Wetlands, wild Bovidae species richness and sheep density delineate risk of Rift Valley fever outbreaks in the African continent and Arabian Peninsula

Artificial Intelligence Models for Zoonotic Pathogens: A Survey

Zoonotic diseases or zoonoses are infections due to the natural transmission of pathogens between species (animals and humans). More than 70% of emerging infectious diseases are attributed to animal origin. Artificial Intelligence (AI) models have been used for studying zoonotic pathogens and the factors that contribute to their spread. The aim of this literature survey is to synthesize and analyze machine learning, and deep learning approaches applied to study zoonotic diseases to understand predictive models to help researchers identify the risk factors, and develop mitigation strategies. Based on our survey findings, machine learning and deep learning are commonly used for the prediction of both foodborne and zoonotic pathogens as well as the factors associated with the presence of the pathogens.

Machine Learning in Infectious Disease for Risk Factor Identification and Hypothesis Generation: Proof of Concept Using Invasive Candidiasis

Background

Machine learning (ML) models can handle large datasets without assuming underlying relationships and can be useful for evaluating disease characteristics; yet, they are more commonly used for predicting individual disease risk rather than identifying factors at the population level. We offer a proof of concept applying random forest (RF) algorithms to Candida-positive hospital encounters in an electronic health record database of patients in the U.S.

Methods

Candida-positive encounters were extracted from the Cerner HealthFacts database; invasive infections were laboratory positive sterile site Candida infections. Features included demographics, admission source, care setting, physician specialty, diagnostic and procedure codes, and medications received prior to the first positive Candida culture. We used RF to assess risk factors for three outcomes: any invasive candidiasis (IC) vs non-IC, within-species IC vs non-IC (e.g. invasive C. glabrata vs non-invasive C. glabrata), and between-species IC (e.g. invasive C. glabrata vs all other IC).

Results

14 of 169 (8%) variables were consistently identified as important features in the ML models. When evaluating within-species IC, for example invasive C. glabrata vs non-invasive C. glabrata, we identified known features like central venous catheters, ICU stay, and gastrointestinal operations. In contrast, important variables for invasive C. glabrata vs all other IC included renal disease and medications like diabetes therapeutics, cholesterol medications, and antiarrhythmics.

Conclusions

Known and novel risk factors for IC were identified using ML, demonstrating the hypotheses generating utility of this approach for infectious disease conditions about which less is known, specifically at the species-level or for rarer diseases.

Scoping future outbreaks: a scoping review on the outbreak prediction of the WHO Blueprint list of priority diseases

BACKGROUND

The WHO's Research and Development Blueprint priority list designates emerging diseases with the potential to generate public health emergencies for which insufficient preventive solutions exist. The list aims to reduce the time to the availability of resources that can avert public health crises. The current SARS-CoV-2 pandemic illustrates that an effective method of mitigating such crises is the pre-emptive prediction of outbreaks. This scoping review thus aimed to map and identify the evidence available to predict future outbreaks of the Blueprint diseases.

METHODS

We conducted a scoping review of PubMed, Embase and Web of Science related to the evidence predicting future outbreaks of Ebola and Marburg virus, Zika virus, Lassa fever, Nipah and Henipaviral disease, Rift Valley fever, Crimean-Congo haemorrhagic fever, Severe acute respiratory syndrome, Middle East respiratory syndrome and Disease X. Prediction methods, outbreak features predicted and implementation of predictions were evaluated. We conducted a narrative and quantitative evidence synthesis to highlight prediction methods that could be further investigated for the prevention of Blueprint diseases and COVID-19 outbreaks.

RESULTS

Out of 3959 articles identified, we included 58 articles based on inclusion criteria. 5 major prediction methods emerged; the most frequent being spatio-temporal risk maps predicting outbreak risk periods and locations through vector and climate data. Stochastic models were predominant. Rift Valley fever was the most predicted disease. Diseases with complex sociocultural factors such as Ebola were often predicted through multifactorial risk-based estimations. 10% of models were implemented by health authorities. No article predicted Disease X outbreaks.

CONCLUSIONS

Spatiotemporal models for diseases with strong climatic and vectorial components, as in River Valley fever prediction, may currently best reduce the time to the availability of resources. A wide literature gap exists in the prediction of zoonoses with complex sociocultural and ecological dynamics such as Ebola, COVID-19 and especially Disease X.

