Disputas - Master of Science Debesh Jha
Master of Science Debesh Jha will on 21st January at 12.15 publically defence his thesis for the PhD degree in Science.
Title of the PhD thesis:
«Machine Learning-based Classification, Detection, and Segmentation of Medical Images»
Popular scientific abstract:
Gastrointestinal (GI) tract cancers are the leading cause of cancer-related death worldwide. Among GI cancers, colorectal cancer is the third most commonly diagnosed cancer. Colonoscopy is the gold standard for screening colorectal cancer, but it is an expensive, time-demanding, and operator-dependent procedure. Studies have reported polyp miss rates of 22%-28% during these procedures. Computer aided diagnosis (CAD) methods can help to highlight suspicious lesions on the screen and alert gastroenterologists in real-time, improving the clinical outcome irrespective of operator experience, potentially saving millions of lives. In this thesis, we have designed machine learning-based architectures for GI tract abnormality detection, classification and segmentation. The availability of public datasets was one of the significant challenges for the development of automated methods. We addressed this problem by collecting, curating, annotating, and publicly releasing several datasets, including the world’s largest publicly available GI endoscopy and video capsule endoscopy datasets. In our experiments, our classification algorithms classify different GI endoscopy findings with an accuracy of 98.07%. For polyp segmentation, our algorithms can identify and segment the potential presence of lesions during colonoscopy with an accuracy of more than 94.93% and at a real-time speed of 182.38 frames per second. We are able to simultaneously identify multiple polyps, including flat and sessile polyps that are often overlooked by endoscopists during the colonoscopy examination. As an important part of the procedure, we also performed instrument segmentation to be able to detect which instruments are used during the examination. The developed method is able to segment different types of surgical instruments in real-time. As another aspect of our work, we also addressed the challenge of generalizability, meaning that our models can perform well on completely new data enabling for example that a model can be moved from one hospital to another. We demonstrated reliable performance and high generalizability compared to baseline algorithms. Due to the challenge of not having access to high-end hardware in the hospitals, we have developed lightweight architectures that can be integrated with low-end hardware devices. Our algorithms are not only designed for polyp segmentation and surgical instrument segmentation but can also be exploited for other medical or non-medical image segmentation tasks. All of our work and algorithms are open-sourced and received very well by the community.
Supervisors:
Evaluation Committee:
Leader of the public defence:
Pro-Dean Cordian Riener, Faculty of Science and Technology, UiT, will lead the public defense.
Opposition ex auditorio:
If you have any questions for the candidate during the public defence, please send an e-mail to leader of the public defence. They will announce your questions during the defence.
Trial lecture:
The trial lecture is held on Friday 21st of January at 10.15 in the same auditorium.
The title of the trial lecture is:
«Opportunities and potential pitfalls on practical machine learning in health care»
Streaming:
The trial lecture and the public defence will be streamed via this link.
Audience:
UiT follows the national guidelines regarding infection control. A maximum of 20 people are allowed in the auditorium during the defence, as long as everybody keeps a distance of 1 meter at all times.