基于深度学习的肺结核检测综述

Review on tuberculosis detection using deep learning

基于医学影像的肺结核病灶自动检测技术成为医学图像处理领域的研究热点。本研究围绕深度学习在肺结核病灶检测方面的相关研究与应用展开综述,首先阐述用于肺结核检测的实验基准,涵盖肺部医学影像的相关公开数据库和肺结核检测与分类竞赛的相关研究进展,然后提出肺结核检测领域中深度学习方法与应用的发展趋势,最后分析深度学习在肺结核诊断中面临的挑战。本研究从技术特性、性能优势、应用前景等方面对这些技术的研究进展以及面临的挑战进行总结和展望。

The automatic detection of tuberculosis lesions based on medical imaging has become a research hotspot in medical image processing.A comprehensive review of relevant researches and applications pertaining to deep learning in tuberculosis lesion detection is provided,which elucidates the experimental benchmarks in tuberculosis analysis,covering public datasets of pulmonary medical images and recent research advancements in tuberculosis detection and classification competitions,introduces emerging trends in deep learning methods and applications in tuberculosis detection,and analyzes the challenges existing in tuberculosis diagnosis using deep learning.The review summarizes and provides insights into the research advances and challenges of these technologies from the aspects of technical characteristics,performance advantages,and application prospects.