English

应用科学学报 ›› 2025, Vol. 43 ›› Issue (4): 672-683.doi: 10.3969/j.issn.0255-8297.2025.04.009

• 信号与信息处理 • 上一篇    

基于分割分类多任务学习的轻量化脑血管造影质量评估模型

黄逸凡1, 陆小锋1,2, 孙军3, 唐嘉吕3, 刘学锋1   

  1. 1. 上海大学 通信与信息工程学院, 上海 200444;
    2. 上海大学温州研究院, 浙江 温州 325000;
    3. 温州市中心医院, 浙江 温州 325000
  • 收稿日期:2024-01-07 发布日期:2025-07-31
  • 通信作者: 陆小锋,高级实验师,研究方向为信号与信息处理。E-mail:luxiaofeng@shu.edu.cn E-mail:luxiaofeng@shu.edu.cn
  • 基金资助:
    上海市科委科研计划基金（No.22511103403,No.22511103304）；上达转化医学基金（No.SDTMF2022KP04）

Lightweight Cerebral Angiography Quality Assessment Model Based on Segmentation-Classification Multi-task Learning

HUANG Yifan1, LU Xiaofeng1,2, SUN Jun3, TANG Jialü3, LIU Xuefeng1   

  1. 1. School of Communication and Information Engineering, Shanghai University, Shanghai 200444, China;
    2. Wenzhou Institute of Shanghai University, Wenzhou 325000, Zhejiang, China;
    3. Wenzhou Central Hospital, Wenzhou 325000, Zhejiang, China
  • Received:2024-01-07 Published:2025-07-31

PDF

6

摘要： 为解决脑血管造影人工质控的不稳定和非实时问题，实现实时脑血管造影质量评估，提出了一种轻量化的分割分类多任务学习模型。该模型分为特征提取主干模块、血管分割模块、造影质量分类模块3个部分，并用深度可分离卷积替代传统卷积以降低参数量；提出了一种局部-全局自注意力模块以增强全局信息的提取能力；在血管分割模块中设计了特征聚合模块以优化特征连接。通过结合质量分类模块和分割结果以及主干特征来评估造影质量，并为模型训练设计了联合损失函数。实验结果表明，提出的模型在参数量仅为3.434 2×106的情况下获得了较好的分割与分类性能，质量评估准确率达到0.8182，且实时性高。

关键词: 脑血管造影, 多任务学习, 局部-全局自注意力, 特征聚合

Abstract: To address the instability and non-real-time problems of manual quality control in cerebral angiography and realize real-time quality assessment, a lightweight segmentation-classification multi-task learning (MTL) model is proposed. This model comprises three main components: a feature extraction backbone module, a blood vessel segmentation module, and an angiography quality classification module. Depth-separable convolution is employed instead of traditional convolution to reduce the number of parameters. Additionally, a local-global self-attention module (L-GSAM) is proposed to enhance the model’s ability to extract global information. A feature aggregation module (FAM) is introduced in the vessel segmentation module to optimize feature connections. The segmentation results are then combined with backbone features to assess the quality of angiography in the classification module, and a joint loss function is designed for model training. Experimental results show that the proposed model achieves good segmentation and classification performance with only 3.4342×106 parameters, the quality assessment accuracy reaches 0.818 2, and the model exhibits high real-time performance.

Key words: cerebral angiography, multi-task learning, local-global self-attention, feature aggregation

中图分类号: