About this document

3D CNN for LGG and HGG Classification by Amol Bankar is a document available to read on EtoBox.

This project utilizes a 3D Convolutional Neural Network (CNN) to classify brain tumors as Low-Grade Glioma (LGG) or High-Grade Glioma (HGG) using multi-modal 3D MRI scans. It includes a preprocessing pipeline, model training, and evaluation scripts, all built with MONAI and PyTorch, and features a ResNet50-based architecture adapted for 3D data. The model demonstrates a strong bias towards predicting HGG, achieving an overall accuracy of approximately 77% on the test set.

Author

Amol Bankar

Language

EN