jeya-maria-jose/Medical-Transformer
A transformer-based deep learning model with gated axial attention for medical image segmentation, published at MICCAI 2021.
Not currently ranked — collecting fresh signals.
star history
This repository provides PyTorch implementations of the Medical Transformer (MedT) and Gated Axial Attention U-Net for medical image segmentation. The model introduces a gating mechanism to control self-attention, addressing the challenge of training transformers on smaller medical imaging datasets. It employs a Local-Global (LoGo) training strategy that operates on both whole images and patches to learn complementary features.
Frequently asked
- What is jeya-maria-jose/Medical-Transformer?
- A transformer-based deep learning model with gated axial attention for medical image segmentation, published at MICCAI 2021.
- Is Medical-Transformer open source?
- Yes — jeya-maria-jose/Medical-Transformer is open source, released under the MIT license.
- What language is Medical-Transformer written in?
- jeya-maria-jose/Medical-Transformer is primarily written in Python.
- How popular is Medical-Transformer?
- jeya-maria-jose/Medical-Transformer has 861 stars on GitHub.
- Where can I find Medical-Transformer?
- jeya-maria-jose/Medical-Transformer is on GitHub at https://github.com/jeya-maria-jose/Medical-Transformer.