东北大学学报(自然科学版) ›› 2026, Vol. 47 ›› Issue (1): 11-19.DOI: 10.12068/j.issn.1005-3026.2026.20259020

• 智慧医疗专栏 • 上一篇    下一篇

医学图像压缩与视觉任务联合优化方法

姚超1(), 高梓轩2, 陈俊如3, 卢奕鹏4   

  1. 1.北京科技大学 计算机与通信工程学院,北京 100083
    2.北京交通大学 计算机科学与技术学院,北京 100044
    3.无锡学院 集成电路科学与工程学院,江苏 无锡 214105
    4.北京大学 集成电路学院,北京 100871
  • 收稿日期:2025-06-06 出版日期:2026-01-15 发布日期:2026-03-17
  • 通讯作者: 姚超
  • 基金资助:
    国家自然科学基金资助项目(62372036);国家自然科学基金资助项目(62120106009);国家自然科学基金资助项目(62332017);国家自然科学基金资助项目(U24B20179);国家自然科学基金资助项目(U22A2022)

Joint Optimization Approach for Medical Image Compression and Vision Tasks

Chao YAO1(), Zi-xuan GAO2, Jun-ru CHEN3, Yi-peng LU4   

  1. 1.School of Computer & Communication Engineering,University of Science and Technology Beijing,Beijing 100083,China
    2.School of Computer Science & Technology,Beijing Jiaotong University,Beijing 100044,China
    3.School of Integrated Circuit Science and Engineering,Wuxi University,Wuxi 214105,China
    4.School of Integrated Circuits,Peking University,Beijing 100871,China. cn
  • Received:2025-06-06 Online:2026-01-15 Published:2026-03-17
  • Contact: Chao YAO

摘要:

针对医学图像处理中依赖独立编码组件无法实现数据压缩与机器视觉任务联合优化的问题,本文构建了一种端到端的机器视觉任务驱动的医学图像压缩网络(machine vision task-driven medical image compression network,MVMICNet)模型,端到端地实现数据压缩与医学图像分析的和谐统一.为了保持医学图像压缩前后机器视觉任务的性能,设计了任务感知的改进码率-准确率损失函数,通过引入任务相关的损失项,在优化过程中动态平衡码率、重建图像失真与机器视觉任务精度三者之间的关系;同时,MVMICNet模型采用分阶段训练的模式,针对机器视觉任务的不同特性进行特定的优化,确保了模型能够精准捕获对诊断至关重要的特征信息,实现了压缩效率与任务性能的同步提升,从而在复杂的医学应用场景中展现出更优越的鲁棒性;最终,本文在语义分割和目标检测任务中验证了该框架的有效性.

关键词: 医学图像压缩, 语义分割, 目标检测, 卷积神经网络(CNN), 任务驱动优化

Abstract:

In medical image processing, the reliance on independent encoding components makes it impossible to achieve joint optimization of data compression and machine vision tasks. To address this issue, an end-to-end machine vision task-driven medical image compression network (MVMICNet) was proposed, achieving harmonious unification of data compression and medical image analysis in an end-to-end manner. To maintain the performance of machine vision tasks before and after medical image compression, a task-aware improved code rate-accuracy loss function was designed. By introducing task-related loss terms, it dynamically balanced the relationship among code rate, reconstructed image distortion, and machine vision task accuracy during the optimization process. Furthermore, the MVMICNet model adopted a stage-wise training approach, specifically optimizing for the different characteristics of machine vision tasks to ensure that the model can accurately capture the feature information crucial for diagnosis. This has achieved a simultaneous improvement in compression efficiency and task performance, thus demonstrating superior robustness in complex medical application scenarios. Finally, the effectiveness of the framework was verified in semantic segmentation and object detection tasks.

Key words: medical image compression, semantic segmentation, object detection, convolutional neural network (CNN), task-driven optimization

中图分类号: