GCGabriel Carrasco-Escobar

Gabriel Carrasco-Escobar

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Gabriel Carrasco-Escobar has a total of 410 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Comparing watershed-based spatial units and administrative boundaries for malaria prediction in the Peruvian Amazon

    by Antony Barja, Edson J. Ascencio, Bryan Fernandez-Camacho, Luis C. Revilla, Jazmin Qquellon, Gabriel Carrasco-Escobar

    Malaria continues to pose a major public health threat in the Peruvian Amazon, where climate variability influences transmission patterns. This study assessed whether watershed-based spatial units (hydrobasins) outperform traditional administrative boundaries in predicting P. vivax and P. falciparum cases in Loreto, Peru. A retrospective ecological analysis was conducted using village-level geocoded malaria case data from 2009 to 2018, aggregated at the district level and three hydrobasin levels (levels 6, 7, and ANA). Environmental and meteorological variables were derived from satellite data and incorporated into Zero-Inflated Negative Binomial regression and machine learning models (Random Forest, Support Vector Machine, and XGBoost). Model performance was evaluated using Root Mean Squared Error (RMSE) and Akaike Information Criteria (AIC). Among 297,324 malaria cases analyzed (79.48% P. vivax , 20.52% P. falciparum ), hydrobasin level 7 consistently showed superior predictive accuracy. For P. falciparum , hydrobasin level 7 consistently produced the lowest prediction errors across modelling approaches. Among machine learning models, XGBoost achieved an RMSE of 15.79 (95% CI: 10.42-21.03), whereas the corresponding ZINB-PCA model achieved an RMSE of 39.19 (95% CI: 17.09-62.24). District- and ANA-based ZINB-PCA models yielded RMSE values of 50.89 (95% CI: 23.65-77.87) and 59.22 (95% CI: 33.26-82.21), respectively. For P. vivax , XGBoost achieved an RMSE of 38.44 (95% CI: 26.90-49.19), whereas the corresponding ZINB-PCA model achieved an RMSE of 63.87 (95% CI: 46.38-80.95). District-, hydrobasin level 6-, and ANA-based ZINB-PCA models produced RMSE values of 102.81 (95% CI: 73.15-131.77), 175.87 (95% CI: 116.08-232.61), and 263.78 (95% CI: 144.98-389.87), respectively. These findings suggest that the choice of spatial unit affects predictive accuracy, with hydrobasin level 7 producing the lowest prediction errors; the associations captured by the models are predictive and are not interpreted as causal effects. Integrating hydrologically defined boundaries into surveillance systems may enhance malaria control strategies in environmentally complex regions.

  • Open Access

    A low-cost environmental monitoring framework for climate-sensitive disease surveillance in the Peruvian Amazon

    by Bryan Fernandez-Camacho, Edson J. Ascencio, Antony Barja, Luis C. Revilla, Diego Villa, Penélope S. Brou, Yomali Ferreyra-Chombo, Lucia Bartolini-Arana, Rachel Lowe, Gabriel Carrasco-Escobar

    Malaria, dengue and leptospirosis are endemic in the Peruvian Amazon, where routine surveillance operates without village-scale environmental observation. We report the yield of that monitoring framework, the completeness of its meteorological record, and its cost, operated at village scale alongside notifiable disease surveillance. Across ten villages along the Iquitos–Nauta Highway, we operated drones, automatic weather stations, low-cost particulate monitors and acoustic recorders, inspected terrain and geolocated cases. Co-creation with the regional health directorate and villages governed deployment and data handling. Streams were harmonized into a spatial database at daily to annual resolution. Nine quarterly campaigns between March 2023 and March 2026 covered 14 flight areas, 14 km 2 per round, and produced 756 orthomosaics from 1.16 million images. Sensors returned 678,564 meteorological and 2,778,363 air-quality records and 4349.75 h of audio. Completely missing outdoor-temperature weeks ranged from 5.1 to 61.4% across sites. Of 10,883 notified individuals, 1489 (13.7%) were georeferenced, with coverage of 8.3% in 2010–2015 and 31.1% in 2021–2025. Capital outlay was about USD 22,050, each campaign about USD 3660. The harmonized series is intended to inform vector control and case detection for local health authorities and stakeholders.

  • land4health: Remote Sensing Metrics for Spatial Health Analysis

    by Antony Barja, Gabriel Carrasco-Escobar

    Calculate and extract remote sensing metrics for spatial analysis in the field of health. The package offers R users a quick and straightforward way to obtain areal or zonal statistics of key environmental indicators, covariates, and vector-borne disease data ideal for modeling infectious diseases within the framework of spatial epidemiology.

