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

Background: Computer-aided detection (CAD) algorithms for automated chest X-ray (CXR) reading have been endorsed by the World Health Organization for tuberculosis (TB) triage, but independent, multi-country assessment and comparison of current products are needed to guide implementation. Methods: We conducted a head-to-head evaluation of five CAD algorithms for TB triage across seven countries. We included CXRs from adults who presented to outpatient facilities with at least two weeks of cough in India, Madagascar, the Philippines, South Africa, Tanzania, Uganda, and Vietnam. The participants completed a standard evaluation for pulmonary TB, including sputum collection for Xpert MTB/RIF Ultra and culture. Against a microbiological reference standard, we calculated and compared the accuracy overall, by country and key groups for five CAD algorithms: CAD4TB (Delft Imaging), INSIGHT CXR (Lunit), DrAid (Vinbrain), Genki (Deeptek), and qXR (qure.AI). We determined the area under the ROC curve (AUC) and if any CAD product could achieve the minimum target accuracy for a TB triage test (≥90% sensitivity and ≥70% specificity). We then applied country- and population-specific thresholds and recalculated accuracy to assess any improvement in performance. Results: Of 3,927 individuals included, the median age was 41 years (IQR 29-54), 12.9% were people living with HIV (PLWH), 8.2% living with diabetes, and 21.2% had a prior history of TB. The overall AUC ranged from 0.774-0.819, and specificity ranged from 64.8-73.8% at 90% sensitivity. CAD4TB had the highest overall accuracy (73.8% specific, 95% CI 72.2-75.4, at 90% sensitivity), although qXR and INSIGHT CXR also achieved the target 70% specificity. There was heterogeneity in accuracy by country, and females and PLWH had lower sensitivity while males and people with a history of TB had lower specificity. The performance remained stable regardless of diabetes status. When country- and population-specific thresholds were applied, at least one CAD product could achieve or approach the target accuracy for each country and sub-group, except for PLWH and those with a history of TB. Conclusions: Multiple CAD algorithms can achieve or exceed the minimum target accuracy for a TB triage test, with improvement when using setting- or population-specific thresholds. Further efforts are needed to integrate CAD into routine TB case detection programs in high-burden communities.

Solicited Cough Sound Analysis for Tuberculosis Triage Testing: The CODA TB DREAM Challenge Dataset.

Cough is a common and commonly ignored symptom of lung disease. Cough is often perceived as difficult to quantify, frequently self-limiting, and non-specific. However, cough has a central role in the clinical detection of many lung diseases including tuberculosis (TB), which remains the leading infectious disease killer worldwide. TB screening currently relies on self-reported cough which fails to meet the World Health Organization (WHO) accuracy targets for a TB triage test. Artificial intelligence (AI) models based on cough sound have been developed for several respiratory conditions, with limited work being done in TB. To support the development of an accurate, point-of-care cough-based triage tool for TB, we have compiled a large multi-country database of cough sounds from individuals being evaluated for TB. The dataset includes more than 700,000 cough sounds from 2,143 individuals with detailed demographic, clinical and microbiologic diagnostic information. We aim to empower researchers in the development of cough sound analysis models to improve TB diagnosis, where innovative approaches are critically needed to end this long-standing pandemic.

Evaluation of the DN-Mini (miniaturized double net) trap for sampling host-seeking Anopheles mosquitoes in malaria-endemic villages of southern Tanzania.

