基于深度学习的软件进行胸部 X 射线分析作为肺结核分诊试验：对培养确诊疾病的诊断准确性的前瞻性研究。
Chest x-ray analysis with deep learning-based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease.
机构信息
McGill International TB Centre, Research Institute of the McGill University Health Centre and McGill University, Montreal, QC, Canada; Respiratory Epidemiology and Clinical Research Unit, Centre for Outcomes Research and Evaluation, Research Institute of the McGill University Health Centre, Montreal, QC, Canada; Department of Medicine and Department of Epidemiology, McGill University, Montreal, Canada.
Interactive Research and Development Pakistan, Karachi, Pakistan.
出版信息
Lancet Digit Health. 2020 Nov;2(11):e573-e581. doi: 10.1016/S2589-7500(20)30221-1. Epub 2020 Oct 19.
BACKGROUND
Deep learning-based radiological image analysis could facilitate use of chest x-rays as triage tests for pulmonary tuberculosis in resource-limited settings. We sought to determine whether commercially available chest x-ray analysis software meet WHO recommendations for minimal sensitivity and specificity as pulmonary tuberculosis triage tests.
METHODS
We recruited symptomatic adults at the Indus Hospital, Karachi, Pakistan. We compared two software, qXR version 2.0 (qXRv2) and CAD4TB version 6.0 (CAD4TBv6), with a reference of mycobacterial culture of two sputa. We assessed qXRv2 using its manufacturer prespecified threshold score for chest x-ray classification as tuberculosis present versus not present. For CAD4TBv6, we used a data-derived threshold, because it does not have a prespecified one. We tested for non-inferiority to preset WHO recommendations (0·90 for sensitivity, 0·70 for specificity) using a non-inferiority limit of 0·05. We identified factors associated with accuracy by stratification and logistic regression.
FINDINGS
We included 2198 (92·7%) of 2370 enrolled participants. 2187 (99·5%) of 2198 were HIV-negative, and 272 (12·4%) had culture-confirmed pulmonary tuberculosis. For both software, accuracy was non-inferior to WHO-recommended minimum values (qXRv2 sensitivity 0·93 [95% CI 0·89-0·95], non-inferiority p=0·0002; CAD4TBv6 sensitivity 0·93 [0·90-0·96], p<0·0001; qXRv2 specificity 0·75 [0·73-0·77], p<0·0001; CAD4TBv6 specificity 0·69 [0·67-0·71], p=0·0003). Sensitivity was lower in smear-negative pulmonary tuberculosis for both software, and in women for CAD4TBv6. Specificity was lower in men and in those with previous tuberculosis, and reduced with increasing age and decreasing body mass index. Smoking and diabetes did not affect accuracy.
INTERPRETATION
In an HIV-negative population, these software met WHO-recommended minimal accuracy for pulmonary tuberculosis triage tests. Sensitivity will be lower when smear-negative pulmonary tuberculosis is more prevalent.
FUNDING
Canadian Institutes of Health Research.
背景
基于深度学习的放射影像学分析可促进在资源有限的环境中使用胸部 X 射线作为肺结核的分诊检测。我们旨在确定市售的胸部 X 射线分析软件是否符合世界卫生组织（WHO）对最小灵敏度和特异性的肺结核分诊检测要求。
方法
我们在巴基斯坦卡拉奇的 Indus 医院招募了有症状的成年人。我们将两种软件（qXR 版本 2.0（qXRv2）和 CAD4TB 版本 6.0（CAD4TBv6））与两种痰液的分枝杆菌培养物进行比较。我们使用制造商针对 qXRv2 预先指定的 X 射线分类阈值评分来评估 X 射线是否存在肺结核。对于 CAD4TBv6，我们使用数据推导的阈值，因为它没有预设的阈值。我们使用 0.05 的非劣效性限值，通过非劣效性检验来测试是否符合 WHO 的预设建议（灵敏度 0.90，特异性 0.70）。我们通过分层和逻辑回归来确定与准确性相关的因素。
结果
我们纳入了 2370 名入组参与者中的 2198 名（92.7%）。2187 名（99.5%）参与者 HIV 检测均为阴性，272 名（12.4%）参与者经培养证实患有肺结核。对于这两种软件，其准确性均不劣于 WHO 推荐的最低值（qXRv2 灵敏度 0.93 [95%CI 0.89-0.95]，非劣效性 p=0.0002；CAD4TBv6 灵敏度 0.93 [0.90-0.96]，p<0.0001；qXRv2 特异性 0.75 [0.73-0.77]，p<0.0001；CAD4TBv6 特异性 0.69 [0.67-0.71]，p=0.0003）。对于两种软件，在痰涂片阴性的肺结核患者中，以及在女性患者中，灵敏度均较低。在男性和既往有肺结核患者中，特异性较低，并且随着年龄的增加和体重指数的降低而降低。吸烟和糖尿病对准确性没有影响。
解释
在 HIV 阴性人群中，这些软件的肺结核分诊检测符合 WHO 推荐的最低准确性。当涂片阴性的肺结核更为普遍时，灵敏度会降低。
资助
加拿大卫生研究院。
Zotero 插件