Artificial intelligence as a fundamental tool in management of infectious diseases and its current implementation in COVID-19 pandemic

The world has never been prepared for global pandemics like the COVID-19, currently posing an immense threat to the public and consistent pressure on the global healthcare systems to navigate optimized tools, equipments, medicines, and techno-driven approaches to retard the infection spread. The synergized outcome of artificial intelligence paradigms and human-driven control measures elicit a significant impact on screening, analysis, prediction, and tracking the currently infected individuals, and likely the future patients, with precision and accuracy, generating regular international and national data on confirmed, recovered, and death cases, as the current status of 3,820,869 infected patients worldwide. Artificial intelligence is a frontline concept, with time-saving, cost-effective, and productive access to disease management, rendering positive results in physician assistance in high workload conditions, radiology imaging, computational tomography, and database formulations, to facilitate availability of information accessible to researchers all over the globe. The review tends to elaborate the role of industry 4.0 technology, fast diagnostic procedures, and convolutional neural networks, as artificial intelligence aspects, in potentiating the COVID-19 management criteria and differentiating infection in SARS-CoV-2 positive and negative groups. Therefore, the review successfully supplements the processes of vaccine development, disease management, diagnosis, patient records, transmission inhibition, social distancing, and future pandemic predictions, with artificial intelligence revolution and smart techno processes to ensure that the human race wins this battle with COVID-19 and many more combats in the future.

Reproducing the Rift Valley fever virus mosquito-lamb-mosquito transmission cycle

Rift Valley fever virus (RVFV) is a mosquito-borne bunyavirus that is pathogenic to ruminants and humans. The virus is endemic to Africa and the Arabian Peninsula where outbreaks are characterized by abortion storms and mortality of newborns, particularly in sheep herds. Vector competence experiments in laboratory settings have suggested that over 50 mosquito species are capable of transmitting RVFV. Transmission of mosquito-borne viruses in the field is however influenced by numerous factors, including population densities, blood feeding behavior, extrinsic incubation period, longevity of vectors, and viremia levels in vertebrate hosts. Animal models to study these important aspects of RVFV transmission are currently lacking. In the present work, RVFV was transmitted to European (Texel-swifter cross-breed) lambs by laboratory-reared Aedes aegypti mosquitoes that were infected either by membrane feeding on a virus-spiked blood meal or by feeding on lambs that developed viremia after intravenous inoculation of RVFV. Feeding of mosquitoes on viremic lambs resulted in strikingly higher infection rates as compared to membrane feeding. Subsequent transmission of RVFV from lamb to lamb by infected mosquitoes was highly efficient in both models. The animal models described here can be used to study mosquito-mediated transmission of RVFV among the major natural target species and to evaluate the efficacy of vaccines against mosquito-mediated RVFV infection.

Farm-Level Risk Factors of Increased Abortion and Mortality in Domestic Ruminants during the 2010 Rift Valley Fever Outbreak in Central South Africa

(1) Background: Rift Valley fever (RVF) outbreaks in domestic ruminants have severe socio-economic impacts. Climate-based continental predictions providing early warnings to regions at risk for RVF outbreaks are not of a high enough resolution for ruminant owners to assess their individual risk. (2) Methods: We analyzed risk factors for RVF occurrence and severity at the farm level using the number of domestic ruminant deaths and abortions reported by farmers in central South Africa during the 2010 RVF outbreaks using a Bayesian multinomial hurdle framework. (3) Results: We found strong support that the proportion of days with precipitation, the number of water sources, and the proportion of goats in the herd were positively associated with increased severity of RVF (the numbers of deaths and abortions). We did not find an association between any risk factors and whether RVF was reported on farms. (4) Conclusions: At the farm level we identified risk factors of RVF severity; however, there was little support for risk factors of RVF occurrence. The identification of farm-level risk factors for Rift Valley fever virus (RVFV) occurrence would support and potentially improve current prediction methods and would provide animal owners with critical information needed in order to assess their herd's risk of RVFV infection.

Informing Rift Valley Fever preparedness by mapping seasonally varying environmental suitability

•

Database of Rift Valley Fever occurrences from 46 countries over 22 years.

•

Predictions of Rift Valley Fever suitability for every month over 1995–2016.

•

Identifies areas at-risk by synthesizing time-series of environmental predictions.

•

We use human and livestock data to identify possible hotspots of disease spillover.

•

We identify places where long-term and routine preparation efforts should be focused.

Background

Rift Valley Fever (RVF) poses a threat to human and animal health throughout much of Africa and the Middle East and has been recognized as a global health security priority and a key preparedness target.

Methods

We combined RVF occurrence data from a systematic literature review with animal notification data from an online database. Using boosted regression trees, we made monthly environmental suitability predictions from January 1995 to December 2016 at a 5 × 5-km resolution throughout regions of Africa, Europe, and the Middle East. We calculated the average number of months per year suitable for transmission, the mean suitability for each calendar month, and the “spillover potential,” a measure incorporating suitability with human and livestock populations.

Results

Several countries where cases have not yet been reported are suitable for RVF. Areas across the region of interest are suitable for transmission at different times of the year, and some areas are suitable for multiple seasons each year. Spillover potential results show areas within countries where high populations of humans and livestock are at risk for much of the year.

Conclusions

The widespread environmental suitability of RVF highlights the need for increased preparedness, even in countries that have not previously experienced cases. These maps can aid in prioritizing long-term RVF preparedness activities and determining optimal times for recurring preparedness activities. Given an outbreak, our results can highlight areas often at risk for subsequent transmission that month, enabling decision-makers to target responses effectively.

Risk assessment for the Rift Valley fever occurrence in China: Special concern in south-west border areas

Rift Valley fever (RVF) is a mosquito-borne zoonotic disease. Since its first outbreak in 1930, RVF epidemics have caused huge economic losses and public health impacts in Africa. In 2000, RVF became a disease of global concern as it spread to the Arabian Peninsula. In our study, a Geographic Information System-based risk assessment for the occurrence of Rift Valley fever in China was established by means of ecological niche modelling. Based on occurrence records (RVF records from FAO EMPRES-i, vector records from literatures and GBIF) and high-resolution environmental layers, the prediction maps of RVF occurrence probability and distribution of five potential RVF vectors in China were modelled using Maxent. An internal validation was adopted for model verification, and high AUC values were obtained (0.918 for RVF and 0.837-0.992 for vectors). By overlaying the RVF prediction map with the combined RVF vector prediction map using Fuzzy overlay tool ('AND' operator) of ArcMap 10.2, we got the first risk map of possible RVF vector transmission. This map was further overlaid with the latest livestock distribution map ('AND' operator) to generate the second risk map of possible RVF threat to domestic livestock. The south-west border provinces in China, Yunnan, Guangxi and Tibet were predicted to have a high possibility of RVF occurrence. Conditions conducive to the local amplification of RVF also exist in these areas. Temperature seasonality, mean temperature of dry season and precipitation of the driest month were considered as key environmental variables for RVF, and common environmental conditions were found conductive for vectors. It is suggested to establish proper surveillance systems in south-west border areas to minimize the possibility of RVF invasion. Our findings can serve as a valuable reference for prevention measures to be implemented.

Neutralizing antibodies against Rift Valley fever virus in wild antelope in far northern KwaZulu-Natal, South Africa, indicate recent virus circulation

Rift Valley fever (RVF) is a zoonotic viral disease of domestic ruminants in Africa and the Arabian Peninsula caused by a mosquito-borne Phlebovirus. Outbreaks in livestock and humans occur after heavy rains favour breeding of vectors, and the virus is thought to survive dry seasons in the eggs of floodwater-breeding aedine mosquitoes. We recently found high seroconversion rates to RVF virus (RVFV) in cattle and goats, in the absence of outbreaks, in far northern KwaZulu-Natal (KZN), South Africa. Here, we report the prevalence of, and factors associated with, neutralizing antibodies to RVFV in 326 sera collected opportunistically from nyala (Tragelaphus angasii) and impala (Aepyceros melampus) culled during 2016-2018 in two nature reserves in the same area. The overall seroprevalence of RVFV, determined using the serum neutralization test, was 35.0% (114/326; 95%CI: 29.8%-40.4%) and tended to be higher in Ndumo Game Reserve (11/20; 55.0%; 95%CI: 31.5%-76.9%) than in Tembe Elephant Park (103/306; 33.6%; 95%CI: 28.4%-39.3%) (p = .087). The presence of antibodies in juveniles (6/21; 28.6%; 95%CI: 11.3%-52.2%) and sub-adults (13/65; 20.0%; 95%CI: 11.1%-37.8%) confirmed that infections had occurred at least until 2016, well after the 2008-2011 RVF outbreaks in South Africa. Odds of seropositivity was higher in adults than in sub-adults (OR = 3.98; 95%CI: 1.83-8.67; p = .001), in males than in females (OR = 2.66; 95%CI: 1.51-4.68; p = .001) and in animals collected ≤2 km from a swamp or floodplain compared with those collected further away (OR = 3.30; 95%CI: 1.70-6.38; p < .001). Under similar ecological conditions, domestic and wild ruminants may play a similar role in maintenance of RVFV circulation and either or both may serve as the mammalian host in a vector-host reservoir system. The study confirms the recent circulation of RVFV in the tropical coastal plain of northern KZN, providing the basis for investigation of factors affecting virus circulation and the role of wildlife in RVF epidemiology.

