This is a link to my GitHub. Here is some description of my old projects:


This software is based on my TMI and MICCAI papers [1] and [2]. The idea is to find optimal basis vectors to decompose brain images such that the basis vectors are clinically interpretable and the projection of the image on those basis yields discriminative features. The method is based on as large-scale constrained matrix factorization and tries to reconstruct the images (generative) while producing good classification (discriminative). The software can handle multi-class and unlabeled cases (semi-supervised learning) [3]. (Software website)


The Build system And Software Implementation Standard (BASIS). I contributed to this project a little bit. This project was initially developed as a standard project development tool in SBIA but later evolved into a toolkit that makes developing with multiple languages (eg C++ and MATLAB) easy. For example, if you are developing a MATLAB project and you would like to write a MEX function which calls external libraries (eg ITK, OpenCV) and you are worried about complex dependencies during compilation; or you would like to compile your code using MCC compiler, BASIS helps you to build a CMAKE project. It also has many other functionalities. (Software website)


[1] N. Batmanghelich, B. Taskar, C. Davatzikos, Generative-Discriminative Basis Learning for Medical Imaging. IEEE Trans Med Imaging. 2012 Jan;31(1):51-69. Epub 2011 Jul 25.
[2] N. Batmanghelich, B. Taskar, C. Davatzikos, . Regularized Tensor Factorization for Multi-Modality Medical Image Classification International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2011), Toronto, Canada
[3] N. Batmanghelich, D. Ye, K. Pohl, B. Taskar, C. Davatzikos, Disease Classification and Prediction via Semi-Supervised Dimensionality Reduction, ISBI 2011,