AI Toolkit for Healthcare Imaging in Keras 3
-
Updated
Sep 24, 2025 - Jupyter Notebook
AI Toolkit for Healthcare Imaging in Keras 3
MPS-accelerated Brain Tumour Segmentation task based on BraTS '21 task dataset. Uses a hybrid Swin-UNETR + CNN architecture. Optimized for Apple Silicon (M-chip) systems using MPS.
Implementation of my Master's thesis "Analyzing Network Architectures and Feature Recovery Methods for Missing Modality Scenarios in Multi-modal Medical Segmentation"
Brain tumor segmentation on BraTS-2023 dataset using nnUNet, SegResNet, and SwinUNETR. Models trained and evaluated on MRI scans to segment tumor sub-regions: WT, TC, ET, and RC, with fold-wise performance visualizations.
Add a description, image, and links to the swin-unetr topic page so that developers can more easily learn about it.
To associate your repository with the swin-unetr topic, visit your repo's landing page and select "manage topics."