the camera. The camera recorded the coordinates of all positions during the movement of the CMM.

Results and Discussion: The jitter in the whole working volume of the camera was calculated. A specific working volume was defined where the jitter is beneath a threshold of 0.2 mm. The positioning accuracy within this working volume was analyzed. The results will be used for a future application of this camera in medical navigation.

A new technology of building instruments for passive optical navigation systems

E. Qose, W. Korb, G. Eggers, H.-J. Schulz, R. Marmulla and S. Hassfeld

Department for Oral- and Cranio-maxillofacial Surgery, University of Heidelberg, Heidelberg, Germany

The feature of navigation systems is to give real-time positional information of a certain digitizer concerning (preoperative) imaging data. For passive navigation systems, mainly sphere markers are used that are covered with retro-reflecting material. It is complex and expensive to produce such markers and they are often damaged after sterilizing them a few times. A new technology of building passive instruments based on disks instead of spheres will be presented here. Such disks offer an alternative way to build the instruments with reduced cost. On the other hand the disadvantage of the disk is its visibility which is approximately 46% of the reflective sphere's visibility. The new technology based on the idea of placing two or three groups of parallel reflective disks on the instrument, allows the camera system to see at least one group of parallel disks. These instruments have the same accuracy as traditional sphere-based instruments, but their visibility is approximately 85% of the traditional instruments visibility. Nevertheless, this is not relevant for the surgeons in praxis when performing interventions. Finally, in the paper a comparative analysis of the two technologies will be shown.

Image-guided surgical microscope with mounted mini-tracker

J. Garica Giraldez1, I.P.I. Pappas1, J. Kowal1, G. Marti2, P. Malik1 and M. Caversaccio3

1MEM Research Center, Bern, Switzerland; 2VRAI Group, EPFL, Lausanne, Switzerland; and 3Inselspital, Bern, Switzerland

Here we present a new image-guided microscope system using augmented reality image overlays directly added onto one of the views of the operating microscope. With this system, segmented objects (2D images or 3D projections), outlines of risk regions or access pathways, previously extracted from preoperative tomographic images can be directly displayed as augmented reality overlays on the microscope image. The novelty of this design consists in mounting a precise mini-tracker directly on the microscope, in parallel to the optical axis, to track the movements of the surgical tools and the patient. Besides the gain in accuracy, this setup offers an improved ergonomy.

 An accuracy study, used as a test-object, a calibration grid, and a phantom head equipped with fiducial markers, has been carried out and the contribution of the individual error sources has been analyzed. The measured overall overlay accuracy for the calibration grid was 1 mm (RMS). The new architecture offers higher overlay accuracy and better ergonomy at lower costs. The results obtained from the presented accuracy study and the experiments with cadaveric models are very promising and indicate that the system would be suitable for clinical use.

Improved accuracy of the “pathfinder” neurosurgical robot

P. Morgan1, J. Holdback2, P. Byrne1 and P. Finlay2

1University of Nottingham, England; and 2Armstrong Healthcare Ltd., High Wycombe, England

Image-guided stereotaxis accurately locates a 3D coordinate in a rigid body, defined using diagnostic imaging. We previously reported a new technique using a phantom of spheres on stalks and a depth gauge, to measure the overall application accuracy. It has been used to measure the accuracy of the PathFinder neurosurgical robot, which performs registration between planning CT images and patient fiducial markers using an arm mounted optical camera.

 Previously, PathFinder performed spatial registration of a patient in two stages: an initial long range registration to identify the approximate position of the markers and the patient's head, followed by close range views of each fiducial to locate its position more precisely. A mean application accuracy of 1.1 mm was measured.

 We have developed an improved technique to identify fiducial markers more accurately using the arm mounted camera. All registration is performed on views acquired with the robot arm at a large distance from the patient's head only, further reducing the risk of collision with the patient or other equipment while performing registration. We have measured the application accuracy of the new single-stage registration technique and found that a mean figure of 1.0 mm was achieved with the new registration methodology.

