A survey on semi-supervised medical image segmentation

PhD Qualifying Examination


Title: "A survey on semi-supervised medical image segmentation"

by

Miss Huimin WU


Abstract:

Deep learning has achieved great successes on a variety of tasks including 
image classification, segmentation and object detection but requires a huge 
amount of labeled data to achieve promising performances, which can be 
difficult to obtain especially in biomedical applications. Semi-supervised 
learning is drawing an increasing research interest on medical image 
segmentation task since it allows to leverage unlabeled data to outperform 
fully supervised counterpart with a small set of labeled data only, making it a 
label-efficient learning paradigm. Before presenting semi-supervised 
techniques, we first introduce background knowledge of medical image 
segmentation. Then we offer a taxonomy of existing semi-supervised medical 
image segmentation methods and have a detailed discussion over these methods by 
exhibiting an outline from earlier methods to recent advances. Finally, we 
conclude this survey with a discussion over existing methods and some prospects 
for the future research.


Date:			Wednesday, 13 October 2021

Time:                  	3:00pm - 5:00pm

Zoom meeting:
https://hkust.zoom.com.cn/j/91871029344?pwd=MG4wV1RHM081US9LeHcyQi9TUENvQT09

Committee Members:	Prof. Tim Cheng (Supervisor)
 			Dr. Xiaomeng Li (Supervisor)
 			Prof. Albert Chung (Chairperson)
 			Dr. Hao Chen


**** ALL are Welcome ****