Recent developments in deep learning-based pulmonary image analysis
Brown Bag Lecture by Dakai Jin, Ph.D. | 8/21/2018 11:00AM – 12PM | 7th Floor Conference Room, Bldg 38A
Join WebEx meeting: https://nih.webex.com/join/bneeriemernih.gov (access code if needed: 629 392 468)
For audio, if joining by phone, please dial 1-650-479-3208 (US/Canada)
In the past decade, deep learning approaches have gained significant success within the natural image domain, due to their unparalleled performance for such challenging data. Deep learning-based methods have also been successfully applied in many medical image analysis and computer-aided diagnosis tasks. We first discuss challenges in the medical domain, and then present recent work in deep learning based pulmonary image analysis and diagnosis – in airway, lung and lobe segmentation, interstitial lung disease (ILD) pattern detection, and lung nodule inpainting via adversarial generative models.
Dr. Dakai Jin received his Ph.D. in Electrical and Computer Engineering from the University of Iowa in 2016. Currently, he is a visiting research fellow in the Department of Radiology and Imaging Science at NIH. His research focuses on medical image analysis, machine learning, and digital topology and geometry. He is particularly interested in incorporating medical domain knowledge with deep learning techniques to solve a broad range of medical imaging problems, such as lesion classification, disease detection, organ segmentation, etc. He has years of experience in pulmonary, brain and bone image analyses from CT and MRI, and expertise in medical image segmentation, computer-aided diagnosis and quantitative image analysis. He has also developed two major skeletonization algorithms for 3D fuzzy objects, which are now broadly used in various medical imaging problems. Dr. Jin has published more than 30 papers in reputable journals and peer-reviewed conferences. He is a regular reviewer for the major journals and conferences in the field, and also currently serves on the editorial board of the journal Computers in Biology and Medicine.