Development of a clinically oriented system for melanoma diagnosis
作者：
Highlights：
• A clinically computer aided diagnosis system is proposed for melanoma diagnosis.
• The system diagnoses the lesions using features that have a medical meaning.
• The system learns to detect various relevant dermoscopic criteria using text labels.
• An image annotation method is used to associate image regions with the criteria.
• Feature fusion is used to combine medical information and diagnose the lesions.
摘要
•A clinically computer aided diagnosis system is proposed for melanoma diagnosis.•The system diagnoses the lesions using features that have a medical meaning.•The system learns to detect various relevant dermoscopic criteria using text labels.•An image annotation method is used to associate image regions with the criteria.•Feature fusion is used to combine medical information and diagnose the lesions.
论文关键词：Melanoma diagnosis,Computer aided diagnosis system,Detection of dermoscopic structures,Image annotation,Corr-LDA,Feature fusion
论文评审过程：Received 10 January 2017, Revised 24 March 2017, Accepted 17 April 2017, Available online 25 April 2017, Version of Record 4 May 2017.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.04.023
