Modelling the seasonal dynamics of Aedes albopictus populations using a spatio-temporal stacked machine learning model

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 30, 2025

Various modelling techniques are available to understand the temporal and spatial variations of phenology species. Scientists often rely on correlative models, which establish a statistical relationship between response variable (such as species abundance or presence-absence) set predominantly abiotic covariates. The choice modeling approach, i.e., algorithm, is itself significant source variability, different algorithms applied same dataset can yield disparate outcomes. This inter-model variability has led adoption ensemble techniques, among stacked generalisation, recently demonstrated its capacity produce robust results. Stacked incorporates predictions from multiple base learners models inputs for meta-learner. meta-learner, in turn, assimilates these generates final prediction by combining information all learners. In our study, we utilized published documenting egg observations Aedes albopictus collected using ovitraps. environmental predictors forecast weekly median number mosquito eggs machine learning model. approach enabled us (i) unearth seasonal egg-laying dynamics Ae. 12 years; (ii) generate spatio-temporal explicit forecasts regions not covered conventional monitoring initiatives. Our work establishes methodological foundation forecasting albopictus, offering flexible framework that be tailored meet specific public health needs related this

Language: Английский

---

Forecasting invasive mosquito abundance in the Basque Country, Spain using machine learning techniques

Parasites & Vectors, Journal Year: 2025, Volume and Issue: 18(1)

Published: March 15, 2025

Abstract Background Mosquito-borne diseases cause millions of deaths each year and are increasingly spreading from tropical subtropical regions into temperate zones, posing significant public health risks. In the Basque Country region Spain, changing climatic conditions have driven spread invasive mosquitoes, increasing potential for local transmission such as dengue, Zika, chikungunya. The establishment mosquito species in new areas, coupled with rising populations viremic imported cases, presents challenges systems non-endemic regions. Methods This study uses models that capture complexities life cycle, by interactions weather variables, including temperature, precipitation, humidity. Leveraging machine learning techniques, we aimed to forecast Aedes abundance provinces Country, using egg count a proxy features key independent variables. A Spearman correlation was used assess relationships between climate variables counts, well their lagged time series versions. Forecasting models, random forest (RF) seasonal autoregressive integrated moving average (SARIMAX), were evaluated root mean squared error (RMSE) absolute (MAE) metrics. Results Statistical analysis revealed impacts humidity on abundance. model demonstrated highest forecasting accuracy, followed SARIMAX model. Incorporating ovitrap counts improved predictions, enabling more accurate forecasts Conclusions findings emphasize importance integrating climate-driven tools predict mosquitoes where data available. Furthermore, this highlights critical need ongoing entomological surveillance enhance contribute development assessment effective vector control strategies expansion. Graphical

Language: Английский

---

Forecasting the abundance of disease vectors with deep learning

Ecological Informatics, Journal Year: 2023, Volume and Issue: 78, P. 102272 - 102272

Published: Aug. 20, 2023

Arboviral diseases such as dengue, Zika, chikungunya or yellow fever are a worldwide concern. The abundance of vector species plays key role in the emergence outbreaks these diseases, so forecasting numbers is fundamental preventive risk assessment. Here we describe and demonstrate novel approach that uses state-of-the-art deep learning algorithms to forecast disease abundances. Unlike classical statistical machine methods, models use time series data directly predictors identify features most relevant from predictive perspective. We for first application this predict short-term temporal trends number Aedes aegypti mosquito eggs across Madeira Island period 2013 2019. Specifically, apply whether, following week, Ae. will remain unchanged, whether it increase decrease, considering different percentages change. obtained high performance all years considered (mean AUC = 0.92 ± 0.05 SD). Our performed better than methods. also found preceding highly informative predictor future trends. Linking our transmission importation contribute operational, early warning systems arboviral risk.

Language: Английский

---

Nowcasting Vector Mosquito Abundance and Determining Its Association With Malaria Epidemics in South Korea

Transboundary and Emerging Diseases, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 1, 2025

Since a resurgence occurred in 1993, malaria has remained an endemic disease the Republic of Korea (ROK). A major challenge is inaccessibility current vector mosquito abundance data due to 2-week reporting delay, which limits timely implementation control measures. We aimed nowcast and assess its utility by evaluating predictive value for epidemic peaks. used machine learning models abundance, employing gradient boosting (GBMs), extreme (XGB), ensemble model combining both. Various meteorological factors served as predictors. The were trained with from collection sites between 2009 2021 tested 2022. To evaluate nowcasting, we calculated effective reproduction number (R t), can indicate Generalized linear (GLMs) then impact on R t. demonstrated best performance nowcasting root mean square error (RMSE) 0.90 R-squared 2) 0.85. GBM showed RMSE 0.91 2 0.84, while XGB had 0.92 Additionally, GLMs predicting t using weeks advance was >0.72 all provinces. coefficients also significant. constructed reliable abundance. These outcomes could potentially be incorporated into early warning system. Our study provides evidence support development management strategies regions where remains public health challenge.

