Identifying pulmonary nodules or masses on chest radiography using deep learning: external validation and strategies to improve clinical practice

• Published in: Clinical radiology Vol. 75; no. 1; pp. 38 - 45
• Main Authors: Liang, C.-H., Liu, Y.-C., Wu, M.-T., Garcia-Castro, F., Alberich-Bayarri, A., Wu, F.-Z.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.01.2020
• ISSN: 0009-9260; 1365-229X; 1365-229X
• DOI: 10.1016/j.crad.2019.08.005

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 (AUCMass: 0.916 versus AUCHeat map: 0.682, p<0.001; AUCMass: 0.916 versus AUCAbnormal: 0.810, p=0.002; AUCMass: 0.916 versus AUCNodule: 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. •Among the four different algorithms, mass algorithm has the best diagnostic accuracy with AUC of 0.916.•Rapid imaging processing time of 94.07 ± 16.54 seconds per case could help make workflow more efficient.•Chest X-ray classifier could be used to automatically evaluate all chest radiographs efficiently.•Implementation of these algorithms could help radiologists improve medical imaging care.