Special posters

Presurgical infant orthopaedics: 3D numerical analysis of morphological changes in the maxilla of patients with cleft lip and palate

C. Bourauel1, L. Keilig1, B. Braumann2 and A. Jäger1

1Department of Orthodontics, Universitätsklinikum Bonn, Bonn, Germany; and 2Department of Orthodontics, University of Cologne, Bonn, Germany

Optimal alignment of cleft palate segments can be achieved before surgical cleft closure by presurgical orthopedics. The question, whether this treatment using a passive appliance according to Hotz and Gnoinski stimulates the growth of the maxilla is still not completely answered. A three-dimensional computer-aided procedure was developed to metrically and visually analyze casts of the edentulous cleft alveolous and maxillary palate of infants. Chronological series of casts of the maxillae of 10 patients (1st week, 3rd, 6th, and 12th month) were digitized using a laser scanner (micromeasure 70, Microdenta). Three-dimensional surface reconstructions of the digitized casts were generated using the software Orthosurf (Surfacer 6.0, Imageware) to visually evaluate growth processes. Subsequently, fitting of the reconstructed surfaces of the chronological series of casts to the first cast was performed by matching and sizing. This procedure gives a change in total volume of the cleft segments and allows for the quantification of growth in the segments. Volume changes, that is, growth processes, in the range of approximately 1.5–1.7 from the first cast to the last one have been determined. An increase in segment volume was obvious in the cleft area after fitting the surface reconstructions.

A biocompatible flow sensor for hydrocephalus

J. Burger1, F. Zumkehr1, B. Neuenschwander2, R. Buser3, H. Keppner4 and Y. Tardy5

1University of Applied Sciences Bern, HTI Biel, Biel, Switzerland; 2Zürcher Hochschule Winterthur, Winterthur, Switzerland; 3Fachhochschule für Technik Buchs, Buchs, Switzerland; 4Ecole d'Ingénieurs du Canton de Neuchâtel, Le Locle, Switzerland; and 5Medos/Codman S.A., Le Locle, Switzerland

A fully biocompatible micromachined thermal anemometric flow sensor with RF passive telemetry using absorption modulation was designed and prototyped. A packaging solution has been developed for long-term implantation in the human body. The sensor was realized as a first prototype with electronics not yet miniaturized and is planned to be implanted together with a shunt catheter for hydrocephalus to monitor the flow of cerebrospinal fluid (CSF).

 The implantable flow sensor will allow a better management of hydrocephalus shunted patients. Furthermore, an improved long-time monitoring of the functioning of the valve as well as a better understanding of the flow of CSF will result in the long-term.

 As the sensor will be implanted for several years and its size is limited, no active power supply using batteries or accumulators could be used. The only way to provide the device with sufficient energy and to read out the measured CSF-flow is to use RF passive telemetry. As the signals measured by the sensor are very small, special care was taken to optimize the shielding of the electronics with respect to the RF power supply.

Multi-modality matching for forensic facial reconstruction

T. Buzug1, P. Hering2 and R.P. Helmer3

1RheinAhrCampus Remagen, Remagen, Germany; 2Holography and Laser Technology, Stiftung caesar Bonn, Germany; and 3Institute of Applied Forensic Medicine and Anthropology, Remagen, Germany

A new method for forensic facial soft-tissue reconstruction is presented. It is based on a nonlinear warping technique using radial basis functions known as thin-plate splines extended to 3D space. As for the manual facial reconstruction procedure, the forensic expert has to attach soft-tissue on a skull find. However, as the conventional, manual 4-step approach of

  • (a)examination of the skull,

  • (b)development of a reconstruction plan,

  • (c)practical sculpturing, and

  • (d)mask design

is very time-consuming, multi-modality elastic matching of 3D MRI soft-tissue onto the 3D CT image of a skull find is proposed.

How to transfer a computer-aided preoperative plan to the patient?

