Neural networks approach for mammography diagnosis using wavelets features
This is an incremental improvement for medical imaging diagnosis, specifically targeting mammogram classification for tumor type and risk level.
The authors tackled mammogram diagnosis by developing a supervised system that uses wavelet multilevel decomposition to extract features and artificial neural networks for classification, reporting enhanced results compared to their previous study.
A supervised diagnosis system for digital mammogram is developed. The diagnosis processes are done by transforming the data of the images into a feature vector using wavelets multilevel decomposition. This vector is used as the feature tailored toward separating different mammogram classes. The suggested model consists of artificial neural networks designed for classifying mammograms according to tumor type and risk level. Results are enhanced from our previous study by extracting feature vectors using multilevel decompositions instead of one level of decomposition. Radiologist-labeled images were used to evaluate the diagnosis system. Results are very promising and show possible guide for future work.