Validating 3D face morphing towards improving pre-operative planning in facial reconstruction surgery

ABSTRACT

Pre-traumatic/pathologic 3D facial imaging is rarely available to guide craniofacial reconstruction. This study aims to independently validate the regional accuracy of a 3D morphable model to estimate face shapes from 2D photographs for craniofacial surgical planning.

3D shape estimates were generated using the Basel Face Model (BFM 2017) from 2D photographs of 100 multi-racial subjects in the Binghamton University 3D Facial Expression database. Accuracy was evaluated by the per-vertex Euclidean distance between the shape estimate and the true 3D scan within defined facial regions.

The 3D estimates’ average RMS distance error across all facial regions was 2.68 ± 0.97 mm, based on photos analysed with 10,000 iterations. The eyes, cheek, chin, forehead and mouth regions fit within ~2.5 mm from the true face shape, representing only marginal perceptible error. The nose and temple regions had lower accuracy (~3.1 mm) for all subjects. Significant differences in the nose region were dependent on race (Caucasian: 2.4 mm, East-Asian: 4.8 mm) and sex (Male: 2.5 mm, Female: 3.6 mm).

On average, 3D face shape estimates using the BFM yielded clinically acceptable accuracy sufficient for use in planning to guide several craniofacial reconstruction regions. However, in individual cases, considerable errors exceeded clinical limits.

KEYWORDS:

• Craniofacial reconstruction
• 3D preoperative planning
• computer-assisted surgery
• morphable models
• Basel face model

Acknowledgments

Support for this work has been provided by the Natural Sciences and Engineering Research Council of Canada and FedDev-Ontario.

Financial Disclosure Statement: Funding for this research was received from NSERC & FedDev-Ontario. None of the authors has a financial interest in any of the products, devices, or drugs mentioned in this manuscript.

Part of this work was presented at the 17th annual Imaging Network Ontario (IMNO) symposium in March 2018 at Toronto, Ontario, Canada. (http://imno.ca/2018-symposium)

Author Contributions

Zachary Fishman – Primary Author. Responsible for study design and implementation, 3D facial data analysis and drafting the paper.

Jerry Liu – Contributed to the implementation of the Basel Face Model and 3D estimation evaluation.

Joshua Pope – Contributed to the 3D estimation evaluation algorithm.

Jeffrey Fialkov – Contributed to study design, provided clinical guidance, and review of the paper.

Cari Whyne – Senior Author. Developed study design, guidance on data analysis and results, critically revised the paper.

Institutional review board & clinical trial registration: Not Applicable. The facial data illustrated in this work are of individuals within the Binghamton University 3D Facial Expression (BU-3DFE) Database, whose 3D scans have been made available to the scientific community for research purposes. The collection of the BU-3DFE database is described in Yin et al. (2006). No new subjects or patients were collected or analyzed in this work.

http://www.cs.binghamton.edu/~lijun/Research/3DFE/3DFE_Analysis.html

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) & FedDev-Ontario.

Notes on contributors

Z. Fishman

Dr. Z. Fishman, PhD, is a Post-Doctoral Fellow in the Orthopaedic Biomechanics Lab at the Sunnybrook Research Institute working under the supervision of Dr. Cari Whyne and Dr. Jeffrey Fialkov. Zachary has an undergraduate and Masters degree in Mechanical Engineering, with a focus on material analysis, 3D design and prototyping. Zachary’s research is investigating new translational tools including 3D scanning, 3D printing, and 2D‐3D image analysis. His research focus is to improve the accuracy and workflow in craniofacial reconstruction.

Jerry Liu

Jerry Liu is a 4th year student at the University of Waterloo. He is currently working in the tech industry as a software engineer/product manager. His background comes from extensive research in the fields of robotics, 3D morphable modeling and computer vision. He completed several internships over the course of his undergraduate studies at Harvard University, Sunnybrook Research Institute, Apple Inc and Google LLC in a variety of roles. He is passionate about building products with positive social impact, as well as mentoring younger students. One of the initiatives he co-founded is 37C Inc, a non-profit organization dedicated to creating and sharing opportunities for students to learn from and engage with experts in the fields of biotechnology and bioengineering.

Joshua Pope

Joshua Pope is a graduate of Biomedical Engineering from the University of Waterloo, inaugural class of 2019. Academically, Joshua has done research at the Lunenfeld-Tanenbaum Research Institute, Centre for Addiction and Mental Health and Sunnybrook Research Institute. After graduating from the University of Waterloo, Joshua founded Trajekt Sports, a robotics company that has developed a baseball pitching robot for elite batter training.

J.A. Fialkov

Dr. J.A. Fialkov MD, MSc, FRCSC, is Head of the Division of Plastic and Reconstructive Surgery at Sunnybrook Health Sciences Centre and an associate professor in the Department of Surgery with cross-appointment to the Institute of Biomedical Engineering at The University of Toronto. As a fellowship trained craniomaxillofacial surgeon, his clinical focus encompasses the reconstruction of post-traumatic, congenital and post-ablative facial deformities. As part of the University of Toronto teaching faculty, he provides post-graduate supervision in adult craniofacial surgery to fellowship trainees from Canada, the United States, Europe and the Middle East. Dr. Fialkov is also an associate scientist in the Holland Bone and Joint Research Program at Sunnybrook Health Sciences Centre. His research focus is on craniomaxillofacial reconstructive techniques and translational technologies, with an emphasis on biomechanics using cadaveric and in silico modelling. His interest in craniofacial trauma, facial reconstruction and craniomaxillofacial biomechanics has led to the publication of numerous peer-reviewed articles and book-chapters on the subject. In addition, his research interests have led to several innovations in the field including novel surgical techniques and technological innovations such as a novel patented osteo-synthetic system.

C.M. Whyne

Dr. C.M. Whyne, PhD, FIOR, is the Susanne and William Holland Chair in Musculoskeletal Research at Sunnybrook Health Sciences Centre in Toronto. She is a Senior Scientist and the Director of the Holland Bone and Joint Research Program at Sunnybrook Research Institute and a Full Professor in the Department of Surgery, Institute of Biomedical Engineering and Institute of Medical Sciences at the University of Toronto. Dr. Whyne received her BSc. in Mechanical Engineering from Queen’s University and her PhD from the University of California Berkeley / University of California San Francisco in Bioengineering. The focus of her work is clinically translational bioengineering research. Dr. Whyne’s research integrates biomechanical analyses with basic science, preclinical and clinical investigations, including extensive work in computational image analysis, micro-imaging, machine learning and finite element modeling techniques. Her work also incorporates design, simulation, evaluation and clinical translation of novel less/minimally invasive surgical techniques and devices. The primary foci of Dr Whyne’s research are cancer in bone, biomechanics (spinal, craniomaxillofacial, upper and lower extremity) and fracture fixation/healing.