A versatile foundation model for cine cardiac magnetic resonance image analysis tasks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Fu, Yunguan, Bai, Wenjia, Yi, Weixi, Manisty, Charlotte, Bhuva, Anish N, Treibel, Thomas A, Moon, James C, Clarkson, Matthew J, Davies, Rhodri Huw, Hu, Yipeng
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915472079847424
author Fu, Yunguan
Bai, Wenjia
Yi, Weixi
Manisty, Charlotte
Bhuva, Anish N
Treibel, Thomas A
Moon, James C
Clarkson, Matthew J
Davies, Rhodri Huw
Hu, Yipeng
author_facet Fu, Yunguan
Bai, Wenjia
Yi, Weixi
Manisty, Charlotte
Bhuva, Anish N
Treibel, Thomas A
Moon, James C
Clarkson, Matthew J
Davies, Rhodri Huw
Hu, Yipeng
contents Here we present a versatile foundation model that can perform a range of clinically-relevant image analysis tasks, including segmentation, landmark localisation, diagnosis, and prognostication. A multi-view convolution-transformer masked autoencoder, named as CineMA, was trained on 15 million cine images from 74,916 subjects. The model was validated on multiple image analysis tasks and compared to existing models on >4,500 images from eight independent datasets with diverse population characteristics, representing the largest benchmark study for cine CMR so far. CineMA consistently outperformed conventional convolutional neural networks (CNNs) in delineating ventricular boundaries and estimating ejection fraction, a key measure of cardiac function. The improved performance was preserved, even when the model only used half of fine-tuning data. CineMA also surpassed CNNs in disease detection and matched their performance in long-axis function measurement. Interestingly, we found that CineMA can also detect cardiac changes in systemic diseases, such as diabetes, hypertension and cancer, and can also predict mortality. Finally, we assessed model fairness and demonstrated consistent model performance across demographic subgroups. These findings highlight CineMA's accuracy, learning efficiency, adaptability, and fairness, underscoring its potential as a foundation model for automated cardiac image analysis to support clinical workflow and cardiovascular research. All training and inference code and models are made publicly available at https://github.com/mathpluscode/CineMA.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00679
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A versatile foundation model for cine cardiac magnetic resonance image analysis tasks
Fu, Yunguan
Bai, Wenjia
Yi, Weixi
Manisty, Charlotte
Bhuva, Anish N
Treibel, Thomas A
Moon, James C
Clarkson, Matthew J
Davies, Rhodri Huw
Hu, Yipeng
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Here we present a versatile foundation model that can perform a range of clinically-relevant image analysis tasks, including segmentation, landmark localisation, diagnosis, and prognostication. A multi-view convolution-transformer masked autoencoder, named as CineMA, was trained on 15 million cine images from 74,916 subjects. The model was validated on multiple image analysis tasks and compared to existing models on >4,500 images from eight independent datasets with diverse population characteristics, representing the largest benchmark study for cine CMR so far. CineMA consistently outperformed conventional convolutional neural networks (CNNs) in delineating ventricular boundaries and estimating ejection fraction, a key measure of cardiac function. The improved performance was preserved, even when the model only used half of fine-tuning data. CineMA also surpassed CNNs in disease detection and matched their performance in long-axis function measurement. Interestingly, we found that CineMA can also detect cardiac changes in systemic diseases, such as diabetes, hypertension and cancer, and can also predict mortality. Finally, we assessed model fairness and demonstrated consistent model performance across demographic subgroups. These findings highlight CineMA's accuracy, learning efficiency, adaptability, and fairness, underscoring its potential as a foundation model for automated cardiac image analysis to support clinical workflow and cardiovascular research. All training and inference code and models are made publicly available at https://github.com/mathpluscode/CineMA.
title A versatile foundation model for cine cardiac magnetic resonance image analysis tasks
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.00679