Data-driven and interpretable machine-learning modeling to explore the fine-scale environmental determinants of malaria vectors biting rates in rural Burkina Faso

Risk assessment of imported malaria in China: a machine learning perspective

BACKGROUND

Following China's official designation as malaria-free country by WHO, the imported malaria has emerged as a significant determinant impacting the malaria reestablishment within China. The objective of this study is to explore the application prospects of machine learning algorithms in imported malaria risk assessment of China.

METHODS

The data of imported malaria cases in China from 2011 to 2019 was provided by China CDC; historical epidemic data of malaria endemic country was obtained from World Malaria Report, and the other data used in this study are open access data. All the data processing and model construction based on R, and map visualization used ArcGIS software.

RESULTS

A total of 27,088 malaria cases imported into China from 85 countries between 2011 and 2019. After data preprocessing and classification, clean dataset has 765 rows (85 * 9) and 11 cols. Six machine learning models was constructed based on the training set, and Random Forest model demonstrated the best performance in model evaluation. According to RF, the highest feature importance were the number of malaria deaths and Indigenous malaria cases. The RF model demonstrated high accuracy in forecasting risk for the year 2019, achieving commendable accuracy rate of 95.3%. This result aligns well with the observed outcomes, indicating the model's reliability in predicting risk levels.

CONCLUSIONS

Machine learning algorithms have reliable application prospects in risk assessment of imported malaria in China. This study provides a new methodological reference for the risk assessment and control strategies adjusting of imported malaria in China.

Characterization of environmental drivers influencing the abundance of Anopheles maculipennis complex in Northern Italy

BACKGROUND

In Italy, malaria was endemic until the 1970s, when it was declared eradicated by WHO. Nowadays, with the persistence of competent mosquito populations, the effect of climate change, and increased possibility of importing malaria parasites from endemic counties due to growing migration, a malaria resurgence in Italy has become more likely. Hence, enhancing the understanding of the current distribution of the Anopheles maculipennis complex and the factors that influence the presence of this malaria vector is crucial, especially in Northern Italy, characterised by a high density of both human population and livestock.

METHODS

To assess the presence and abundance of malaria vectors, a 4-year field survey in the plain areas of Lombardy and Emilia-Romagna region in Italy was conducted. Every sampling point was characterised in space by the land use in a 500-m radius and in time considering meteorological data collected in the short and long time periods before sampling. We combined the results of a linear regression model with a random forest analysis to understand the relative importance of the investigated niche dimensions in determining Anopheles mosquito presence and abundance.

RESULTS

The estimated normalised variable importance indicates that rice fields were the most important land use class explaining the presence of Anopheles, followed by transitional woodlands and shrubland. Farm buildings were the third variable in terms of importance, likely because of the presence of animal shelters, followed by urbanised land. The two most important meteorological variables influencing the abundance of Anopheles in our study area were mean temperature in the 24 h before the sampling date and the sum of degree-days with temperature between 18 °C and 30 °C in the 14 days before the sampling date.

CONCLUSIONS

The results obtained in this study could be helpful in predicting the risk of autochthonous malaria transmission, based on local information on land cover classes that might facilitate the presence of malaria vectors and presence of short- and medium-term meteorological conditions favourable to mosquito development and activity. The results can support the design of vector control measures through environmental management.

Anopheles sampling collections in the health districts of Korhogo (Côte d'Ivoire) and Diébougou (Burkina Faso) between 2016 and 2018

Characterizing the entomological profile of malaria transmission at fine spatiotemporal scales is essential for developing and implementing effective vector control strategies. Here, we present a fine-grained dataset of Anopheles mosquitoes (Diptera: Culicidae) collected in 55 villages of the rural districts of Korhogo (Northern Côte d'Ivoire) and Diébougou (South-West Burkina Faso) between 2016 and 2018. In the framework of a randomized controlled trial, Anopheles mosquitoes were periodically collected by Human Landing Catches experts inside and outside households, and analyzed individually to identify the genus and, for a subsample, species, insecticide resistance genetic mutations, Plasmodium falciparum infection, and parity status. More than 3,000 collection sessions were carried out, achieving about 45,000 h of sampling efforts. Over 60,000 Anopheles were collected (mainly A. gambiae s.s., A. coluzzii, and A. funestus). The dataset is published as a Darwin Core archive in the Global Biodiversity Information Facility, comprising four files: events, occurrences, mosquito characterizations, and environmental data.

Repeat Ivermectin Mass Drug Administrations for Malaria Control II: Protocol for a Double-blind, Cluster-Randomized, Placebo-Controlled Trial for the Integrated Control of Malaria

BACKGROUND

The gains made against malaria have stagnated since 2015, threatened further by increasing resistance to insecticides and antimalarials. Improvement in malaria control necessitates a multipronged strategy, which includes the development of novel tools. One such tool is mass drug administration (MDA) with endectocides, primarily ivermectin, which has shown promise in reducing malaria transmission through lethal and sublethal impacts on the mosquito vector.

