152. DeepImplant – Modeling of personalized cranial implants in real-time using artificial intelligence
Marek Wodziński, AGH University of Kraków, Poland / HES-SO Valais-Wallis, Switzerland
Abstract:
Every year, thousands of people suffer from cranial damages due to traffic accidents, surgeries, or warfare, requiring personalized cranial implants to close the cranial cavity. The modeling and manufacturing of personalized cranial implants is a costly and time-consuming process. Nowadays, in clinical practice the implants are prepared by experienced mechanical engineers and manufactured by external companies, leading to high costs and weeks of waiting time for the reconstructive surgery. The presentation will discuss the current status of the DeepImplant project automating personalized implant modeling by proposing novel AI methods. The talk will present recent advances handling high heterogeneity of the skull data. Finally, further research directions will be presented that will eventually lead to fully automatic system operating in real-time.
About the author:
Marek Wodziński is an Assistant Professor at AGH University of Kraków, Deep Learning Expert at Shen.AI, and Postdoctoral Researcher at HESSO Valais. He has more than 7 years of experience in medical image analysis, computer vision, and deep learning. His current research interests combine automatic shape modeling, medical image registration, and digital pathology. He won 6 international scientific challenges related to medical image analysis and was on the podium more than 15 times. He is currently Principal Investigator in the DeepImplant project related to automatic modeling of personalized cranial implants, funded by NCBR under the LIDER grant programme, and Deep Learning Expert in several other projects. .
