LoRe-MIM: Lost-or-Replaced Masked Image Modeling for Self-Supervised Representation Learning in Whole-Body PET/CT
作者: Amartya Bhattacharya, Bashirul Azam Biswas, Biratal Raj Wagle, Michael S. Leapman, Matthew E. Maeder, James B. Yu, Indrani Bhattacharya
分类: cs.CV
发布日期: 2026-10-29
备注: 25 pages, 9 figures
💡 一句话要点
提出LoRe-MIM以解决全身PET/CT图像自监督学习问题
🎯 匹配领域: 支柱二:RL算法与架构 (RL & Architecture) 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 自监督学习 多模态学习 医学影像 PET/CT 深度学习 图像重建 疾病检测
📋 核心要点
- 现有的全身PET/CT模型在专家注释的时间和成本上存在显著瓶颈,限制了其广泛应用。
- 本文提出的LoRe-MIM框架通过掩蔽图像重建,利用多模态信息学习更为有效的表示。
- 实验结果显示,LoRe-MIM在多项任务中超越了现有的最先进方法,展现出良好的性能提升。
📝 摘要(中文)
全身正电子发射计算机断层扫描(PET/CT)深度学习模型在疾病检测、治疗规划和反应评估中具有重要的临床潜力。然而,专家注释的时间和成本是开发稳健且可推广模型的主要瓶颈。尽管自监督学习(SSL)已成为缓解注释稀缺的有希望的方法,但由于缺乏大规模数据集和学习互补跨模态表示的挑战,其在全身PET/CT中的应用仍相对未被充分探索。为了解决这些限制,本文构建了一个异质的多示踪剂多机构队列,包含6531个全身PET/CT扫描,并提出了LoRe-MIM(Lost-Or-Replaced Masked Image Modelling),一种通过掩蔽图像重建学习多模态表示的新型自监督框架。LoRe-MIM引入了一种掩蔽策略,其中图像区域被替换为零(Lost)或用相应的互补成像模态替代(Replaced),鼓励模型捕捉模态特征和跨模态关系。实验结果表明,LoRe-MIM在多个下游任务中始终优于相应的最先进基线,突显了其作为全身PET/CT分析的可扩展和可推广自监督学习框架的潜力。
🔬 方法详解
问题定义:本文旨在解决全身PET/CT图像自监督学习中的注释稀缺问题,现有方法在跨模态表示学习上存在困难。
核心思路:LoRe-MIM通过引入掩蔽策略,利用图像区域的丢失或替换来促进模型学习模态特征和跨模态关系,从而提升表示学习的效果。
技术框架:该框架包括数据预处理、掩蔽策略实施、模型训练和下游任务评估等主要模块,整体流程旨在通过自监督学习优化模型性能。
关键创新:LoRe-MIM的主要创新在于其独特的掩蔽策略,结合了丢失和替换两种方式,显著提升了模型对多模态信息的理解能力,与传统方法相比具有本质区别。
关键设计:在模型设计中,采用了特定的损失函数以平衡模态特征的学习,同时在网络结构上进行了优化,以适应多模态数据的处理需求。具体参数设置和网络结构细节在实验部分进行了详细描述。
🖼️ 关键图片
📊 实验亮点
实验结果表明,LoRe-MIM在多模态PET/CT图像重建、癌症与非癌症的分类、全身病灶分割等任务中均显著优于现有最先进基线,具体提升幅度达到XX%,展示了其强大的学习能力和应用潜力。
🎯 应用场景
该研究的潜在应用领域包括医学影像分析、疾病早期检测和个性化治疗规划等。通过提高全身PET/CT图像的分析能力,LoRe-MIM有望在临床实践中实现更高效的疾病管理和治疗效果评估,未来可能推动相关领域的技术进步。
📄 摘要(原文)
Deep learning models for whole-body positron emission tomography/computed tomography (PET/CT) hold significant clinical potential for disease detection, treatment planning, and response assessment. However, the time and cost associated with expert annotations remain major bottlenecks in developing robust and generalizable models. Although self-supervised learning (SSL) has emerged as a promising approach to mitigating annotation scarcity, its application to whole-body PET/CT remains relatively underexplored due to the limited availability of large-scale datasets and the challenges associated with learning complementary cross-modal representations. To address these limitations, we curate a heterogeneous, multi-tracer, multi-institutional cohort comprising 6,531 whole-body PET/CT scans and propose LoRe-MIM (Lost-Or-Replaced Masked Image Modelling), a novel self-supervised framework designed to learn multimodal representations through masked image reconstruction. Specifically, LoRe-MIM introduces a masking strategy in which image regions are either replaced with zeros (Lost) or substituted with corresponding complementary imaging modality (Replaced), encouraging the model to capture both modality-specific characteristics and cross-modal relationships. We evaluate the learned representations across multiple downstream tasks, including multimodal PET/CT image reconstruction, patch-based classification into cancer vs. non-cancer, whole-body lesion segmentation, label-efficient training, and robustness assessment. Experimental results demonstrate that LoRe-MIM consistently outperforms the corresponding state-of-the-art baselines across these tasks, highlighting its potential as a scalable and generalizable self-supervised learning framework for whole-body PET/CT analysis.