J. Chapuis1, I.P.I. Pappas1, T. Rudolph1, A. Schramm2, W. Hallermann3 and M. Caversaccio4

1MEM Research Center, Bern, Switzerland; 2Department of Oral and Maxillofacial Surgery, University of Freiburg, Germany; 3CMF Department, Inselspital, Bern, Switzerland; and 4ENT Department, Inselspital, Bern, Switzerland

New computer-aided tools allowing 3D planning of CMF surgeries have been developed in recent years. However, executing the surgical plan with high intra-operative fidelity is still a great challenge. In this presentation, we present an environment for cranio-maxillofacial preoperative planning and intra-operative navigation. Preoperative planning is realized interactively on 3D surface model of the facial and cranial bone extracted from the patient's CT data. Assessment of the pathology is performed using 3D cephalometry. Different surgical scenarios can be planned by performing virtual osteotomies, placing implants and fixation devices, and relocating bone segments. In case of asymmetric deformities, the healthy side can be mirrored and used as a template. The preoperative surgical plan can be loaded intra-operatively into the navigation module, which provides guidance to the surgeon to execute the surgical plan. Surgical tools and bone fragments can be tracked and their motion visualized as contour overlays on standard CT cut-views and in a 3D surface model representation. The system enables the surgeon to identify osteotomy lines and position bone fragments according to the preoperative plan. We present clinical cases of surgical planning for mandibular distraction, bimaxillary osteotomy, and orbital reconstruction as well as results of a navigated maxillary repositioning on a phantom model.

A graphical evaluation tool for application accuracy

W. Freysinger, G. Diakov and F. Kral

Medical University Innsbruck, Innsbruck, Austria

The goal in CAS is the best application accuracy using simple, fast, and reliable strategies for correlating the patient with his preoperative data set. Many strategies are available (pair-point, surface matching, and so on). These modalities are used in many variations, as stand alones or in combination, depending on the system used and the experience of the user. Most commonly, the achievable application accuracy with these systems is presented as the root-mean-square error. This is as we all know not equivalent to the application accuracy of a navigation system under clinical conditions. The presented evaluation tool is easy to understand and has a graphical tool to compare different referencing strategies. The spatial statistical error distribution and accuracy, the number of measurements, mean and standard deviations are shown. This tool allows analyzing huge measurement data and gives surgeons the possibility to choose the right referencing strategies for their specific needs. For developers and experimental users it allows comparing the results and benefits of upcoming new technologies. Therefore, this script will be made available for everyone.

Resection of a fronto-orbital skull base tumor and orbital wall reconstruction using navigational planning and guidance

J. Hoffmann1, J. Kaminsky2, U. Ernemann3, C. Westendorff1, M. Tatagiba2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery, 2Department of Neurosurgery, and 3Department of Neuroradiology, University Hospital, Tübingen, Germany

Introduction: Ossifying fibroma is a rare tumor entity commonly found in the mandible. When located in the paranasal sinuses, complete en bloc excision is required to prevent recurrence. We outline the clinical advantages of an image-guided fronto-orbito-zygomatic craniotomy and a navigation-assisted orbital wall reconstruction in frontobasal giant ossifying fibromas.

Materials and Methods: Our experience includes a patient with a giant ossifying fibroma located at the left frontobasis with involvement of the left ethmoid sinus and medial orbital wall leading to exophthalmos and impaired vision. The patient was scheduled for image-guided surgery by use of a wireless passive infrared surgical navigation system (VectorVision™, BrainLAB).

Results: After fronto-orbital approach, the tumor margins were located by navigational guidance. Removal was performed minimally invasive. Primary reconstruction of the left medial orbital wall was performed by image-guidance with calvarial bone. Bone chips were implanted following an accurate preoperative treatment plan. No major pre- and post-operative complications were observed.

Conclusion: Image-guidance allows a minimized approach and reconstruction of the medial orbital wall provides a safe and accurate surgical treatment.

