Background Artificial intelligence (AI) algorithms can be trained to recognise tuberculosis-related abnormalities on chest radiographs. Various AI algorithms are available commercially, yet there is little impartial evidence on how their performance compares with each other and with radiologists. We aimed to evaluate five commercial AI algorithms for triaging tuberculosis using a large dataset that had not previously been used to train any AI algorithms.Methods Individuals aged 15 years or older presenting or referred to three tuberculosis screening centres in Dhaka, Bangladesh, between May 15, 2014, and Oct 4, 2016, were recruited consecutively. Every participant was verbally screened for symptoms and received a digital posterior-anterior chest x-ray and an Xpert MTB/RIF (Xpert) test. All chest x-rays were read independently by a group of three registered radiologists and five commercial AI algorithms: CAD4TB (version 7), InferRead DR (version 2), Lunit INSIGHT CXR (version 4.9.0), JF CXR-1 (version 2), and qXR (version 3). We compared the performance of the AI algorithms with each other, with the radiologists, and with the WHO's Target Product Profile (TPP) of triage tests (≥90% sensitivity and ≥70% specificity). We used a new evaluation framework that simultaneously evaluates sensitivity, proportion of Xpert tests avoided, and number needed to test to inform implementers' choice of software and selection of threshold abnormality scores. Findings Chest x-rays from 23 954 individuals were included in the analysis. All five AI algorithms significantly outperformed the radiologists. The areas under the receiver operating characteristic curve were 90•81% (95% CI 90•33-91•29) for qXR, 90•34% (89•81-90•87) for CAD4TB, 88•61% (88•03-89•20) for Lunit INSIGHT CXR, 84•90% (84•27-85•54) for InferRead DR, and 84•89% (84•26-85•53) for JF CXR-1. Only qXR (74•3% specificity [95% CI 73•3-74•9]) and CAD4TB (72•9% specificity [72•3-73•5]) met the TPP at 90% sensitivity. All five AI algorithms reduced the number of Xpert tests required by 50% while maintaining a sensitivity above 90%. All AI algorithms performed worse among older age groups (>60 years) and people with a history of tuberculosis.Interpretation AI algorithms can be highly accurate and useful triage tools for tuberculosis detection in high-burden regions, and outperform human readers.Funding Government of Canada.

Whether an individual becomes infected in an infectious disease outbreak depends on many interconnected risk factors, which may relate to characteristics of the individual (e.g., age, sex), his or her close relatives (e.g., household members), or the wider community. Studies monitoring individuals in households or schools have helped elucidate the determinants of transmission in small social structures due to advances in statistical modeling; but such an approach has so far largely failed to consider individuals in the wider context they live in. Here, we used an outbreak of chikungunya in a rural community in Bangladesh as a case study to obtain a more comprehensive characterization of risk factors in disease spread. We developed Bayesian data augmentation approaches to account for uncertainty in the source of infection, recall uncertainty, and unobserved infection dates. We found that the probability of chikungunya transmission was 12% [95% credible interval (CI): 8-17%] between household members but dropped to 0.3% for those living 50 m away (95% CI: 0.2-0.5%). Overall, the mean transmission distance was 95 m (95% CI: 77-113 m). Females were 1.5 times more likely to become infected than males (95% CI: 1.2-1.8), which was virtually identical to the relative risk of being at home estimated from an independent human movement study in the country. Reported daily use of antimosquito coils had no detectable impact on transmission. This study shows how the complex interplay between the characteristics of an individual and his or her close and wider environment contributes to the shaping of infectious disease epidemics. data augmentation | Bayesian | chikungunya | outbreaks | spatial spread

Serostudies are needed to answer generalizable questions on disease risk. However, recruitment is usually biased by age or location. We present a nationally-representative study for dengue from 70 communities in Bangladesh. We collected data on risk factors, trapped mosquitoes and tested serum for IgG. Out of 5866 individuals, 24% had evidence of historic infection, ranging from 3% in the north to >80% in Dhaka. Being male (aOR:1.8, [95%CI:1.5–2.0]) and recent travel (aOR:1.3, [1.1–1.8]) were linked to seropositivity. We estimate that 40 million [34.3–47.2] people have been infected nationally, with 2.4 million ([1.3–4.5]) annual infections. Had we visited only 20 communities, seropositivity estimates would have ranged from 13% to 37%, highlighting the lack of representativeness generated by small numbers of communities. Our findings have implications for both the design of serosurveys and tackling dengue in Bangladesh.

