发表机构
Technical University of Munich (TUM); Munich Center for Machine Learning (MCML); Imperial College London; Florida State University; Institute for Advanced Study, TUM (TUM-IAS); TUM University Hospital（慕尼黑工业大学; 慕尼黑机器学习中心; 帝国理工学院; 佛罗里达州立大学; 慕尼黑工业大学高等研究院; 慕尼黑工业大学医院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对大规模多模态MRI的存储与I/O问题，采用JPEG2000或JPEG-LS压缩三维脑肿瘤MRI，在20:1压缩率下训练Wavelet Flow Matching模型，证实其合成质量与未压缩数据训练的模型相当，为可扩展三维MRI生成建模提供了实用方案。
AI 中文摘要
大规模多模态MRI数据集带来了巨大的存储和I/O成本，限制了在商用基础设施上训练三维生成模型。虽然有损压缩被证实能保持判别式分割网络的准确性，但它对生成模型的影响尚未被探索——生成模型需要学习完整的数据分布，而非决策边界。本研究探讨标准图像编解码器能否有效压缩语义丰富的脑肿瘤MRI，同时保留训练和部署三维MRI生成模型所需的保真度。每个三维体积分别用JPEG2000或近无损JPEG-LS流水线压缩；随后，基于BraTS图像序列（T1n、T1c、T2、T2f）条件化的Wavelet Flow Matching模型在压缩数据上训练，所得模型在验证集上评估。在20:1的压缩率下，合成质量与未压缩数据训练的模型在预先设定的边际内统计等效（ΔPSNR<1dB，ΔSSIM<0.02；配对TOST p=[[p]]）：跨模态平均PSNR为27.3dB对比27.0dB，平均SSIM为0.95对比0.96。结果表明，JPEG2000压缩是实现可扩展三维MRI生成建模的实用步骤，且不会降低合成质量。代码库可在该https URL获取。
英文摘要
Large-scale multi-modal MRI datasets impose substantial storage and I/O costs, limiting the training of 3D generative models on commodity infrastructure. While lossy compression is known to preserve accuracy for discriminative segmentation networks, its effect on generative models, which must learn the full data distribution rather than a decision boundary, is unexplored. We study whether standard image codecs can effectively compress semantically rich brain tumor MRI while preserving the fidelity required to train and deploy a 3D MRI generative model. Each 3D volume is compressed with JPEG2000 or a near-lossless JPEG-LS pipeline. Next, a Wavelet Flow Matching model, conditioned on BraTS image sequences (T1n, T1c, T2, T2f), is trained on compressed data, and the resulting models are evaluated on the validation set. At a 20:1 compression ratio, synthesis quality is statistically equivalent to a model trained on uncompressed data within a pre-specified margin ($Δ$PSNR $<1$,dB, $Δ$SSIM $<0.02$; paired TOST $p=[[p]]$): mean PSNR is 27.3,dB vs. 27.0,dB and mean SSIM is 0.95 vs. 0.96 across modalities. Our results indicate that JPEG2000 compression is a practical step toward scalable 3D MRI generative modeling without degrading synthesis quality. The codebase is available at https://github.com/lisafis/MRIComp4Flow .
CommentsAccepted: MICCAI 2026 SASHIMI workshop