OBJECTIVE

The primary objective of the study is to assess the impact of repeated ivermectin MDA on malaria incidence in children aged ≤10 years.

METHODS

Repeat Ivermectin MDA for Malaria Control II is a double-blind, placebo-controlled, cluster-randomized, and parallel-group trial conducted in a setting with intense seasonal malaria transmission in Southwest Burkina Faso. The study included 14 discrete villages: 7 (50%) randomized to receive standard measures (seasonal malaria chemoprevention [SMC] and bed net use for children aged 3 to 59 months) and placebo, and 7 (50%) randomized to receive standard measures and monthly ivermectin MDA at 300 μg/kg for 3 consecutive days, provided under supervision to all eligible village inhabitants, over 2 successive rainy seasons. Nonpregnant individuals >90 cm in height were eligible for ivermectin MDA, and cotreatment with ivermectin and SMC was not permitted. The primary outcome is malaria incidence in children aged ≤10 years, as assessed by active case surveillance. The secondary safety outcome of repeated ivermectin MDA was assessed through active and passive adverse event monitoring.

RESULTS

The trial intervention was conducted from July to November in 2019 and 2020, with additional sampling of humans and mosquitoes occurring through February 2022 to assess postintervention changes in transmission patterns. Additional human and entomological assessments were performed over the 2 years in a subset of households from 6 cross-sectional villages. A subset of individuals underwent additional sampling in 2020 to characterize ivermectin pharmacokinetics and pharmacodynamics. Analysis and unblinding will commence once the database has been completed, cleaned, and locked.

CONCLUSIONS

Our trial represents the first study to directly assess the impact of a novel approach for malaria control, ivermectin MDA as a mosquitocidal agent, layered into existing standard-of-care interventions. The study was designed to leverage the current SMC deployment infrastructure and will provide evidence regarding the additional benefit of ivermectin MDA in reducing malaria incidence in children.

TRIAL REGISTRATIONS

ClinicalTrials.gov NCT03967054; https://clinicaltrials.gov/ct2/show/NCT03967054 and Pan African Clinical Trials Registry PACT201907479787308; https://pactr.samrc.ac.za/TrialDisplay.aspx?TrialID=8219.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/41197.

An illustration of model agnostic explainability methods applied to environmental data

Historically, two primary criticisms statisticians have of machine learning and deep neural models is their lack of uncertainty quantification and the inability to do inference (i.e., to explain what inputs are important). Explainable AI has developed in the last few years as a sub-discipline of computer science and machine learning to mitigate these concerns (as well as concerns of fairness and transparency in deep modeling). In this article, our focus is on explaining which inputs are important in models for predicting environmental data. In particular, we focus on three general methods for explainability that are model agnostic and thus applicable across a breadth of models without internal explainability: "feature shuffling", "interpretable local surrogates", and "occlusion analysis". We describe particular implementations of each of these and illustrate their use with a variety of models, all applied to the problem of long-lead forecasting monthly soil moisture in the North American corn belt given sea surface temperature anomalies in the Pacific Ocean.

A meta-epidemiological assessment of transparency indicators of infectious disease models

Mathematical models have become very influential, especially during the COVID-19 pandemic. Data and code sharing are indispensable for reproducing them, protocol registration may be useful sometimes, and declarations of conflicts of interest (COIs) and of funding are quintessential for transparency. Here, we evaluated these features in publications of infectious disease-related models and assessed whether there were differences before and during the COVID-19 pandemic and for COVID-19 models versus models for other diseases. We analysed all PubMed Central open access publications of infectious disease models published in 2019 and 2021 using previously validated text mining algorithms of transparency indicators. We evaluated 1338 articles: 216 from 2019 and 1122 from 2021 (of which 818 were on COVID-19); almost a six-fold increase in publications within the field. 511 (39.2%) were compartmental models, 337 (25.2%) were time series, 279 (20.9%) were spatiotemporal, 186 (13.9%) were agent-based and 25 (1.9%) contained multiple model types. 288 (21.5%) articles shared code, 332 (24.8%) shared data, 6 (0.4%) were registered, and 1197 (89.5%) and 1109 (82.9%) contained COI and funding statements, respectively. There was no major changes in transparency indicators between 2019 and 2021. COVID-19 articles were less likely to have funding statements and more likely to share code. Further validation was performed by manual assessment of 10% of the articles identified by text mining as fulfilling transparency indicators and of 10% of the articles lacking them. Correcting estimates for validation performance, 26.0% of papers shared code and 41.1% shared data. On manual assessment, 5/6 articles identified as registered had indeed been registered. Of articles containing COI and funding statements, 95.8% disclosed no conflict and 11.7% reported no funding. Transparency in infectious disease modelling is relatively low, especially for data and code sharing. This is concerning, considering the nature of this research and the heightened influence it has acquired.