基于CNN的图像识别酸快细菌在唾液涂抹中,用于增强结核病诊断
Chandrakant Kokane1, Neeta A Deshpande2, Nitin Dhawas3
1Department of Computer Science and Engineering (Artificial Intelligence), Vishwakarma Institute of Technology, Pune, Maharashtra, India.
The Indian journal of tuberculosis
|December 16, 2025
概括
这项研究引入了一个卷积神经网络 (CNN),用于在结核病诊断中自动检测酸快细菌 (AFB). 人工智能模型显著提高了准确性和效率,有助于早期发现疾病.
科学领域:
- 医疗成像医学成像
- 人工智能的人工智能
- 微生物学 微生物学
背景情况:
- 结核病 (TB) 仍然是主要的传染病杀手,特别是在低至中等收入国家.
- 酸快细菌 (AFB) 的唾液涂抹显微镜是标准的,但受到变量灵敏度和人为错误的影响.
- 卷积神经网络 (CNN) 的进步为自动和准确的AFB识别提供了潜力.
研究的目的:
- 开发和评估基于CNN的图像识别系统,用于在唾液涂抹图像中自动检测AFB.
- 为了比较不同CNN架构的性能,包括转移学习模型 (ResNet50,VGG16),用于AFB识别.
- 通过人工智能提高结核病检测的诊断准确性和一致性.
主要方法:
- 来自临床实验室的大量高分辨率唾液涂抹图像 (AFB阳性和阴性) 的数据集被策划.
- 专家微生物学家手动标记图像,以创建一个基本真相数据集.
- 使用数据增强技术来提高模型稳定性和概括性.
- 多个CNN模型被训练并使用分割方法验证,采用亚当优化器和二进制交叉损失.
主要成果:
- 转移学习模型ResNet50实现了卓越的性能,AUC-ROC为95.5%,准确率为93.2%,精度为91.7%,回忆率为92.4%.
- 一个定制的CNN模型也以超过90%的精度和强大的F1分数展示了竞争性表现.
- 数据增强和转移学习有效地改善了模型的概括性,并减少了过拟合.
- ResNet50表现出增强的灵敏度和特异性,这对于最小化结核病诊断中的错误阳性至关重要.
结论:
- 开发的CNN系统可以自动检测唾液涂抹中的AFB,从而提高诊断准确性和规律性.
- 这种人工智能驱动的方法可以支持显微镜,减少诊断错误,并改善结核病管理,特别是在资源有限的环境中.
- 建议进一步研究涉及更大的数据集和与其他诊断工具的集成,以获得现实世界的临床实用性.
更多相关视频
09:54Synthesis, Characterization, and Application of Superparamagnetic Iron Oxide Nanoprobes for Extrapulmonary Tuberculosis Detection
Published on: February 16, 2020
6.1K
06:52Optimized Workflow for Iterative Bleaching Extends Multiplexity Imaging of Highly Autofluorescent Clinical Samples
Published on: July 11, 2025
740
相关概念视频
01:26Sputum Studies I: Gram Stain, cytology, and Acid-fast smear and culture
536
Sputum studies are a critical part of diagnosing and treating numerous respiratory conditions. These studies involve obtaining sputum samples for analysis to identify pathogenic organisms and assess the presence of abnormal cells indicative of malignant conditions. This lesson will delve into three fundamental sputum studies: Gram Stain, Cytology, and Acid-fast Smear and Culture.
Gram Stain
The Gram Stain is an integral part of sputum studies. It involves the staining of sputum, which permits...
Gram Stain
The Gram Stain is an integral part of sputum studies. It involves the staining of sputum, which permits...
536
01:26Pulmonary Tuberculosis IV
468
Tuberculosis, more commonly referred to as TB, is an infectious disease stemming from Mycobacterium tuberculosis. While it primarily impacts the lungs, TB can also affect other body areas. Given its severity and global impact, timely and accurate diagnosis is crucial for controlling its spread and improving patient outcomes.
Several diagnostic approaches are used to detect TB. The conventional method is the Tuberculin Skin Test (TST), also known as the Mantoux test. However, this method has...
Several diagnostic approaches are used to detect TB. The conventional method is the Tuberculin Skin Test (TST), also known as the Mantoux test. However, this method has...
468
01:20Sputum Studies II: Culture and Sensitivity
1.1K
Description
Sputum culture and sensitivity is a medical procedure used to diagnose bacterial infections in the respiratory tract and select the most appropriate antibiotics for treatment. This process involves analyzing sputum samples of thick and opaque secretions produced in the lungs and airways. These samples are collected from patients and then sent to the laboratory for analysis.
The test can identify various pathogens responsible for respiratory infections, including Streptococcus,...
Sputum culture and sensitivity is a medical procedure used to diagnose bacterial infections in the respiratory tract and select the most appropriate antibiotics for treatment. This process involves analyzing sputum samples of thick and opaque secretions produced in the lungs and airways. These samples are collected from patients and then sent to the laboratory for analysis.
The test can identify various pathogens responsible for respiratory infections, including Streptococcus,...
1.1K