Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhu, Xiyue, Kwark, Dou Hoon, Zhu, Ruike, Hong, Kaiwen, Tao, Yiqi, Luo, Shirui, Li, Yudu, Liang, Zhi-Pei, Kindratenko, Volodymyr
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2501.07430
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916602374520832
author Zhu, Xiyue
Kwark, Dou Hoon
Zhu, Ruike
Hong, Kaiwen
Tao, Yiqi
Luo, Shirui
Li, Yudu
Liang, Zhi-Pei
Kindratenko, Volodymyr
author_facet Zhu, Xiyue
Kwark, Dou Hoon
Zhu, Ruike
Hong, Kaiwen
Tao, Yiqi
Luo, Shirui
Li, Yudu
Liang, Zhi-Pei
Kindratenko, Volodymyr
contents In volume-to-volume translations in medical images, existing models often struggle to capture the inherent volumetric distribution using 3D voxelspace representations, due to high computational dataset demands. We present Score-Fusion, a novel volumetric translation model that effectively learns 3D representations by ensembling perpendicularly trained 2D diffusion models in score function space. By carefully initializing our model to start with an average of 2D models as in TPDM, we reduce 3D training to a fine-tuning process and thereby mitigate both computational and data demands. Furthermore, we explicitly design the 3D model's hierarchical layers to learn ensembles of 2D features, further enhancing efficiency and performance. Moreover, Score-Fusion naturally extends to multi-modality settings, by fusing diffusion models conditioned on different inputs for flexible, accurate integration. We demonstrate that 3D representation is essential for better performance in downstream recognition tasks, such as tumor segmentation, where most segmentation models are based on 3D representation. Extensive experiments demonstrate that Score-Fusion achieves superior accuracy and volumetric fidelity in 3D medical image super-resolution and modality translation. Beyond these improvements, our work also provides broader insight into learning-based approaches for score function fusion.
format Preprint
id arxiv_https___arxiv_org_abs_2501_07430
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Introducing 3D Representation for Medical Image Volume-to-Volume Translation via Score Fusion
Zhu, Xiyue
Kwark, Dou Hoon
Zhu, Ruike
Hong, Kaiwen
Tao, Yiqi
Luo, Shirui
Li, Yudu
Liang, Zhi-Pei
Kindratenko, Volodymyr
Computer Vision and Pattern Recognition
Artificial Intelligence
In volume-to-volume translations in medical images, existing models often struggle to capture the inherent volumetric distribution using 3D voxelspace representations, due to high computational dataset demands. We present Score-Fusion, a novel volumetric translation model that effectively learns 3D representations by ensembling perpendicularly trained 2D diffusion models in score function space. By carefully initializing our model to start with an average of 2D models as in TPDM, we reduce 3D training to a fine-tuning process and thereby mitigate both computational and data demands. Furthermore, we explicitly design the 3D model's hierarchical layers to learn ensembles of 2D features, further enhancing efficiency and performance. Moreover, Score-Fusion naturally extends to multi-modality settings, by fusing diffusion models conditioned on different inputs for flexible, accurate integration. We demonstrate that 3D representation is essential for better performance in downstream recognition tasks, such as tumor segmentation, where most segmentation models are based on 3D representation. Extensive experiments demonstrate that Score-Fusion achieves superior accuracy and volumetric fidelity in 3D medical image super-resolution and modality translation. Beyond these improvements, our work also provides broader insight into learning-based approaches for score function fusion.
title Introducing 3D Representation for Medical Image Volume-to-Volume Translation via Score Fusion
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2501.07430