Diagnostic accuracy of three computer-aided detection systems for detecting pulmonary tuberculosis on chest radiography when used for screening: Analysis of an international, multicenter migrants screening study

Abstract

The aim of this study was to independently evaluate the diagnostic accuracy of three artificial intelligence (AI)-based computer aided detection (CAD) systems for detecting pulmonary tuberculosis (TB) on global migrants screening chest x-ray (CXR) cases when compared against both microbiological and radiological reference standards (MRS and RadRS, respectively). Retrospective clinical data and CXR images were collected from the International Organization for Migration (IOM) pre-migration health assessment TB screening global database for US-bound migrants. A total of 2,812 participants were included in the dataset used for analysis against RadRS, of which 1,769 (62.9%) had accompanying microbiological test results and were included against MRS. All CXRs were interpreted by three CAD systems (CAD4TB v6, Lunit INSIGHT v4.9.0, and qXR v2) in offline setting, and re-interpreted by two expert radiologists in a blinded fashion. The performance was evaluated using receiver operating characteristics curve (ROC), estimates of sensitivity and specificity at different CAD thresholds against both microbiological and radiological reference standards (MRS and RadRS, respectively), and was compared with that of the expert radiologists. The area under the curve against MRS was highest for Lunit (0.85; 95% CI 0.83−0.87), followed by qXR (0.75; 95% CI 0.72−0.77) and then CAD4TB (0.71; 95% CI 0.68−0.73). At a set specificity of 70%, Lunit had the highest sensitivity (81.4%; 95% CI 77.9–84.6); at a set sensitivity of 90%, specificity was also highest for Lunit (54.5%; 95% CI 51.7–57.3). The CAD systems performed comparable to the sensitivity (98.3%), and except CAD4TB, to specificity (13.7%) of the expert radiologists. Similar trends were observed when using RadRS. Area under the curve against RadRS was highest for CAD4TB (0.87; 95% CI 0.86–0.89) and Lunit (0.87; 95% CI 0.85–0.88) followed by qXR (0.81; 95% CI 0.80–0.83). At a set specificity of 70%, CAD4TB had highest sensitivity (84.1%; 95% CI 82.3−85.8) followed by Lunit (80.9%; 95% CI 78.9−82.7); and at a set sensitivity of 90%, specificity was also highest for CAD4TB (54.6%; 95% CI 51.3−57.8). In conclusion, the study demonstrated that the three CAD systems had broadly similar diagnostic accuracy with regard to TB screening and comparable accuracy to an expert radiologist against MRS. Compared with different reference standards, Lunit performed better than both qXR and CAD4TB against MRS, and CAD4TB and Lunit better than qXR against RadRS. Moreover, the performance of the CADs can be impacted by characteristics of subgroup of population. The main limitation was that our study relied on retrospective data and MRS was not routinely done in individuals with a low suspicion of TB and a normal CXR. Our findings suggest that CAD systems could be a useful tool for TB screening programs in remote, high TB prevalent places where access to expert radiologists may be limited. However, further large-scale prospective studies are needed to address outstanding questions around the operational performance and technical requirements of the CAD systems.

An independent, multi-country head-to-head accuracy comparison of automated chest x-ray algorithms for the triage of pulmonary tuberculosis

“…One study compared Genki to other CAD algorithms and noted similar specificity to CAD4TB and qXR, while we found it to be overall less specific. 11 However, that study assessed CAD in a screening cohort and was conducted in Vietnam where we also found Genki had higher specificity, highlighting the importance to conduct a multi-country evaluation to assess performance. To our knowledge this is the first published work to assess and compare DrAid, and although lower accuracy than the above three algorithms, overall it performed well with 68% specificity at 90% sensitivity.…”

“…Multiple analyses have found that the CAD threshold to classify TB may need to be adjusted for different settings and populations, but head-to-head comparisons of CAD products with these thresholds have been limited to one or two countries, 8,9 retrospective meta-analyses, 10 or for the screening use-case. 11,12 An independent, head-to-head comparison of the diagnostic accuracy of CAD algorithms in a large, diverse, multi-country cohort of individuals with presumptive pulmonary TB is needed to address these issues. We thus conducted a prospective diagnostic accuracy study across seven countries in sub-Saharan Africa, South Asia, and Southeast Asia.…”

A Comparison of CXR-CAD Software to Radiologists in Identifying COVID-19 in Individuals Evaluated for Sars CoV 2 Infection in Malawi and Zambia

“…This analysis mirrors a similar approach that has been described in other studies evaluating CXR-CAD for TB. (16, 21) Subgroup analysis was performed based on country, site, age group, sex, symptoms, diabetes status (if known), and HIV status (if known). Results were presented for CXR-CAD sensitivity and specificity with 95% confidence intervals.…”

“…Prior to sharing of DICOM images with FIND, identifying information was removed by use of a previously described DICOM anonymizing tool. (16) Patient clinical data and radiologist interpretations were aggregated and duplicate study subjects were excluded prior to analysis.…”

Head-to-head comparison of diagnostic accuracy of TB screening tests: Chest-X-ray, Xpert TB host response, and C-reactive protein

“…CXR implementation for TB screening has been limited in resource-constrained settings due to centralized CXR infrastructure [ 3 ] and scarcity of skilled personnel for interpretation [ 4 ]. The advent of digital CXR with computer-aided detection (CAD) analysis may reduce barriers to CXR screening by quickly enabling screening without the need for a skilled reader [ 5 , 6 ]. However, performance varies by population screened, and limited data are available for children [ 7 – 10 ].…”