南方医科大学学报

南方医科大学学报 ›› 2025, Vol. 45 ›› Issue (1): 197-205.doi: 10.12122/j.issn.1673-4254.2025.01.23

• • 上一篇    

手工提取的视觉特征与深度特征的融合模型用于消化性溃疡再出血风险分级

周沛珊1,2(), 阳维1, 李青原2, 郭小芳3, 傅蓉1(), 刘思德2()   

  1. 1.南方医科大学生物医学工程学院，广东 广州 510515
    2.南方医科大学南方医院消化内科，广东 广州 510515
    3.赣州市人民医院消化内科，江西 赣州 341000
  • 收稿日期:2024-09-30 出版日期:2025-01-20 发布日期:2025-01-20
  • 通讯作者: 傅蓉,刘思德 E-mail:2279626154@qq.com;furong@smu.edu.cn;liuside2011@163.com
  • 作者简介:周沛珊，在读硕士研究生，E-mail: 2279626154@qq.com
  • 基金资助:
    广东省重点领域研发计划项目(2022B0303020003);广州市科技计划项目(2024A04J9948)

A fusion model of manually extracted visual features and deep learning features for rebleeding risk stratification in peptic ulcers

Peishan ZHOU1,2(), Wei YANG1, Qingyuan LI2, Xiaofang GUO3, Rong FU1(), Side LIU2()   

  1. 1.School of Biomedical Engineering, Southern Medical University, Guangzhou 510515, China
    2.Department of Gastroenterology, Nanfang Hospital, Southern Medical University, Guangzhou 510515, China
    3.Department of Gastroenterology, Ganzhou People's Hospital, Ganzhou 341000, China
  • Received:2024-09-30 Online:2025-01-20 Published:2025-01-20
  • Contact: Rong FU, Side LIU E-mail:2279626154@qq.com;furong@smu.edu.cn;liuside2011@163.com

RichHTML

2

PDF

139

摘要：

目的 提出一种基于电子内镜图像的多特征融合模型，结合深度学习与手工特征的优势，用于消化性溃疡再出血风险的分级。 方法 根据溃疡的内镜表现，提取颜色特征以区分活动性出血（Forrest I）与非出血溃疡（Forrest II、III），并利用边缘和纹理特征描述不同级别溃疡的形态与外观。通过融合深度学习网络提取的深度特征与手工提取的视觉特征，形成电子内镜图像的多特征表达，最终用于预测消化性溃疡的再出血风险。 结果 在包含708例患者、3573张图像的Forrest分级数据集上，提出的多特征融合模型在消化性溃疡再出血风险六分级任务中取得了74.94%的准确率，优于进修医生59.9%的分级准确性（P<0.05）。在Ib、IIa和III级溃疡的识别中，F1得分为90.16%、75.44%和77.13%，其中Ib级表现尤为突出。与首个进行溃疡再出血分级研究的模型相比，提出模型的F1得分提升了5.8%。在简化的3类风险分级任务中，模型在高风险、低风险和无需内镜治疗级别上的F1得分为93.74%、81.30%和73.59%。 结论 本文提出的多特征融合模型有效融合卷积神经网络（CNN）提取的深度特征与手工提取的视觉特征，提升了消化性溃疡再出血风险分级的准确性，为临床提供了高效的诊断辅助工具。

关键词: 电子内镜图像, 多特征融合, 消化性溃疡, Forrest再出血风险分级

Abstract:

Objective We propose a multi-feature fusion model based on manually extracted features and deep learning features from endoscopic images for grading rebleeding risk of peptic ulcers. Methods Based on the endoscopic appearance of peptic ulcers, color features were extracted to distinguish active bleeding (Forrest I) from non-bleeding ulcers (Forrest II and III). The edge and texture features were used to describe the morphology and appearance of the ulcers in different grades. By integrating deep features extracted from a deep learning network with manually extracted visual features, a multi-feature representation of endoscopic images was created to predict the risk of rebleeding of peptic ulcers. Results In a dataset consisting of 3573 images from 708 patients with Forrest classification, the proposed multi-feature fusion model achieved an accuracy of 74.94% in the 6-level rebleeding risk classification task, outperforming the experienced physicians who had a classification accuracy of 59.9% (P<0.05). The F1 scores of the model for identifying Forrest Ib, IIa, and III ulcers were 90.16%, 75.44%, and 77.13%, respectively, demonstrating particularly good performance of the model for Forrest Ib ulcers. Compared with the first model for peptic ulcer rebleeding classification, the proposed model had improved F1 scores by 5.8%. In the simplified 3-level risk (high-risk, low-risk, and non-endoscopic treatment) classification task, the model achieved F1 scores of 93.74%, 81.30%, and 73.59%, respectively. Conclusions The proposed multi-feature fusion model integrating deep features from CNNs with manually extracted visual features effectively improves the accuracy of rebleeding risk classification for peptic ulcers, thus providing an efficient diagnostic tool for clinical assessment of rebleeding risks of peptic ulcers.

Key words: endoscopic images, multi-feature fusion, peptic ulcer, Forrest rebleeding risk classification