Talk by Dr. Lin Gu (National Institute of Informatics)

Title: Insufficient Label, Transparency and Robustness of Medical Image Analysis

Abstract:
Though deep learning has shown successful performance in the medical image analysis in the tasks of classifying the label and severity stage of certain disease. However, the CNN based methods suffer the bottleneck of lacking training label and interpretability. For example, most of them give few evidence on how to make prediction. To make it worse, the ubiquitous adversarial attack has posed even more serious challenge on its real application. This talk would introduce the recent progress for these challenges on various medical image domains.

Bio: Dr. Gu joined National Institute of Informatics in Tokyo in June 2016. From October 2016 to March 2019, he was also a regular visiting scholar of Kyoto University. Before moving to Japan, he was a Postdoctoral Research Fellow at Bioinformatics Institute, A*STAR, Singapore. He completed his PhD studies at the Australian National University and NICTA (Now Data61) in 2014. At that time, he was working on the hyperspectral imaging and colour science. After that, from 2014 to 2016, he was mainly focusing on the application of machine learning on the biomedical imaging. He is now working on both medical imaging analysis and computational photography.