Purpose
Chest X-ray (CXR) is one of the most common diagnostic imaging tests to identify and monitor various chest findings, including pulmonary lesions that may be indicative of, among other pathologies, lung cancer. However, the effectiveness of X-ray imaging in detecting both primary and secondary tumors is not always reliable [1]. The aim of our study was to demonstrate the effectiveness of the proposed deep learning-based algorithm (DLAD) for detecting pulmonary lesions on CXR images and to compare its performance with that of radiologists with different...
Methods and materials
A pulmonary lesion can usually be observed as a rounded, well-defined opacity on CXR. The margins may be smooth or convoluted and it is surrounded by normally aerated lung tissue. A lesion is defined by its size, location, margins, and density. Generally, if the lesion is less than 30 mm in diameter, it is considered a nodule; if more than 30 mm, it is considered a mass [2,3]. Occasionally, nodules as small as 5-6 mm in size can be visible on CXR [4]. Despite this,...
Results
The proposed DLAD (Carebot AI CXR v2.00) has achieved significantly higher Se (0.910 (0.854-0.966)) than that of all assessed radiologists (RAD 1 0.290 (0.201-0.379), p < 0.001, RAD 2 0.450 (0.352-0.548), p < 0.001, RAD 3 0.670 (0.578-0.762), p < 0.001, RAD 4 0.810 (0.733-0.887), p = 0.025, RAD 5 0.700 (0.610-0.790), p < 0.001), and would therefore help to identify patients with pulmonary lesions which the radiologists evaluated as without the suspicious finding.
[Fig 2]
However, the DLAD achieved a lower Sp (0.775 (0.717-0.833))...
Conclusion
The proposed DLAD demonstrated improved detection performance compared to existing conventional imaging-based diagnostics, as it showed a significantly lower false-negative rate while also providing relatively high specificity.
Personal information and conflict of interest
D. Kvak:
CEO: Carebot s.r.o.
A. Chromcova:
Employee: Carebot s.r.o.
R. Hrubý:
Employee: Carebot s.r.o.
M. Biroš:
Employee: Carebot s.r.o.
M. Pajdaković:
Employee: Carebot s.r.o.
K. Kvaková:
Employee: Carebot s.r.o.
S. Strukov:
Nothing to disclose
P. Polášková:
Nothing to disclose
E. Janu:
Nothing to disclose
References
[1] Sone, S., Takashima, S., Li, F., Yang, Z., Honda, T., Maruyama, Y., Hasegawa, M., Yamanda, T., Kubo, K., Hanamura, K. & Others. Mass screening for lung cancer with mobile spiral computed tomography scanner. The Lancet. 351, 1242-1245 (1998)
[2] Hansell D.M., Bankier A.A., MacMahon H., McLoud T.C., Müller N.L., Remy J. Fleischner Society: glossary of terms for thoracic imaging. Radiology (2008)
[3] Herring, W. Learning radiology: recognizing the basics. Elsevier Health Sciences (2019)
[4] Gamboa, A., Ethun, C., Switchenko, J., Lipscomb, J., Poultsides, G.,...