Here at http://2016dfdp.blogspot.co.uk/ you can find our Y2 project blog titled "Diagonstics from disease profiling". This project is finished by Nawei.Chen, Jiahuan.Lu, Tianhua.Xu, Jitong.Liu and supervised by Dowsey. Andrew.
This project aims to investigate different signal processing and machine learning methods to find differences between healthy and diseased sample (Cirrhosis and HCC) just by comparing series of images. In the project, the noise signal in images is filtrated by wavelet transform, threshold filtration and inverse wavelet transform. Then, the peaks in image are detected and picked out, where the features are concentrated. After that, a list contains of features' value is generated and different kinds of image could be classified with Mann-Whitney test.
All the process are conducted by MATLAB. The functions are written and combined by the members in project group. To conduct it, some parameter need to be set according to different input samples.