BackgroundChest radiography (chest X-ray or CXR) plays an important role in the early detection of active pulmonary tuberculosis (TB). In areas with a high TB burden that require urgent screening, there is often a shortage of radiologists available to interpret the X-ray results. Computer-aided detection (CAD) software employed with artificial intelligence (AI) systems may have the potential to solve this problem.ObjectiveWe validated the effectiveness and safety of pulmonary tuberculosis imaging screening software that is based on a convolutional neural network algorithm.MethodsWe conducted prospective multicenter clinical research to validate the performance of pulmonary tuberculosis imaging screening software (JF CXR-1). Volunteers under the age of 15 years, both with or without suspicion of pulmonary tuberculosis, were recruited for CXR photography. The software reported a probability score of TB for each participant. The results were compared with those reported by radiologists. We measured sensitivity, specificity, consistency rate, and the area under the receiver operating characteristic curves (AUC) for the diagnosis of tuberculosis. Besides, adverse events (AE) and severe adverse events (SAE) were also evaluated.ResultsThe clinical research was conducted in six general infectious disease hospitals across China. A total of 1,165 participants were enrolled, and 1,161 were enrolled in the full analysis set (FAS). Men accounted for 60.0% (697/1,161). Compared to the results from radiologists on the board, the software showed a sensitivity of 94.2% (95% CI: 92.0–95.8%) and a specificity of 91.2% (95% CI: 88.5–93.2%). The consistency rate was 92.7% (91.1–94.1%), with a Kappa value of 0.854 (P = 0.000). The AUC was 0.98. In the safety set (SS), which consisted of 1,161 participants, 0.3% (3/1,161) had AEs that were not related to the software, and no severe AEs were observed.ConclusionThe software for tuberculosis screening based on a convolutional neural network algorithm is effective and safe. It is a potential candidate for solving tuberculosis screening problems in areas lacking radiologists with a high TB burden.

The spatiotemporal characteristics of dry-wet trends were identified and assessed, and the dominant meteorological factors were identified for the climate of Jiangsu province in humid southeastern China for the period 1960–2019. We conducted the research using data for the entire Jiangsu province as well as three major regions in Jiangsu (Huaibei, Jianghuai, and Sunan) with different regional climates. The results showed that decreased precipitation and relative humidity in spring and autumn over the study period were mainly responsible for the dry trends of the climates of all three regions and the entire province. Precipitation had a greater influence in spring and relative humidity in autumn. Decreases in sunshine hours and wind speed were responsible for the summer wet trends of the climates of Huaibei and Jianghuai and the entire province. However, precipitation increased significantly in the summer and was responsible for the increasing wet trend in Sunan. Significantly increased precipitation in winter was primarily responsible for the increasing wetness in Jianghuai and Sunan and the entire province in that season. However, the wet trend in northern Huaibei in winter was mainly caused by the decrease in wind speed over the study period. For the growing season and annually, the positive effects of changes in wind speed, sunshine hours, and precipitation led to increased humidity index in Jianghuai, Sunan, and the entire province. Precipitation showed a decreasing trend that countered the positive effects of decreases in wind speed and sunshine hours, which resulted in a slight decrease in the humidity index in Huaibei for both the growing season and annually. Sensitivity analysis indicated that the humidity index was positively sensitive to precipitation and relative humidity and negatively sensitive to air temperature, wind speed, and sunshine hours in Jiangsu province during 1960–2019. Overall, the humidity index in this region of southeastern China was most sensitive to changes in precipitation followed, in order of sensitivity, by sunshine hours, air temperature, wind speed, and relative humidity. Our findings provide a theoretical basis for adjusting irrigation programs and efficient utilization of water resources at the regional scale in humid southeastern China.

BackgroundThe recombinant mycobacterium tuberculosis fusion protein ESAT6-CFP10 skin test (ECST) is a novel test for tuberculosis (TB) infection; however, its accuracy in active tuberculosis (ATB) remains uncertain. This study aimed to evaluate the accuracy of ECST in the differential diagnosis of ATB for an early real-world assessment.MethodsThis prospective cohort study recruited patients suspected of ATB in Shanghai Public Health Clinical Center from January 2021 to November 2021. The diagnostic accuracy of the ECST was evaluated under the gold standard and composite clinical reference standard (CCRS) separately. The sensitivity, specificity, and corresponding confidence interval of ECST results were calculated, and subgroup analyses were conducted.ResultsDiagnostic accuracy was analyzed using data from 357 patients. Based on the gold standard, the sensitivity and specificity of the ECST for patients were 72.69% (95%CI 66.8%-78.5%) and 46.15% (95%CI 37.5%-54.8%), respectively. Based on the CCRS, the sensitivity and specificity of the ECST for patients were 71.52% (95%CI 66.4%-76.6%) and 65.45% (95%CI 52.5%-78.4%), respectively. The consistency between the ECST and the interferon-γ release (IGRA) test is moderate (Kappa = 0.47).ConclusionThe ECST is a suboptimum tool for the differential diagnosis of active tuberculosis. Its performance is similar to IGRA, an adjunctive diagnostic test for diagnosing active tuberculosis.Clinical trial registrationhttp://www.chictr.org.cn, identifier ChiCTR2000036369.