SimChest: a novel anomaly agnostic model for similarity measurement in follow-up chest radiograph pairs via a supervised contrastive learning model
This is a PyTorch implementation of the LDH under reivew paper:
CUDA_VISIBLE_DEVICES=0,1,2,3 python main_supcon.py --dataset real --name SupCon_scratch \
--print_freq=5 --save_freq 1 --num_workers 8 --aug True --warm \
--batch_size 8 --model resnet50 --method SupCon --epochs 100
If you want to get CXR image similarity logit, run the code below.
python model_inference.py
Page: https://mi2rl.co
Email: kjcho.amc@gmail.com
@Article{khosla2020supervised,
title = {Supervised Contrastive Learning},
author = {Prannay Khosla and Piotr Teterwak and Chen Wang and Aaron Sarna and Yonglong Tian and Phillip Isola and Aaron Maschinot and Ce Liu and Dilip Krishnan},
journal = {arXiv preprint arXiv:2004.11362},
year = {2020},
}