Use of artificial intelligence in infectious diseases

Infectious diseases are caused by microorganisms belonging to the class of bacteria, viruses, fungi, or parasites. These pathogens are transmitted, directly or indirectly, and can lead to epidemics or even pandemics. The resulting infection may lead to mild-to-severe symptoms such as life-threatening fever or diarrhea. Infectious diseases may be asymptomatic in some individuals but may lead to disastrous effects in others. Despite the advances in medicine, infectious diseases are a leading cause of death worldwide, especially in low-income countries. With the advent of mathematical tools, scientists are now able to better predict epidemics, understand the specificity of each pathogen, and identify potential targets for drug development. Artificial intelligence and its components have been widely publicized for their ability to better diagnose certain types of cancer from imaging data. This chapter aims at identifying potential applications of machine learning in the field of infectious diseases. We are deliberately focusing on key aspects of infection: diagnosis, transmission, response to treatment, and resistance. We are proposing the use of extreme values as an avenue of interest for future developments in the field of infectious diseases. This chapter covers a series of applications selectively chosen to showcase how artificial intelligence is moving the field of infectious disease further and how it helps institutions to better tackles them, especially in low-income countries.

Rift Valley fever virus in small ruminants in the Democratic Republic of the Congo

Rift Valley fever (RVF) is a zoonotic viral disease caused by the RVF phlebovirus (RVFV) that infects a variety of animal species including sheep and goats. Sera (n = 893) collected between 2013 and 2015 from randomly selected indigenous sheep and goats in seven provinces of the Democratic Republic of the Congo (DRC) were tested for the presence of specific immunoglobulin G (IgG) and M (IgM) against RVFV, using two commercially available enzyme-linked immunosorbent assays. The reverse transcription polymerase chain reaction (RT-PCR) was also used to detect RVFV nucleic acid. There was significant variation in true seroprevalence of RVFV for both sheep and goats between the seven provinces investigated. Values ranged from 0.0 (95% confidence interval [CI] 0.0-6.55) to 23.81 (95% CI 12.03-41.76) for goat and 0.0 (95% CI 0.0-7.56) to 37.11 (95% CI 15.48-65.94) for sheep, respectively. One serum (1.85%) out of 54 that tested positive for IgG was found to be IgM-positive. This same sample was also positive by RT-PCR indicating an active or recent infection. These findings report the presence of RVFV in small ruminants in the DRC for the first time and indicate variations in exposure to the virus in different parts of the country.

Interspecific network centrality, host range and early-life development are associated with wildlife hosts of Rift Valley fever virus

Rift Valley fever virus (RVFV) is responsible for a substantive disease burden in pastoralist communities and the agricultural sector in the African continent and Arabian Peninsula. Enzootic, epizootic and zoonotic RVFV transmission dynamics remain ill-defined, particularly due to a poor understanding of the role of mammalian hosts in the epidemiology and infection ecology of this arbovirus. Using a piecewise structural equation model, this study sought to identify associations between biological and ecological characteristics of mammalian species and documented RVFV infection to highlight species-level traits that may influence wildlife host status. Interspecific network centrality, size of species home range and reproductive life-history traits were all associated with being an RVFV host. The identification of these species-level characteristics may help to provide ecological context for the role of wildlife amplification hosts in the epidemiology of spillover to livestock and humans and may also help to identify specific points of vulnerability at the wildlife-livestock interface.