Improving classification performance of breast lesions on ultrasonography

作者：

Highlights：

• 1491 features are evaluated for classifying breast lesions on ultrasound.

• Feature selection is based on mutual information and statistical tests.

• 5 morphological and 4 texture features achieve the best classification performance.

• 11 feature sets from the literature are surpassed by the 5 morphological features.

摘要

Highlights•1491 features are evaluated for classifying breast lesions on ultrasound.•Feature selection is based on mutual information and statistical tests.•5 morphological and 4 texture features achieve the best classification performance.•11 feature sets from the literature are surpassed by the 5 morphological features.

论文关键词：Ultrasonography,Breast lesions,Computer-aided diagnosis,Feature selection,Classification performance

论文评审过程：Received 19 December 2013, Revised 5 May 2014, Accepted 9 June 2014, Available online 18 June 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.06.006