Computational Anatomy Toolbox for SPM12 or SPM25

A brain MRI segmentation tool that provides accurate robust segmentation of problematic brain regions across the neurodegenerative spectrum. The methodology is generalisable to perform well with the typical variance in MRI acquisition parameters and other factors that influence image contrast.

This GitHub repository was created for research focusing on the development of deep learning-based segmentation models for fetal brain tissue.

Survival Prediction of GlioBlastoma Patients using Ensemble architecture of random forest, xgboost and logistic regression classifiers. Uses Optuna for tuning, SMOTE for imbalances, CNN for feature extraction, LDA for feature pre-processing, MPL and Seaborn for visualizations and concordance index as the performance metrics.

U-Net from Scratch for Brain Tumor Segmentation

This project applies segmentation techniques to a set of brain imaging data to identify and analyze different brain regions. The Deep Learning model accurately predicts the midslice of the brain MRI image.