Machine Learning Project

In this project, we were given a training dataset of ~200 3D MRI brain scans. Throughout the three milestones, we had to predict gender, age & cognitive disabilities.

We used various features:

1. A principle component analysis (PCA) to reduce the dimensionality
2. Histograms of values of partitioned brains
3. Histograms of canny edges of partitioned brains

We tried out several ML methods from sklearn such as random forests, SVM (kernelized) or KNN. We used cross validation to tune the resp. hyper parameters.