BACKGROUND: Surveillance of malaria vectors is crucial for assessing the transmission risk and impact of control measures. Human landing catches (HLC) directly estimate the biting rates but raise ethical concerns due to the exposure of volunteers to mosquito-borne pathogens. A common alternative is the CDC-light trap, which is effective for catching host-seeking mosquitoes indoors but not outdoors. New, exposure-free methods are needed for sampling mosquitoes indoors and outdoors in ways that reflect their natural risk profiles. The aim of this study was therefore to evaluate the efficacy of the miniaturized double net trap (DN-Mini) for sampling host-seeking mosquitoes in south-eastern Tanzania, where malaria transmission is dominated by Anopheles funestus. METHODS: Adult mosquitoes were collected from 222 randomly selected houses across three villages (74 per village) in Ulanga district, south-eastern Tanzania, using the DN-Mini traps, CDC-Light traps, and Prokopack aspirators. First, we compared CDC-light and DN-Mini traps for collecting indoor host-seeking mosquitoes, while Prokopack aspirators were used for indoor-resting mosquitoes. Second, we deployed the DN-Mini and Prokopack aspirators to collect host-seeking and resting mosquitoes indoors and outdoors. Generalized linear mixed models (GLMM) with a negative binomial distribution were used to compare the effectiveness of the traps for catching different mosquito species. RESULTS: The DN-Mini was 1.53 times more efficient in collecting An. funestus indoors (RR = 1.53, 95% CI: 1.190-1.98) compared to the CDC-Light trap. However, for Anopheles arabiensis, the DN-Mini caught only 0.32 times as many mosquitoes indoors as the CDC-Light traps (RR = 0.32, 95% CI: 0.183-0.567). Both An. funestus and An. arabiensis were found to be more abundant indoors than outdoors when collected using the DN-Mini trap. Similarly, the Prokopack aspirator was greater indoors than outdoors for both An. funestus and An. arabiensis. CONCLUSION: The DN-Mini outperformed the CDC-light trap in sampling the dominant malaria vector, An. funestus species, but was less effective in capturing An. arabiensis, and for both vector species, the biting risk was greater indoors than outdoors when measured using the DN-Mini trap. These findings highlight the importance of selecting appropriate trapping methods based on mosquito species and behaviors.

BOHEMIA: Broad One Health Endectocide-based Malaria Intervention in Africa-a phase III cluster-randomized, open-label, clinical trial to study the safety and efficacy of ivermectin mass drug administration to reduce malaria transmission in two African settings.

BACKGROUND: Residual malaria transmission is the result of adaptive mosquito behavior that allows malaria vectors to thrive and sustain transmission in the presence of good access to bed nets or insecticide residual spraying. These behaviors include crepuscular and outdoor feeding as well as intermittent feeding upon livestock. Ivermectin is a broadly used antiparasitic drug that kills mosquitoes feeding on a treated subject for a dose-dependent period. Mass drug administration with ivermectin has been proposed as a complementary strategy to reduce malaria transmission. METHODS: A cluster randomized, parallel arm, superiority trial conducted in two settings with distinct eco-epidemiological conditions in East and Southern Africa. There will be three groups: human intervention, consisting of a dose of ivermectin (400 mcg/kg) administered monthly for 3 months to all the eligible population in the cluster (>15 kg, non-pregnant and no medical contraindication); human and livestock intervention, consisting human treatment as above plus treatment of livestock in the area with a single dose of injectable ivermectin (200 mcg/kg) monthly for 3 months; and controls, consisting of a dose of albendazole (400 mg) monthly for 3 months. The main outcome measure will be malaria incidence in a cohort of children under five living in the core of each cluster followed prospectively with monthly RDTs DISCUSSION: The second site for the implementation of this protocol has changed from Tanzania to Kenya. This summary presents the Mozambique-specific protocol while the updated master protocol and the adapted Kenya-specific protocol undergo national approval in Kenya. BOHEMIA will be the first large-scale trial evaluating the impact of ivermectin-only mass drug administration to humans or humans and cattle on local malaria transmission TRIAL REGISTRATION: ClinicalTrials.gov NCT04966702 . Registered on July 19, 2021. Pan African Clinical Trials Registry PACTR202106695877303.

Passive Outdoor Host Seeking Device (POHD): Designing and Evaluation against Outdoor Biting Malaria Vectors.