Percutaneous navigation-guided needle puncture and sclerotherapy of periorbital vascular malformations

J. Hoffmann1, U. Ernemann2, D. Troitzsch1, C. Westendorff1 and S. Reinert1

1Department of Oral and Maxillofacial Surgery and 2Department of Neuroradiology, University Hospital, Tübingen, Germany

Introduction: Percutaneous navigation-assisted puncture is a new minimally invasive approach for the treatment of vascular malformations, especially those originating in the orbital region. We studied the feasibility and the outcome of image-data-guided puncture and needle placement in a patient with orbital vascular malformation.

Material and Methods: The study was performed in a patient with retrobulbar vascular malformation. Preoperative magnetic resonance tomography image data was acquired using a standard protocol. The puncture locations and puncture route was planned according to 3D image data reconstructions. A passive infrared light based tracking technique (VectorVision, BrainLAB) was applied to visualize the puncture needle.

Results: Under navigation-assisted guidance the needle was successfully navigated to multiple pre-planned target locations. All anticipated vascular structures could be reached accurately with no complications.

Conclusion: Percutaneous navigation-assisted puncture is a new route of access for vascular malformation treatment with many further potential applications.

Computer simulation of cerebrospinal fluid flow

V. Kurtcuoglu1, M. Söllinger1, P. Summers2, K. Boomsma1, Y. Ventikos3 and D. Poulikakos1

1Swiss Federal Institute of Technology, Zurich, Switzerland; 2University Hospital Zurich, Zurich, Switzerland; and 3University of Oxford, Oxford, England

To improve existing treatments and develop new therapies for managing diseases such as hydrocephalus that directly or indirectly involve the flow of cerebrospinal fluid (CSF) and transport processes therein, a better understanding of the fluid dynamics of CSF in the human brain is needed. This research project aims at combining MRI flow and anatomy measurements with computational fluid dynamic (CFD) calculations to obtain quantitative data on patient-specific CSF flow features. 3D anatomical data were acquired on a 3T MRI scanner. The ventricles and their connecting pathways were manually segmented and a computational grid was generated within the obtained 3D geometry. The displacement of the ventricular walls owing to brain motion was measured using MRI tagging. The maximal brain displacements were found to be in the range of 0.2 mm caudally. The associated motion was incorporated as a time-variant stretching and compression of the computational grid. Flow and pressure data were obtained by means of finite-volume calculations. We present results in the form of flow and pressure fields within the ventricular system that can be viewed as a first step towards a computer simulation environment for interventions in the CSF space.

A new method of coarse registration for 3d surface data in oral and maxillofacial surgery

T. Maier1, M. Benz1, N. Schön1, E. Nkenke2, F.W. Neukam2 and G. Häusler1

1Chair for Optics, Institute for Optics, Information and Photonics, University of Erlangen-Nuremberg, Erlangen, Germany; and 2Department of Oral and Maxillofacial Surgery, University of Erlangen-Nuremberg, Erlangen, Germany

The zygomatic fracture associated with a dislocation of the eye ball is one of the most frequent traumata to the cranium. We have developed a system that supports the surgeon in adjusting the correct eye position. The system's core is an optical range sensor that allows fast 3D data acquisition. We preoperatively compute 3D target data of the eye position. During the operation, the surgeon can acquire actual data at any time. The system compares the actual with the target data in order to give practical feedback to the surgeon. The data have to be registered for comparison. Fine registration is done by default via the ICP-algorithm. A necessary preceding coarse registration is often accomplished by user interaction. The reliable real time automation of coarse registration still remains a challenge. Moreover, we need an algorithm that deals with data that are subject to form changes by the operation. This contribution is about a coarse registration method that exploits the Gaussian image of a face and is an advancement of T. Maier, M. Benz, G. Häusler, E. Nkenke, F.W. Neukam, F. Vogt. Automatische Grobregistrierung intraoperativ akquirierter 3D-Daten von Gesichtsoberflächen anhand ihrer Gaußschen Abbilder. Procs BVM'03, pp.11–15, 2003. It allows the fast automation of this process.

