Computational and Mathematical Methods in Medicine / 2017 / Article / Tab 19 / Review Article
Involvement of Machine Learning for Breast Cancer Image Classification: A Survey Table 19 -means Cluster Algorithm and Self-Organizing Map for breast image classification.
Reference Descriptor Image type Number of images Key findings Moftah et al. [142 ] Intensity distribution used as feature.MRI — Three types of evaluation measures performed: (a) Accuracy, (b) feature based, (c) shape based measure. This can classify the data as well as identify the target. The obtained best Accuracy of the segmented ROI is 90.83%Lee et al. [143 ] 1734 signal patterns.MRI 322 Available signal patterns have been classified into 10 classes.Dalmiya et al. [144 ] Discrete Wavelet Transform.Mammogram — Cancer tumor masses have been segmented.Elmoufidi et al. [145 ] Local Binary Pattern.Mammogram 322 Image enhancing. Generation of number of clusters Detection of regions of interest. Mean detection of regions of interest is 85.00%.Samundeeswari et al. [146 ] Ultrasound — Utilizing ant colony and regularization parameters. This method obtained 96.00% similarity between segmented and reference tumors. Rezaee [147 ] Discrete Wavelet Transform. Mammogram 120 Early detection of tumors from the breast image. Tumor detection Accuracy 92.32%, Sensitivity 90.24%.Chandra et al. [148 ] Gray intensity values.Mammogram — Mammogram image has been clustered using SOM along with the Quadratic Neural Network.