Odor-baited devices are increasingly needed to compliment long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) for control of residual malaria transmission. However, the odor-baited devices developed so far are bulky, dependent on the source of electricity and carbon dioxide (CO2), and they are logistically unsuitable for scaling up in surveillance and control of malaria vectors. We designed a passive and portable outdoor host seeking device (POHD) and preliminarily evaluated suitable components against Anopheles arabiensis that maintains residual malaria transmission. Experiments were conducted using semifield reared An. arabiensis within the semifield system at Ifakara Health Institute (IHI) in southeastern Tanzania. These mosquitoes were exposed to Suna traps® baited with BG lures or source of light and augmented with carbon dioxide (CO2) in view of identifying best attractants necessary to improve attractiveness of designed POHD. Two Suna traps® were hanged at the corner but outside the experimental hut in a diagonal line and rotated between four corners to control for the effect of position and wind direction on mosquito catches. Furthermore, mosquitoes were also exposed to either a bendiocarb-treated or bendiocarb-untreated POHD baited with Mbita blend, Ifakara blend, and worn socks and augmented with warmth (i.e., 1.5 liter bottle of warm water) inside an experimental hut or a screened rectangular box. This study demonstrated that mosquitoes were more strongly attracted to Suna trap® baited with BG lures and CO2 relative to those traps baited with a source of light and CO2. The POHD baited with synthetic blends attracted and killed greater proportion of An. arabiensis compared with POHD baited with worn socks. Efficacy of the POHD was unaffected by source of warmth, and it was reduced by about 50% when the device was tested inside a screened rectangular box relative to closed experimental hut. Overall, this study demonstrates that the POHD baited with synthetic blends (Mbita and Ifakara blends) and bendiocarb can effectively attract and kill outdoor biting malaria vector species. Such POHD baited with synthetic blends may require the source of CO2 to enhance attractiveness to mosquitoes. Further trials are, therefore, ongoing to evaluate attractiveness of improved design of POHD baited with slow-release formulation of synthetic blends and sustainable source of CO2 to malaria vectors under semifield and natural environments.

The Use of Granular Cyclopentanone as Alternative to Artificial Source of Carbon Dioxide in Improved Passive Outdoor Host Seeking Device (POHD).

Reliable sources of CO2 that are relatively cheap, obtainable, and easy to sustain are immediately required for scaling up of odor-baited mosquito surveillance and control devices. Several odor-baited devices are in the pipeline; however, their scale-up against residual malaria transmission, particularly in resource poor areas, is limited by the unavailability of reliable sources of CO2 and reliance on electrical power sources among other factors. We evaluated the use of granular cyclopentanone as an alternative to artificial or yeast fermentation-produced CO2 in passive outdoor host seeking device (POHD). Experiments were conducted against semifield reared An. arabiensis within the semifield system (SFS) at Ifakara Health Institute. Mosquitoes were tested against odor-baited POHDs augmented with yeast fermentation-produced CO2, granular cyclopentanone, attractive blends (Mbita or Ifakara), or their combinations. An insecticide, bendiocarb, was a killing agent used as a proxy for marking the mosquitoes visit the POHDs. Relative attractiveness of different treatment combinations was compared based on the proportion of dead mosquitoes that visited the POHD. The POHD augmented with granules of cyclopentanone alone was attractive to An. arabiensis as much as, or more than, POHDs augmented with yeast fermentation-produced CO2. The POHD baited with CO2 attracted more mosquitoes than those POHDs baited with synthetic blends alone; when these blends are combined with CO2, they attracted more mosquitoes than individual blends. More importantly, such POHDs baited with cyclopentanone attracted far greater proportion of mosquitoes than the POHD baited with either Mbita or Ifakara blend alone. The granular cyclopentanone strongly enhanced/potentiated the attractiveness of POHD baited with Mbita blends against mosquitoes compared to that of POHD baited with Ifakara blend. Moreover, the granular cyclopentanone retained its residual activity against An. arabiensis for up to 2 months after application particularly when used in combination with Mbita blend. In conclusion, this study demonstrates that cyclopentanone granules have the potential to substitute sources of CO2 in outdoor-based surveillance and control devices, thus warranting evaluation of such alternative under realistic field conditions.

