多区域和多图像卷积神经网络模型用于检测胃杆菌感染
Jie Dai1, Zhijian Li2, Xigang Zhang3
1Suzhou Wellomen Information Technology Co., Ltd, Suzhou, China.
Surgical endoscopy
|December 1, 2025
概括
一个新的多区域,多图像卷积神经网络 (CNN) 显著提高了人工智能 (AI) 检测胃杆菌 (HP) 感染的准确性. 这种先进的AI模型通过整合多区域胃数据来提高诊断能力,以实现更可靠的HP检测.
科学领域:
- 医疗诊断中的人工智能
- 机器学习用于传染病检测和检测.
- 计算病理学和成像分析分析
背景情况:
- 目前用于胃杆菌 (HP) 感染检测的AI模型仅限于单图像分析.
- 现有的方法缺乏整合多区域胃数据,这可能会降低诊断准确度.
- 需要先进的人工智能模型,这些模型包含全面的胃成像数据,以改进HP检测.
研究的目的:
- 开发和评估一个多区域,多图像卷积神经网络 (CNN) 模型,用于增强胃HP感染诊断.
- 将拟议的多区域CNN模型的诊断性能与传统的单图像CNN模型进行比较.
- 利用来自多家医院的外部验证数据集,评估多区域CNN模型的通用性.
主要方法:
- 来自南方医院的5169例病例 (104437张图像) 的数据集被分为训练 (80%) 和测试 (20%) 集.
- 单图像CNN和已开发的多区域,多图像CNN模型都经过了HP感染诊断的培训和测试.
- 对来自三个独立医院的696例病例 (20,948张图像) 进行了外部验证,以评估模型的概括性.
主要成果:
- 在内部测试组中,与单图像CNN相比,多区域CNN实现了更高的准确性 (95.1% vs. 93.3%) 和灵敏性 (96.5% vs. 94.4%).
- 外部验证显示,多区域CNN的性能明显改善,准确度为89.7%与77.6% (P < 0.01) 相比,灵敏度为90.2%与82.4% (P < 0.01).
- 多区域CNN模型在内部和外部验证数据集中显示出优越的特异性和曲线下面面积 (AUC) 值.
结论:
- 开发的多区域,多图像CNN模型显著提高了基于AI的胃HP感染诊断的准确性,灵敏性和特异性.
- 该模型表现出卓越的概括性,在来自不同医院来源的外部数据集上表现优于单图像CNN.
- 这种方法代表了医疗成像AI的重大进步,为胃HP感染提供了更好的诊断能力.
相关概念视频
Treating Helicobacter pylori in Peptic Ulcers: Antimicrobial Therapy
1.1K
Helicobacter pylori, a resilient gram-negative bacterium, can thrive in the stomach's harsh, acidic environment. Infection with H. pylori leads to a cascade of events within the stomach lining. One of the critical disruptions caused by this bacterium is the interference with somatostatin production, a hormone responsible for regulating acid secretion. This interference tips the balance, escalating acid secretion and diminishing bicarbonate levels. This imbalance compromises the defensive...
1.1K
Peptic Ulcer Disease III: Clinical Manifestations and Diagnostic Studies
507
Peptic ulcer disease (PUD) presents with diverse symptoms depending on the location and severity of the ulcer. Clinical manifestations of peptic ulcer include dull pain and a burning sensation in the mid-epigastric region.
Few clinical manifestations differentiate gastric ulcers from duodenal ulcers. Distinctions in the location, timing, and pain relief are crucial for healthcare providers in differentiating between gastric and duodenal ulcers during clinical assessments.
Few clinical manifestations differentiate gastric ulcers from duodenal ulcers. Distinctions in the location, timing, and pain relief are crucial for healthcare providers in differentiating between gastric and duodenal ulcers during clinical assessments.
507