Facial skin shift and face recognition in computer-assisted surgery

R. Marmulla1, J. Mühling1, T.C. Lüth2, S. Hassfeld1

1University of Heidelberg, Heidelberg, Germany; and 2Charité, Berlin, Germany

In cranio-maxillofacial surgery markerless patient registration through the facial skin becomes increasingly accepted for registration between preoperative imaging and the intraoperative surgical site. However, a changing position or mimic activity may change the skin's geometry, generating an incongruence between the preoperatively and intraoperatively recorded facial contours, which leads to inaccuracies during the computer-assisted intervention.

 In the present study, the influence of skin shift on the accuracy of patient registration in CAS was evaluated. Skin laser-scans were made of 12 conscious persons, both in sitting and in supine positions, as well as under mimic activity. Then the laser-scans were referenced to the corresponding CT data set. Markerless patient registration was reduced by 0.4 mm during severe facial mimic activity. Mass-related skin shifts reduced the mean accuracy of patient registration by 0.5 mm.

 Thus, it is important to make sure that the position during CT-acquisition is identical to the position during intraoperative laser-scan registration. Mimic activity during CT acquisition should be avoided.

Accurate needle positioning for interstitial brachytherapy treatment of tumors in the skull base

I.P.I. Pappas1, P. Ryan1, P. Cossmann2, B. Borgeson1 and M. Caversaccio3

1MEM Research Center, Bern, Switzerland; 2Hirslanden Klinik Aarau, Aarau, Switzerland; and 3Inselspital, ENT-Department, Bern, Switzerland

Treatment of tumors in the frontal or lateral skull base area with interstitial brachytherapy requires high targeting accuracy of the needles to avoid harming vital anatomical structures. To enable safe placement of the needles in this area, we developed an image-guided planning and navigation system for brachytherapy. The system consists of a two-stage positioning mechanism, an infra-red tracking camera (Optotrak), a non-invasive head-fixation frame with a vacuum cushion and a computer-aided planning and navigation software, which provides intuitive visual guidance for the fine adjustment of the mechanical degrees of freedom. The positioning mechanism consists of a passive mechanical arm with six gross degrees of freedom and a microscrew positioning system with five fine adjustable degrees of freedom. The two-stage positioning mechanism combines the advantages of a large working volume with high positioning resolution. A limiting factor for the targeting accuracy is the mechanical bending of the needles. To avoid bending because of the weight of the navigation markers, the number of markers on the needles was reduced to a single, axially-mounted LED to track the tail of the needle. A navigation shield attached onto the needle-holder gives the orientation of the needle. With this set-up, an experimental ex vivo targeting study showed that a positioning accuracy of 0.8 mm (RMS) can be achieved. The presented system provides a promising tool for the safe treatment of tumors in the skull base area.

Acoustic monitoring of bone ablation using pulsed CO2 lasers

A. Rätzer-Scheibe1, M. Ivanenko1, M. Klasing1, M. Werner1 and P. Hering1,2

1center of advanced european studies and research (caesar), Bonn, Germany; and 2University of Düsseldorf, Germany

Laser osteotomy offers remarkable advantages, for example free cut geometry, over the conventional mechanical saw. Unlike the saw, however, the laser lacks haptic feedback during cutting. Based on ablation noise analysis, we are developing an acoustic-feedback system for laser osteotomy to obtain in situ information on the ablation within the tissue.

 We used a pulsed TEA CO2 laser (wavelength, 9.57 µm; pulse length, 1 µs; pulse energy, 25 mJ) and piezoelectric transducers for sound detection. Various bone specimens and reference materials were studied. For the determination of the ablation crater depth, we analyzed the time delay between the laser-induced acoustic signal from the surface of the specimen and the bottom of the ablation crater. The possibility of controlling the cut-depth in material with known acoustical properties with high precision is demonstrated.

