Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taipei, Taiwan; Faculty of Medicine, School of Medicine, National Yang Ming University, Taipei, Taiwan; Institute of Clinical Medicine, National Yang Ming University, Taipei, Taiwan.
Department of Diagnostic Radiology, Xiamen Chang Gung Hospital, China.
Clin Radiol. 2020 Jan;75(1):38-45. doi: 10.1016/j.crad.2019.08.005. Epub 2019 Sep 11.
To test the diagnostic performance of a deep learning-based system for the detection of clinically significant pulmonary nodules/masses on chest radiographs.
Using a retrospective study of 100 patients (47 with clinically significant pulmonary nodules/masses and 53 control subjects without pulmonary nodules), two radiologists verified clinically significantly pulmonary nodules/masses according to chest computed tomography (CT) findings. A computer-aided diagnosis (CAD) software using a deep-learning approach was used to detect pulmonary nodules/masses to determine the diagnostic performance in four algorithms (heat map, abnormal probability, nodule probability, and mass probability).
A total of 100 cases were included in the analysis. Among the four algorithms, mass algorithm could achieve a 76.6% sensitivity (36/47, 11 false negative) and 88.68% specificity (47/53, six false-positive) in the detection of pulmonary nodules/masses at the optimal probability score cut-off of 0.2884. Compared to the other three algorithms, mass probability algorithm had best predictive ability for pulmonary nodule/mass detection at the optimal probability score cut-off of 0.2884 (AUC: 0.916 versus AUC: 0.682, p<0.001; AUC: 0.916 versus AUC: 0.810, p=0.002; AUC: 0.916 versus AUC: 0.813, p=0.014).
In conclusion, the deep-learning based computer-aided diagnosis system will likely play a vital role in the early detection and diagnosis of pulmonary nodules/masses on chest radiographs. In future applications, these algorithms could support triage workflow via double reading to improve sensitivity and specificity during the diagnostic process.
测试基于深度学习的系统在胸部 X 光片上检测临床显著肺结节/肿块的诊断性能。
使用 100 例患者（47 例有临床显著肺结节/肿块，53 例对照无肺结节）的回顾性研究，两名放射科医生根据胸部 CT（CT）结果验证临床显著肺结节/肿块。使用基于深度学习的计算机辅助诊断（CAD）软件来检测肺结节/肿块，以确定四种算法（热图、异常概率、结节概率和肿块概率）中的诊断性能。
共有 100 例病例纳入分析。在四种算法中，肿块算法在最佳概率评分截止值为 0.2884 时，对肺结节/肿块的检测可达到 76.6%的灵敏度（36/47，11 个假阴性）和 88.68%的特异性（47/53，6 个假阳性）。与其他三种算法相比，肿块概率算法在最佳概率评分截止值为 0.2884 时对肺结节/肿块的检测具有最佳的预测能力（AUC：0.916 与 AUC：0.682，p<0.001；AUC：0.916 与 AUC：0.810，p=0.002；AUC：0.916 与 AUC：0.813，p=0.014）。
总之，基于深度学习的计算机辅助诊断系统可能在胸部 X 光片上肺结节/肿块的早期检测和诊断中发挥重要作用。在未来的应用中，这些算法可以通过双重阅读来支持分诊工作流程，以提高诊断过程中的敏感性和特异性。
Zotero 插件