Home > Research > Colloquium > Content

“Xue Cui Academic Forum" -- Application of Deep Federated Learning in Medical Imaging

DATEDecember 6, 2022

“Xue Cui Academic Forum" adheres to the concept of learning from scholars and extracting the essence of science and technology. It takes learning as the soul and education as the foundation. It pursues technological innovation, improves academic taste, creates a rich academic atmosphere, and shares a feast of science and technology!

Report Title: Application of deep federal learning in medical imaging

Reported by: Shenjun Zhong

Report time: 9:00 on Friday, December 9, 2022

Reporting place: Zoom Conference:

Meeting ID: 899 4952 2393

Password: 475666

Sponsored by: Academy of Science and Technology

Youth Science and Technology Association of Harbin Engineering University

Organized by: College of Computer Science and Technology

Report content: In the context of deep learning, especially supervised learning, a large amount of labeled data is often required for model training to achieve ideal model accuracy and robustness. However, medical image data, as private data of medicine and research institutes, has high privacy requirements, which leads to homogeneous data being separated on the data island of each site. It is difficult to train a model with high stability relying on the limited data of a single site. Federated Learning can protect privacy by learning shared models while keeping training data locally, thus providing a solution to the problem of data islands. This lecture will introduce Federal Learning, its application and current situation in the field of medical imaging.

About the Reporter: Dr. Shenjun Zhong, now a researcher in the Medical Imaging Center of Monash University in Australia and the National Imaging Center in Australia, is mainly engaged in the research and development of medical imaging AI technology and bio information system. He received his doctorate from Monash University in Australia in 2016. Dr. Shenjun Zhong has worked in academia and industry for many years. He led his team to develop and launch a large-scale dialogue system based on deep learning in Australia Telecom and published many works in the field of medical imaging. At present, the main research directions include the application of deep learning technology in medical imaging, and the application of deep federated learning architecture in medical imaging.

Academy of Science and Technology

Dec 6th, 2022