This repository contains PyTorch implementation of "CardiacCLIP: Video-based CLIP Adaptation for LVEF Prediction in a Few-shot Manner" (MICCAI 2025).
Created by Du Yao, Guo Jiarong, Li Xiaomeng*
CardiacCLIP is a novel adaptation of CLIP models for few-shot echocardiogram video analysis, capturing crucial temporal dynamics and localized cardiac structures essential for accurate diagnosis.
-
Multi-Frame Learning (MFL)
An attention-based aggregation mechanism that prioritizes diagnostically relevant frames instead of simple averaging. -
EchoZoom
A multi-scale input representation strategy that enhances modeling of fine-grained cardiac structures.
-
Change the dataset path in
/echoclip/runner/data.py(around line 330). -
Run the training script:
sh scripts/run.shResults and logs will be saved in the results/ and wandb/ folders.
If you find this repository useful, please cite our work:
@article{du2025cardiacclip,
title = {CardiacCLIP: Video-based CLIP Adaptation for LVEF Prediction in a Few-shot Manner},
author = {Du, Yao and Guo, Jiarong and Li, Xiaomeng},
journal = {arXiv preprint arXiv:2509.17065},
year = {2025},
eprint = {2509.17065},
archivePrefix = {arXiv},
primaryClass = {cs.CV}
}