Decoupling multi-task causality for improved skin lesion segmentation and classification

作者：

Highlights：

• Decouple the causality among skin lesion classification, detection and segmentation from the Pareto efficiency view, solving the common trade-off dilemma in multiple tasks.

• New paradigm boosts the diagnosis with the investigated favorable fusing manner.

• Effective model compression scheme overcomes the redundant complexity, achieving the raised performance under fewer parameters.

摘要

•Decouple the causality among skin lesion classification, detection and segmentation from the Pareto efficiency view, solving the common trade-off dilemma in multiple tasks.•New paradigm boosts the diagnosis with the investigated favorable fusing manner.•Effective model compression scheme overcomes the redundant complexity, achieving the raised performance under fewer parameters.

论文关键词：Skin lesion analysis,Multi-task decoupled,Deep learning,Task causality

论文评审过程：Received 1 March 2022, Revised 16 July 2022, Accepted 20 August 2022, Available online 28 August 2022, Version of Record 1 September 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.108995