---
tags:
- Multi-task Learning
- Explainable AI
- Prognosis
---
# Multi-task Learning Approach for Intracranial Hemorrhage Prognosis

This repository contains the checkpoints of deep learning intracranial hemorrhage (ICH) prognosis models, used in the paper: 'Multi-task Learning Approach for Intracranial Hemorrhage Prognosis'. 
Particularly, it contains the checkpoints of our proposed method, where we introduced a multi-task learning approach to jointly predict ICH, Glasgow Coma Scale and age, improving performance and interpretability. 

The checkpoints are trained on the open-source Head-CT 2D/3D images with and without ICH prepared for Deep Learning dataset. You can access the dataset at this [link](https://digital.csic.es/handle/10261/275792).

Please refer to the [original GitHub repository](https://github.com/Trustworthy-AI-UU-NKI/MultitaskLearning_ICH_Prognosis) for the code.