Semifield Evaluation of Improved Passive Outdoor Host Seeking Device (POHD) for Outdoor Control of Anopheles arabiensis Mosquitoes.

Despite the considerable progress made so far, the effectiveness and mass application of odour-baited outdoor mosquito control devices in pipelines is limited by several factors. These include the design and size of the devices, optimal placement of attractive blends, and nature of materials into which the blends are impregnated. The primary aim of this study was to manipulate these factors to improve the attractiveness of our recently developed passive outdoor host seeking device (POHD) to outdoor biting Anopheles arabiensis. Specifically, the study aimed to determine optimal placement of odour blends and killing bioactives in POHD for maximum attraction and killing of An. arabiensis and to assess the effects of blend types, formulation, and residual activity on attractiveness of the POHD to An. arabiensis. The POHDs baited with attractive blends, carbon dioxide (CO2), and bendiocarb-treated electrostatic netting were placed either towards the top or bottom openings, and other modifications were exposed to An. arabiensis under the semifield system at Ifakara Health Institute (IHI). Each night, a total of 100 starved female, 3-7-day-old, semifield reared An. arabiensis mosquitoes were released, collected the next morning (alive or dead), counted, and recorded. Live mosquitoes were maintained in the semifield insectary and monitored for 24 hours mortality. Each treatment combination of the POHD was tested in three replicates. Overall, the results indicated that the proportion of mosquitoes attracted to and killed in the POHD varied with position of attractants and killing agent (bendiocarb). The POHD with bottom placed attractants and bendiocarb attracted and killed higher proportion of mosquitoes compared to the POHD with top placed attractants and bendiocarb. The highest mortalities were observed when the POHD was baited with a combination of attractive blends and CO2. Moreover, the residual activity of attractive blends applied inside POHD varied with type and formulation of attractive blend. The POHD packed with Mbita and Ifakara blend in microencapsulated pellets (granules) attracted higher proportion of mosquitoes than that baited with soaked nylon-strip formulation of either blends. Interestingly, POHD baited with Mbita blend in microencapsulated pellets (granules) formulation attracted and killed higher proportion of mosquitoes (>90%) than that baited with Ifakara blend even 9 months after application. Conclusively, the POHD remained effective for a relatively longer period of time when baited with bottom placed synthetic blends and CO2 combination, thus warranting further trials under real life situations.

Mesocosm experiments reveal the impact of mosquito control measures on malaria vector life history and population dynamics.

The impact of control measures on mosquito vector fitness and demography is usually estimated from bioassays or indirect variables in the field. Whilst indicative, neither approach is sufficient to quantify the potentially complex response of mosquito populations to combined interventions. Here, large replicated mesocosms were used to measure the population-level response of the malaria vector Anopheles arabiensis to long-lasting insecticidal nets (LLINs) when used in isolation, or combined with insecticidal eave louvers (EL), or treatment of cattle with the endectocide Ivermectin (IM). State-space models (SSM) were fit to these experimental data, revealing that LLIN introduction reduced adult mosquito survival by 91% but allowed population persistence. ELs provided no additional benefit, but IM reduced mosquito fecundity by 59% and nearly eliminated all populations when combined with LLINs. This highlights the value of IM for integrated vector control, and mesocosm population experiments combined with SSM for identifying optimal combinations for vector population elimination.

Ivermectin-treated cattle reduces blood digestion, egg production and survival of a free-living population of Anopheles arabiensis under semi-field condition in south-eastern Tanzania.

