Longitudinal infant functional connectome prediction and fingerprinting

Framework:

Papers:

This repository provides a PyTorch implementation of the models adopted in the following papers:

• Hu, D., et al. "Disentangled intensive triplet autoencoder for infant functional connectome fingerprinting." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 2020.
• Yu, X., et al. "Longitudinal infant functional connectivity prediction via conditional intensive triplet network." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 2022.

Code:

main_pcc_mae.py

You need to run this file to start. The hyper-parameters of loss weight can be defined in this file.

Run python ModelCode\main_pcc_mae.py.

NNFunctions.py

We created a triple construction method in this file.

Tested with:

• PyTorch 1.9.0
• Python 3.7.0

Citation:

If you used the code or data of this project, please cite:

@inproceedings{hu2020disentangled,
title={Disentangled intensive triplet autoencoder for infant functional connectome fingerprinting},
author={Hu, Dan and Wang, Fan and Zhang, Han and Wu, Zhengwang and Wang, Li and Lin, Weili and Li, Gang and Shen, Dinggang and UNC/UMN Baby Connectome Project Consortium},
booktitle={Medical Image Computing and Computer Assisted Intervention--MICCAI 2020: 23rd International Conference, Lima, Peru, October 4--8, 2020, Proceedings, Part VII 23},
pages={72--82},
year={2020},
organization={Springer}
}

@inproceedings{yu2022longitudinal,
title={Longitudinal infant functional connectivity prediction via conditional intensive triplet network},
author={Yu, Xiaowei and Hu, Dan and Zhang, Lu and Huang, Ying and Wu, Zhengwang and Liu, Tianming and Wang, Li and Lin, Weili and Zhu, Dajiang and Li, Gang},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={255--264},
year={2022},
organization={Springer}
}