Background intensity independent texture features for assessing breast cancer risk in screening mammograms

Image intensity and texture in screening mammograms are thought to be associated with the risk of breast cancer. Studies on developing automatic breast cancer risk assessment schemes tend to employ texture measures which are correlated to local background intensity. Accordingly, the contribution of texture alone to risk assessment is not known. Here background intensity independent texture measures are used to assess cancer risk. Moreover risk assessment based on background intensity independent texture outperforms intensity dependent texture suggesting that local image background intensity may confound risk assessment. Performance seems to depend on the view of the breast and so suggests that optimizing schemes for different views may improve risk assessment.

► Background intensity independent texture features were proposed for mammogram classification. ► Subspace ensemble k-nearest neighbor classifiers were used in the final classification. ► The superiority of the background intensity independent texture analysis is shown by comparison. ► The performance of CC view images was better than that of other view images.