BACKGROUND: Anopheles arabiensis feed on cattle and contributes to residual transmission of malaria in areas with high coverage of long-lasting insecticide-treated nets and indoor residual spraying in East Africa. This study aimed to evaluate the effects of ivermectin-treated cattle as a complementary vector control tool against population of An. arabiensis under the semi-field conditions in south-eastern Tanzania. METHODS: The free-living population of An. arabiensis was allowed to forage on untreated or ivermectin-treated cattle in alternating nights within the semi-field system in south-eastern Tanzania. Fresh blood fed mosquitoes were collected in the morning using mouth aspirators and assessed for their blood meal digestion, egg production, and survivorship. The residual activity of ivermectin-treated cattle was also determined by exposing mosquitoes to the same treatments after every 2 days until day 21 post-treatments. These experiments were replicated 3 times using different individual cattle. RESULTS: Overall, the ivermectin-treated cattle reduced blood meal digestion in the stomach of An. arabiensis, and their subsequent egg production and survival over time. The ivermectin-treated cattle halved blood meal digestion in mosquitoes, but reduced their egg production for up to 15 days. The ivermectin-treated cattle reduced the survival, and median survival times (1-3 days) of An. arabiensis than control cattle. The daily mortality rates of mosquitoes fed on ivermectin-treated cattle increased by five-fold relative to controls in the first week, and it gradually declined up to 21 days after treatment. CONCLUSION: This study demonstrates that long-lasting effects of ivermectin-treated cattle on egg production and survival of An. arabiensis may sustainably suppress their vector density, and reduce residual transmission of malaria. This study suggests that ivermectin-treated non-lactating cattle (i.e. calves, heifers and bulls) could be suitable option for large-scale malaria vector control without limiting consumption of milk and meat by communities in rural settings. Furthermore, simulation models are underway to predict the impact of ivermectin-treated cattle alone, or in combination with LLIN/IRS, the frequency of treatment, and their coverage required to significantly suppress population of An. arabiensis and reduce residual transmission of malaria.

Eave tubes for malaria control in Africa: initial development and semi-field evaluations in Tanzania.

BACKGROUND: Presented here are a series of preliminary experiments evaluating "eave tubes"-a technology that combines house screening with a novel method of delivering insecticides for control of malaria mosquitoes. METHODS: Eave tubes were first evaluated with overnight release and recapture of mosquitoes in a screened compartment containing a hut and human sleeper. Recapture numbers were used as a proxy for overnight survival. These trials tested physical characteristics of the eave tubes (height, diameter, angle), and different active ingredients (bendiocarb, LLIN material, fungus). Eave tubes in a hut with closed eaves were also compared to an LLIN protecting a sleeper in a hut with open eaves. Eave tubes were then evaluated in a larger compartment containing a self-replicating mosquito population, vegetation, and multiple houses and cattle sheds. In this "model village", LLINs were introduced first, followed by eave tubes and associated house modifications. RESULTS: Initial testing suggested that tubes placed horizontally and at eave height had the biggest impact on mosquito recapture relative to respective controls. Comparison of active ingredients suggested roughly equivalent effects from bendiocarb, LLIN material, and fungal spores (although speed of kill was slower for fungus). The impact of treated netting on recapture rates ranged from 50 to 70 % reduction relative to controls. In subsequent experiments comparing bendiocarb-treated netting in eave tubes against a standard LLIN, the effect size was smaller but the eave tubes with closed eaves performed at least as well as the LLIN with open eaves. In the model village, introducing LLINs led to an approximate 60 % reduction in larval densities and 85 % reduction in indoor catches of host-seeking mosquitoes relative to pre-intervention values. Installing eave tubes and screening further reduced larval density (93 % relative to pre intervention values) and virtually eliminated indoor host-seeking mosquitoes. When the eave tubes and screening were removed, larval and adult catches recovered to pre-eave tube levels. CONCLUSIONS: These trials suggest that the "eave tube" package can impact overnight survival of host-seeking mosquitoes and can suppress mosquito populations, even in a complex environment. Further testing is now required to evaluate the robustness of these findings and demonstrate impact under field conditions.