Language: Английский

---

Modelling the small spatial scale questing abundance of Hyalomma lusitanicum Koch, 1844 (Acari: Ixodidae), vector of Crimean-Congo haemorrhagic fever virus

International Journal for Parasitology, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

A climate and population dependent diffusion model forecasts the spread of Aedes Albopictus mosquitoes in Europe

Communications Earth & Environment, Journal Year: 2025, Volume and Issue: 6(1)

Published: April 9, 2025

Language: Английский

---

Spatiotemporal dynamics of Ixodes ricinus abundance in northern Spain

Ticks and Tick-borne Diseases, Journal Year: 2024, Volume and Issue: 15(6), P. 102373 - 102373

Published: July 3, 2024

Ixodes ricinus is the most medically relevant tick species in Europe because it transmits pathogens that cause Lyme borreliosis and tick-borne encephalitis. Northern Spain represents southernmost margin of its main European range has highest rate hospitalisations country. Currently, environmental determinants spatiotemporal patterns I. abundance remain unknown this region these may differ from drivers highly favourable areas for Europe. Therefore, our study aimed to understand factors modulating questing population dynamics map northern Spain. From 2012 2014, monthly/fortnightly samplings were conducted at 13 sites two regions estimate variation abundance. Local was modelled relation local biotic abiotic conditions by constructing generalised linear mixed models with a zero-inflated negative binomial distribution overdispersed data. The different developmental stages active times year. Adults nymphs showed peak spring, while larvae more frequent summer. affecting related humidity temperature. For adults larvae, summer seemed be influential period their abundance, nymphs, winter those preceding months determining factors. abundances predicted hospitalisations. Our could basis on which build accurate predictive identify windows greatest potential interaction between animals/humans lead transmission ricinus-borne pathogens.

Language: Английский

---

Molecular Epidemiology of Travel-Associated and Locally Acquired Dengue Virus Infections in Catalonia, Spain, 2019

Viruses, Journal Year: 2025, Volume and Issue: 17(5), P. 621 - 621

Published: April 26, 2025

Dengue virus (DENV) is the most important arbovirus worldwide. In 2019, a significant increase in dengue cases was reported worldwide, resulting peak of imported some European countries such as Spain. We aimed to describe travel-associated and locally acquired DENV strains detected 2019 Catalonia region (northeastern Spain), hotspot for introduction Europe. Through sequencing phylogenetic analysis envelope gene, 75 viremic two local were described. Autochthonous transmission events included an infection mosquito with strain human from infected mosquito. Overall, all four serotypes up 10 different genotypes detected. Phylogenetic revealed transcontinental circulations associated DENV-1 DENV-2 presence DENV-4 genotype I Indonesia, where few had been previously A molecular study autochthonous determined that Ae. albopictus mosquitoes by African V strain, while case caused DENV-3 Asian origin. These findings underline wide variability high risk into this territory, emphasizing importance usefulness characterization phylogenetics both global surveillance disease.

Language: Английский

---

An innovative model for capturing seasonal patterns of train passenger movement using exogenous variables and fuzzy time series hybridization

Journal of Open Innovation Technology Market and Complexity, Journal Year: 2024, Volume and Issue: 10(1), P. 100232 - 100232

Published: Feb. 10, 2024

Train is a popular mode of ground transportation due to the ability accommodate large number passenger, save time, avoid traffic congestion, offer cost-effective fares, and provide relatively high level safety. These benefits contribute an annual increase in passenger numbers, particularly during holidays year-end period. Consequently, it essential for management anticipate potential capacity constraints faced by train operators. Detecting this challenge encompasses observing count trends at end each year, which can be effectively analyzed using Seasonal Autoregressive Integrated Moving Average (SARIMA) model account seasonal effects. surges also align with Eid Al-Fitr holiday Indonesia, event that varies annually according Hijri calendar. To address issue, SARIMA was adapted include exogenous effects form calendar variations, producing Exogenous Variables (SARIMAX). Furthermore, novel numerical proposed Fuzzy Time Series Markov Chain (FTSMC). vital operators might face. This innovation introduced through hybrid SARIMA-FTSMC SARIMAX-FTSMC. The results showed delivered highest accuracy smallest error value, providing more precise insights into movement patterns. modeling offered valuable recommendations risk limitations period, enabling optimize services effectively.

Language: Английский

---

VectAbundance: a spatio-temporal database of Aedes mosquitoes observations

Scientific Data, Journal Year: 2024, Volume and Issue: 11(1)

Published: June 15, 2024

Abstract Modelling approaches play a crucial role in supporting local public health agencies by estimating and forecasting vector abundance seasonality. However, the reliability of these models is contingent on availability standardized, high-quality data. Addressing this need, our study focuses collecting harmonizing egg count observations mosquito Aedes albopictus , obtained through ovitraps monitoring surveillance efforts across Albania, France, Italy, Switzerland from 2010 to 2022. We processed raw obtain continuous time series allowing for an extensive geographical temporal coverage Ae. population dynamics. The resulting post-processed are stored open-access database VectAbundance.This initiative addresses critical need accessible, data, enhancing modelling bolstering preparedness.

Language: Английский

---