Abstract #2129

Evaluation of the fairness and effectiveness of nnU-Net on multi-organ segmentation

Qing Li¹, Yan Li², Longyu Sun¹, Mengting Sun¹, Meng Liu¹, Xumei Hu¹, Xinyu Zhang¹, Xueqin Xia³, Shuo Wang⁴, Yinghua Chu⁵, and Chengyan Wang¹

¹Human Phenome Institute, Fudan University, Shanghai, China, ²Department of Radiology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, ³Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China, ⁴Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, Shanghai, China, ⁵Simens Healthineers Ltd, Shanghai, China

Synopsis

Keywords: Visualization, Visualization, Fairness; Bias

Motivation: A systematic analysis of the segmentation effectiveness for fairness helps enhance the effectiveness of artificial intelligence(AI) model, which has not been done before.

Goal(s): This study aims to compile statistics the relation between the segmentation effectiveness and aging, gender as well as anatomical regions.

Approach: The nnU-Net model is used for organ segmentation while the DICE was computed to evaluate the relation between the effectiveness with aging and gender and the heatmap was used to visualize the spatial error distribution regarding anatomical regions.

Results: The result demonstrates variations in nnU-Net's effectiveness within subgroups, highlighting the significance of attention mechanisms for segmentation model enhancement.

Impact: This study comprehensively evaluated the fairness and effectiveness of nnU-Net across multiple organs within the body. An analysis was conducted to investigate the relationship between segmentation errors and age, gender as well as anatomical regions for organ segmentation.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.