论文部分内容阅读
中国有世界上第二大的结核病疫情(仅次于印度),具有较高的结核病感染、发病、耐药以及病死率。因此,快速、准确的诊断,以及时采取治疗,是控制这种疾病的关键问题。涂片荧光显微镜与金胺-O染色是在病理实验室检测肺结核抗酸杆菌(acid fast bacilli,AFB)最常用的诊断方法。但是以眼睛观察显微镜来筛查检测抗酸杆菌是一项烦琐、劳动密集型任务。低质量、不一致的痰涂片染色技术,标本碎屑,人眼视觉的变异和疲劳等因素会导致灵敏度低至40%。使用基于人工智能(artificial intelligence,AI)计算机辅助诊断(computer aided diagnostic,CAD)技术的自动显微镜系统对结核病进行自动诊断,提供了一个有效解决当前痰涂片诊断结核病所存在的缺陷。胸部X射线片也是世界卫生组织认定的非常有效的快速分流和转诊检测方法。但是,在不发达地区,因放射医生缺乏,它无法为大量的感染人群提供服务。为解决这一需要,将先进的数字化医学影像精准诊断应用于肺结核(pulmonary tuberculosis,PTB)检测,基于人工智能的CAD自动化智能系统,为肺结核的数字化医学影像精准诊断开辟了一条新路。
China has the second largest tuberculosis epidemic in the world (second only to India), with a high incidence of tuberculosis, morbidity, resistance and mortality. Therefore, rapid and accurate diagnosis, and timely treatment is the key issue to control this disease. Smear fluorescence microscopy with Auramine-O staining is the most commonly used diagnostic method for the detection of acid fast bacilli (AFB) in pathology laboratories. However, screening microscopy for the detection of acid-fast bacilli is a cumbersome and labor-intensive task. Low-quality, inconsistent sputum smear staining techniques, specimen debris, variability in human vision, and fatigue can cause as little as 40% sensitivity. Automated diagnosis of tuberculosis using an automated microscopy system based on artificial intelligence (AI) computer aided diagnostic (CAD) technology provides an effective solution to the shortcomings of current sputum smears in the diagnosis of tuberculosis. Chest X-ray films are also very effective rapid triage and referral testing methods identified by the World Health Organization. However, in underdeveloped areas, radiopharmaceuticals lack capacity to serve a large number of infected people. In order to solve this problem, the advanced digital medical image accurate diagnosis is applied to the detection of pulmonary tuberculosis (PTB) and the artificial intelligence-based CAD automation intelligent system, which opens up a new path for the accurate diagnosis of digital medical images of pulmonary tuberculosis.