Intrinsic Bias Identification on Medical Image Datasets

• DOI: 10.48550/arxiv.2203.12872
• Publication Date: 2022-01-01
• Authors: Shijie Zhang; Lanjun Wang; Lian Ding; An‐An Liu; Senhua Zhu; Dandan Tu
• Keywords: Identification

Machine learning based medical image analysis highly depends on datasets. Biases in the dataset can be learned by model and degrade generalizability of applications. There are studies debiased models. However, scientists practitioners difficult to identify implicit biases datasets, which causes lack reliable unbias test datasets valid To tackle this issue, we first define data intrinsic bias attribute, then propose a novel identification framework for The contains two major components, KlotskiNet Bias Discriminant Direction Analysis(bdda), where KlostkiNet is build mapping makes backgrounds distinguish positive negative samples bdda provides theoretical solution determining attributes. Experimental results three show effectiveness attributes discovered framework.