 For acoustic tissue differentiation, we analyzed acoustic spectra initiated by laser pulses in different materials. The spectra shows specific material-based features. This will prompt surgeons with information about the transition from compact bone to other materials, for example soft-tissue.

 An acoustic feedback system will eventually provide online control over the depth of the incision and ancillary recognition of the tissue exposed to the laser.

Vredmed, a tool for 3D-visualization of augmented and virtual reality in surgery

A. Schäffer1, A. Schäffer1, S. Petra1, A. Zieringer2 and A. Kozak3

1PolyDimensions GmbH, Bickenbach, Germany; 2AMZ GmbH, Bernsheim, Germany; and 3SeeReal Technologies GmbH, Dresden, Germany

Purpose: Medical-imaging delivers 3D anatomical information, beneficial for surgery, but there is no appropriate visualization software for use in the OR.

Material: VREDmed, a 3D-rendering engine, was developed to visualize patient's data from CT, MRI, ultrasound, and surface models from preoperative planning. VREDmed supports viewing with an auto-stereoscopic display by SeeReal Technologies, surgical guidance via interfaces for tracking devices and interaction through speech recognition.

Methods: CT, MRI, and Ultrasound were imported in VREDmed for preoperative planning in stereo-3D and integrated in a coordinate system by tracking, if required. Surgical planning and further information was gained through postprocessing with segmentation- and CAD-software. Obtained data was visualized on monitors, auto-stereoscopic displays, and Head-Mounted-Displays. Results were evaluated to find fast and good information contents. OR testing during interventions is scheduled soon.

Results: VREDmed is a powerful tool that presents various imaging data for preoperative planning, allows intuitive inspection, interaction during surgical procedures with interfaces for tracking, haptic devices, and stereoscopic displays.

Conclusion: AR and VR software is very helpful for surgeons, if extendable for specific needs and situations. Documented interfaces for fast adaptations to track instruments, medical-imaging, and force-feedback are advantageous.

Technical realization of teleconsultation in cranio-maxillofacial surgery

K. Schicho, A. Wagner, R. Seemann, M. Truppe and R. Ewers

Medizinische Universität Wien, Universitätsklinik für Mund-, Kiefer- und Gesichtschirurgie, Vienna, Austria

Computer-assisted navigation technology and augmented reality environments can be combined with telecommunication and networking to enable numerous types of interactive support for the treatment planning as well as for intraoperative advice by remote experts. Beginning from the middle of the 1990s the internet as a new medium is used for telemedical support of surgical interventions in the field of cranio-maxillofacial surgery. In addition to the internet also conventional videoconferencing set-ups are successfully applied for teleconsultation. Based on the experience from more than 50 interactive teleconsultations and live videostreamings of operations this lecture summarizes five different technical set-ups that were tested and have proved practical for clinical routine, that is, conventional videoconferencing via ISDN-connections without remote control of the navigation computer, TCP/IP-based interactive teleconsultation with remote control of the navigation computer via ISDN or via network-connection and combinations of these types. The third generation of mobile communication is the UMTS-technology. Recent studies investigated UMTS for the transmission of endoscopic videostreams. Possible applications, technical expenditure, limitations, and advantages/disadvantages will be discussed in this presentation.

Real-time navigation for laser interstitial thermotherapy in head and neck vascular malformations

D. Troitzsch1, J. Hoffmann1, C. Westendorff1, U. Ernemann2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery and 2Department of Neuroradiology, University Hospital, Tübingen, Germany

Purpose: Image-guided navigation-controlled LITT offers a non-invasive safe treatment option.

Materials and Methods: Multiple image data navigation-guided LITT was performed in patients (five procedures) with giant venous malformations of the maxillofacial area.

Results: As a result of the 3D reconstruction for laser surgical planning and the defined target areas for laser probe navigation, the application of the interstitial laser treatment was performed accurately. In all cases, follow-up examination clearly showed a diminished tumor volume and all patients reported subjective amelioration.

