ABSTRACT
The morphological assessment of anatomical structures is clinically relevant, but often falls short of quantitative or standardised criteria. Whilst human observers are able to assess morphological characteristics qualitatively, the development of robust shape features remains challenging. In this study, we employ psychometric and radiomic methods to develop quantitative models of the perceived irregularity of intracranial aneurysms (IAs). First, we collect morphological characteristics (e.g. irregularity, asymmetry) in imaging-derived data and aggregated the data using rank-based analysis. Second, we compute regression models relating quantitative shape features to the aggregated qualitative ratings (ordinal or binary). We apply our method for quantifying perceived shape irregularity to a dataset of 134 IAs using a pool of 179 different shape indices. Ratings given by 39 participants show good agreement with the aggregated ratings (Spearman rank correlation ). The best-performing regression model based on quantitative shape features predicts the perceived irregularity with
Acknowledgments
We would like to thank Diana Sapina for her support in preparing the database of intracranial aneurysms, and Victor Garcia and John Bennett for critically proofreading the manuscript.
Disclosure statement
No potential conflict of interest was reported by the authors.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Raw 3D-DRA of the AneuX test data set were provided by the University Hospital of Geneva and collected with formal patient consent according to the @neurIST protocol and ethics authorisation PB_2018-00073 (previously CER 07-05) released 1 June 2007 and renewed April 13th 19 August 2010th 2014 and 28 February 2018 initially by the Geneva Cantonal Ethics Commission for Research involving Humans and renewed by swissethics in 2018.
Supplementary material
Supplemental data for this article can be accessed here.
Additional information
Funding
Notes on contributors
Norman Juchler
Norman Juchler, MSc – received in 2009 a master’s degree in Mechanical Engineering from ETH Zurich, Switzerland. He re-entered academia after working several years in a software start-up and is currently pursuing his PhD at the Institute of Applied Simulation (Zurich University of Applied Sciences, Waedenswil) and Institute of Physiology (University of Zurich). His research focuses on 3D-shape analysis of intracranial aneurysms with the goal to improve disease prognostics. Norman contributed to the AneuX project whose primary goal was to improve the understanding of the formation process of intracranial aneurysms using data-driven methodologies.
Sabine Schilling
Dr Sabine Schilling – studied physics at the University of Heidelberg and obtained her PhD in Theoretical Particle Physics from the University Zurich. Besides gaining experience in the financial industry as a strategy consultant for Oliver Wyman, she has been pursuing an interdisciplinary research record with focus on statistics, machine learning and numerical simulations and several years of teaching experience of statistics. She is currently working at the Lucerne University of Applied Sciences and Arts.
Stefan Glüge
Dr Stefan Glüge – received the diploma degree in Information Technology in 2008 from the Otto-von-Guericke University Magdeburg, Germany. From 2009 to 2013, he was PhD Student at the Cognitive Systems group at the Otto-von-Guericke-University at the Institute for Information Technology and Communications. Since 2013, he is an assistant researcher at the Institute of Applied Simulation at the Zurich University of Applied Sciences, Switzerland. Stefan worked in various research projects with a focus on statistical modelling and machine learning and published several peer-reviewed papers and articles in the fields of recurrent neural networks and automatic speech processing.
Philippe Bijlenga
Prof. Philippe Bijlenga, MD PhD – obtained a degree in medicine, medical biology and a doctorate in science and medicine from the Faculty of Medicine and Sciences of the University of Geneva in 1995, 1997 and 2000, respectively. He received further training in neurosurgery and vascular neurosurgery during the following years in Geneva, Switzerland, and Cambridge, England. Besides his occupation as a neurosurgeon at Geneva University Hospitals, Philippe has been researching on clinically relevant topics to improve the effectivity and safety of neurosurgical procedures and to advance the knowledge of cerebrovascular diseases. From 2006 to 2011, he was contributing as a leading scientist to the European @neurIST project. Between 2015 and 2018, he was principal investigator for the Swiss AneuX project.
Daniel Rüfenacht
Prof. Daniel Rüfenacht, MD PhD – is contributing since 1984 to the field of minimally invasive endovascular therapies of neurovascular diseases. He graduated from the Medical School of the University of Berne, Switzerland, and obtained clinical and academic experience during extensive work stays in Switzerland, France and the USA. Since 2008, he has been developing a diagnostic and interventional neuroradiology expert group at the SwissNeuroInstitute, Klinik Hirslanden, Zurich, in collaboration with Prof. Isabel Wanke, UKE Essen, Germany. He has co-authored over 255 peer-reviewed publications, has been involved in 23 national and 4 European research grants (@neurist, THROMBUS, VPH-SHARE, VPH-DARE), and has co-founded several educational/academic projects and initiatives: ICS (2004), ESMINT society (2008), ESMINT/ANIC teaching course (2008), SwissNeuroFoundation (2013), iNEW (2017).
Vartan Kurtcuoglu
Prof. Dr Vartan Kurtcuoglu – is the head of the The Interface Group at the Institute of Physiology, University of Zurich, with joint appointments at the faculties of medicine and science. He received his PhD in 2006 from ETH Zurich on the topic of cerebrospinal fluid dynamics. His research aims to answer fundamental questions of physiology and address clinical needs through the convergence of engineering, biological and medical research, focusing on the fluid systems of the human body. His current studies address challenges posed by pathologies in the kidneys, the brain and the cardiovascular system.
Sven Hirsch
Prof. Dr Sven Hirsch – is a trained physicist, specialised in image processing and physiology, lecturing physics and complex systems. His Biomedical Simulation Group at the ZHAW has a strong focus on vessel wall pathologies and their lifecycle, applying continuum mechanical and agent-based techniques to simulate biological structures, as well as statistical modelling and machine learning to approach the related medical research questions. A particular field of attention is the treatment of cerebral aneurysms and the identification of image biomarkers to judge the disease state.