BackgroundDengue virus (DENV) activity has been reported in Dhaka, Bangladesh since the early 1960s with the greatest burden of dengue fever and dengue hemorrhagic fever cases observed in 2000. Since this time, the intensity of dengue activity has varied from year to year, and its determining factors remained relatively unknown. In light of such gaps in knowledge, the main objectives of this study were to determine the magnitude of seroprevalence and seroconversion among the surveyed population, and establish the individual/household level risk factors for the presence of DENV antibodies among all age groups of target populations in the city of Dhaka.Methodology/Principal findingsConsidering the lack of fine scale investigations on the factors driving dengue activity in Bangladesh, a prospective cohort study involving serological surveys was undertaken with participant interviews and blood donation across the city of Dhaka in 2012. Study participants were recruited from 12 of 90 wards and blood samples were collected during both the pre-monsoon (n = 1125) and post-monsoon (n = 600) seasons of 2012. The findings revealed that the seroprevalence in all pre-monsoon samples was 80.0% (900/1125) while the seropositivity in the pre-monsoon samples that had paired post-monsoon samples was 83.3% (503/600). Of the 97 paired samples that were negative at the pre-monsoon time point, 56 were positive at the post-monsoon time point. This resulted in a seroprevalence of 93.2% (559/600) among individuals tested during the post-monsoon period. Seroprevalence trended higher with age with children exhibiting a lower seropositivity as compared to adults. Results from this study also indicated that DENV strains were the only flaviviruses circulating in Dhaka in 2012. A multivariate analysis revealed that age, possession of indoor potted plants, and types of mosquito control measures were significant factors associated with DENV seroprevalence; while attendance in public/mass gatherings, and use of mosquito control measures were significantly associated with DENV seroconversion after adjusting for all other variables.Conclusions/SignificanceOur study suggests that there is a high level of endemic dengue virus circulation in the city of Dhaka which has resulted in significant DENV seroprevalence among its residents. Seropositivity increased with age, however, a substantial proportion of children are at risk for DENV infections. Our serological analysis also documents considerable DENV seroconversion among study participants which indicates that a large proportion of the population in the city of Dhaka were newly exposed to DENV during the study period (pre-and post-monsoon 2012). High levels of seroconversion suggest that there was an intense circulation of DENV in 2012 and this may have resulted in a significant risk for viral associated illness. Findings of our study further indicated that home-based interventions, such as removing indoor potted plants and increased bed net use, in addition to vector control measures in public parks...

Dengue viruses are responsible for over 100 million infections a year worldwide and are a public health concern in Bangladesh. Although risk of transmission is high, data on vector population characteristics are scanty in Bangladesh; therefore, a comprehensive prediction of the patterns of local virus transmission is not possible. Recognizing these gaps, multi-year entomological surveys were carried out in Dhaka, where the disease is most frequently reported. The specific objectives of the present study are threefold: i) to determine the risk factors for the presence of Aedes mosquitoes; ii) to identify the types of most productive and key containers; and iii) to estimate the effects of climatic factors on Aedes abundance in the city of Dhaka, Bangladesh. Entomological surveys were conducted in 12 out of 90 wards in Dhaka. These wards were selected using a probability proportional sampling procedure during the monsoon seasons in 2011, 2012 and 2013 and in the dry season in 2012. All containers inside and around sampled households were inspected for mosquito larvae and pupae, and containers were classified according to their relative size, use pattern, and materials of construction. During the study period (2011–2013), 12,680 larvae and pupae were collected. About 82% of the identified immature mosquitoes were Aedes aegypti, while the remainder were Ae. albopictus and other mosquito species. The largest number of immature mosquitoes was collected from tires and refrigerator trays during 2011 and 2012 monsoon seasons. Conversely, plastic drums were the most productive during the 2012 dry and 2013 monsoon season. Vehicle parts and discarded construction materials were the most efficient producers of Aedes mosquitoes in all surveys. The presence of Aedes mosquitoes was significantly (p < 0.05) higher in low socio-economic zones of Dhaka. Container location, presence of vegetation, and availability of shade for containers were also significantly associated with finding immature Aedes mosquitoes, based on multivariable analysis after confounder adjustment. Rainfall, temperature, and relative humidity also significantly affected the mean abundance of mosquitoes. Proper use, disposal, and recycling of the containers that effectively produce large numbers of Aedes vector mosquitoes may decrease the risk of arboviral transmission.