Hi,
I am Sourabh Jain, I am interested in the project. I have good experience with machine learning algorithms like ANN and Bayesian method. In this project, I will use labeled legion map as input for training of algorithm. I will use, most likely, ANN (feedforward with backpropagation) for purpose of classification. After classification, feature extraction will be performed using voxel intensities on T1, T2 and FLAIR. This data is required to be normalized, which is a tricky part as further processing (data smoothing etc) will depend upon it. I am using following references for this project. " A Comparison of Supervised Machine Learning Algorithms and Feature Vectors for MS Lesion Segmentation Using Multimodal Structural MRI" by Sweeney et al AND "White Matter Lesion Segmentation Using Machine Learning and Weakly Labeled MR Images" by Xie and Tao. Please let me know if you have other references. If you would like to discuss this further, please let me know.
Regards
Sourabh