Conclusion: The results suggest that navigation-guided LITT can be performed safely by preserving vital structures from collateral thermal damage and can be effective in the treatment of complex vascular malformations.

Computer-assisted virtual planning and intraoperative surgical navigation for treatment of complex orbito-zygomatico-maxillary fractures

C. Westendorff1, J. Hoffmann1, D. Troitzsch1, F. Dammann2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery and 2Department of Diagnostic Radiology, University Hospital, Tübingen, Germany

Introduction: Accurate assessment of the position of the zygomatic bone in relation to the cranial base posteriorly and the midface anteriorly is the key to the acute repair of midfacial fractures as secondary reconstruction of post-traumatic deformities of the orbito-zygomatico-maxillary complex remains a major surgical challenge.

Materials and Methods: Non-comparative series of 10 consecutive patients with zygomatic fractures were scheduled for virtual reality-planning and image-guided repositioning by use of a wireless passive infrared surgical navigation system (VectorVision™, BrainLAB).

Results: In all cases, preoperative virtual fracture reduction allowed a precise anatomic repositioning of the zygomatic segment. Fixation was performed using miniplates and screws. Post-operative X-ray revealed an accurate anatomy in all patients and there was no requirement for secondary corrections.

Conclusion: Image-data-based virtual planning and navigation-assisted zygomatic trauma surgery provides effective treatment options.

Minimal access and navigation-guided surgery for resection of a craniofacial bone lesion

C. Westendorff1, J. Hoffmann1, D. Troitzsch1, F. Dammann2 and S. Reinert1

1Department of Oral and Maxillofacial Surgery; and 2Department of Diagnostic Radiology, University Hospital, Tübingen, Germany

Introduction: We introduce and outline the clinical advantages of a navigation-assisted approach to a lateral skull bone lesion using image-guided surgery.

Materials and Methods: Our experience includes one patient with a skull bone tumor. Skull radiographic computed tomography revealed a bone lesion in the right temporo-parietal region. Computed tomographic scans indicated a bone destructive lesion involving external and internal calvarial layer. The patient was scheduled for image-guided surgery by use of a wireless passive infrared surgical navigation system (VectorVision™, BrainLAB).

Results: The procedure was successful and the tumor was minimally invasive, removed with no pre- and post-operative complications. The patient was ready for discharge two days after surgery. Post-operative imaging scans showed no recurrent tumor process.

Conclusion: Traditional surgical approaches to skull bone tumors may result in significant cranial deformity and morbidity. Image-guided excision with surgical navigation techniques provide a safe minimally invasive surgical treatment option.

Improved craniofacial endosseous implant positioning using image-guided surgical navigation

C. Westendorff, J. Hoffmann, D. Troitzsch and S. Reinert

Department of Oral and Maxillofacial Surgery, University Hospital, Tübingen, Germany

Introduction: Craniofacial implants provide excellent stability and retention for auricular prosthetic rehabilitation. We used image-guided navigation for implant positioning to test feasibility and practical impact.

Materials and Methods: All patients undergoing navigation-assisted craniofacial implant treatment were included. Image-guided surgery was performed by use of a passive infrared surgical navigation system (VectorVision™, BrainLAB). The preoperative computed tomography (CT) data was obtained using the Somatom Sensation 16 multislice-scanner (Siemens).

Results: A total of eight implants was placed in the mastoid area and various other craniofacial locations. The implant positions were conventionally planned and updated after image-guided measurements in conjunction with the bone thickness. After registration, axial, coronal, and sagittal reconstructions of the pointer tip position in the regions of interest were displayed in real-time. The proper location and positioning of implants in the craniofacial area could strongly be improved and treatment was well controlled.

Conclusion: Navigation-assisted surgical techniques have the ability to assist in a proper positioning of craniofacial implants, which in turn complements the prosthetic result.

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.