Park Jongsoo, Hwang Eui Jin, Lee Jong Hyuk, Hong Wonju, Nam Ju Gang, Lim Woo Hyeon, Kim Jae Hyun, Goo Jin Mo, Park Chang Min
Department of Radiology, Seoul National University Hospital.
Department of Radiology, Yeungnam University Medical Center, Daegu.
J Thorac Imaging. 2023 May 1;38(3):145-153. doi: 10.1097/RTI.0000000000000691. Epub 2023 Feb 3.
To evaluate the accuracy of a deep learning-based computer-aided detection (CAD) system in identifying active pulmonary tuberculosis on chest radiographs (CRs) of patients with positive interferon-gamma release assay (IGRA) results in different scenarios of clinical implementation.
We collected the CRs of consecutive patients with positive IGRA results. Findings of active pulmonary tuberculosis on CRs were independently evaluated by the CAD and a thoracic radiologist, followed by interpretation using the CAD. Sensitivity and specificity were evaluated in different scenarios: (a) radiologists' interpretation, (b) radiologists' CAD-assisted interpretation, and (c) CAD-based prescreening (radiologists' interpretation for positive CAD results only). We conducted a reader test to compare the accuracy of the CAD with those of 5 radiologists.
Among 1780 patients (men, 53.8%; median age, 56 y), 44 (2.5%) were diagnosed with active pulmonary tuberculosis. The CAD-assisted interpretation exhibited a higher sensitivity (81.8% vs. 72.7%; P =0.046) but lower specificity than the radiologists' interpretation (84.1% vs. 85.7%; P <0.001). The CAD-based prescreening exhibited a higher specificity than the radiologists' interpretation (88.8% vs. 85.7%; P <0.001) at the same sensitivity, with a workload reduction of 85.2% (1780 to 263). In the reader test, the CAD exhibited a higher sensitivity than radiologists (72.7% vs. 59.5%; P =0.005) at the same specificity (88.0%), and CAD-assisted interpretation significantly improved the sensitivity of radiologists' interpretation (72.3%; P <0.001).
For identifying active pulmonary tuberculosis among patients with positive IGRA results, deep learning-based CAD can enhance the sensitivity of interpretation. CAD-based prescreening may reduce the radiologists' workload at an improved specificity.
评估基于深度学习的计算机辅助检测（CAD）系统在不同临床应用场景下，对干扰素-γ释放试验（IGRA）结果呈阳性的患者胸部X线片（CR）上活动性肺结核的识别准确性。
我们收集了连续的IGRA结果呈阳性患者的CR。CAD和一位胸科放射科医生分别独立评估CR上的活动性肺结核表现，随后使用CAD进行解读。在不同场景下评估敏感性和特异性：（a）放射科医生的解读，（b）放射科医生的CAD辅助解读，以及（c）基于CAD的预筛查（仅对CAD结果为阳性的进行放射科医生解读）。我们进行了一项阅片者测试，以比较CAD与5位放射科医生的准确性。
在1780例患者中（男性占53.8%；中位年龄56岁），44例（2.5%）被诊断为活动性肺结核。与放射科医生的解读相比，CAD辅助解读表现出更高的敏感性（81.8%对72.7%；P =0.046），但特异性较低（84.1%对85.7%；P <0.001）。基于CAD的预筛查在相同敏感性下表现出比放射科医生解读更高的特异性（88.8%对85.7%；P <0.001），工作量减少了85.2%（从1780例降至263例）。在阅片者测试中，在相同特异性（88.0%）下，CAD表现出比放射科医生更高的敏感性（72.7%对59.5%；P =0.005），并且CAD辅助解读显著提高了放射科医生解读的敏感性（72.3%；P <0.001）。
对于识别IGRA结果呈阳性患者中的活动性肺结核，基于深度学习的CAD可提高解读的敏感性。基于CAD的预筛查可能在提高特异性的同时减少放射科医生的工作量。