  • Association of mortality and combined oxidative capacity of ozone and nitrogen dioxide

    by Yue Niu, Renjie Chen, Francesco Sera, Aleš Urban, Ana Maria Vicedo-Cabrera, Yasushi Honda, Veronika Huber, Yuming Guo, Shilu Tong, Micheline de Sousa Zanotti Staglior Coelho, et al.


  • Open Access

    Investigating antibody cross-reactivity and transmission dynamics of alphaviruses and flaviviruses using a multiplex serological assay

    by Victor Yman, Jason Rosado, Noé Ochida, Laura Garcia, Marie-Fabrice Gasasira, Gaëlle Baudemont, Estee Cramer, Myrielle Dupont-Rouzeyrol, Karl Huet, Maylis Douine, et al.

    Accurate serological tools are essential for monitoring the transmission of arboviruses with pandemic potential, yet cross-reactivity between closely related viruses hampers diagnostics and surveillance. Here, we develop a high-throughput multiplex serological assay to quantify antibody responses to 28 antigens from nine arboviruses (dengue, Zika, yellow fever, West Nile, Usutu, Japanese encephalitis, chikungunya (CHIKV), Mayaro (MAYV), and O'nyong-nyong virus) and apply it to over 4000 samples from epidemiologically distinct sites on four continents. We implement a flexible analytical method based on Bayesian finite mixture models and Receiver Operating Characteristic analysis to evaluate assay performance and define seropositivity thresholds. As a case study, we resolve cross-reactive and virus-specific responses for CHIKV and the emerging MAYV by combining competitive immunoassays with mathematical modelling of multiplex serological and epidemiological data. This approach yields cross-reactivity-adjusted estimates of local transmission dynamics, in agreement with existing epidemiological evidence, and reveals that CHIKV is more prone to induce cross-reactive antibody responses than MAYV. Our results demonstrate the power of combining multiplex serology with experimental validation and modelling to disentangle exposure histories in the face of serological cross-reactivity. This integrative approach holds promise for improving arbovirus surveillance, particularly in settings with overlapping transmission of multiple viruses and limited diagnostic capacity.


  • Open Access

    Drone-based geospatial prediction modeling identifies Fasciola hepatica infection risk in the Cusco Highlands of Peru

    by Bryan Fernandez-Camacho, Antony Barja, Luis C. Revilla, Rodrigo A. Ore, Jose L. Alccacontor-Muñoz, Maria L. Morales, Melinda B. Tanabe, Gabriel Carrasco-Escobar, Miguel M. Cabada

    Background: Fascioliasis is a neglected infectious disease affecting agricultural communities worldwide, with the Peruvian Andes among the most severely affected regions. Identifying fine-scale environmental risk patterns could support targeted surveillance and control. We aimed to develop predictive models of Fasciola hepatica infection in humans and sheep using drone-derived environmental indices in a rural Andean community. Methods: We conducted a cross-sectional study in the Huayllapata community, Cusco, Peru. Demographic, socioeconomic, and georeferenced infection data were collected from households and livestock with fascioliasis diagnosed by stool microscopy. High-resolution multispectral and thermal drone surveys were performed in April 2023 to derive environmental, topographic, and climatic indices. Logistic regression, random forest (RF), XGBoost (XGB), and deep learning models were trained using literature-based or principal component analysis (PCA)-based variable selection strategies. Model performance was evaluated using standard and spatial cross validation approaches. Fine-scale probability surface maps were generated across the study area. Results: Human fascioliasis prevalence was 21.3% of households, while sheep prevalence reached 80%. Under standard cross validation, RF achieved the best performance for human infection using the literature-based approach (accuracy = 0.89, sensitivity = 0.99, specificity = 0.88), while XGB performed best using the PCA-based approach (accuracy = 0.85, sensitivity = 0.75, specificity = 0.85). For sheep infection, XGB achieved the highest performance (accuracy = 0.93, sensitivity = 0.65, specificity = 0.93) with literature-based variables and RF performed best under the PCA-based approach (accuracy = 0.85, sensitivity = 0.75, specificity = 0.86). Spatial cross-validation reduced accuracy and specificity across models but preserved high sensitivity. Probability maps revealed marked spatial heterogeneity in predicted risk within the community, with shifts in the location and magnitude of risk zones when spatial dependence was accounted for. Conclusions: In this single Andean community, machine learning models integrating drone-derived environmental, topographic and climatic indices, successfully identified F. hepatica infection occurrence in humans and sheep. RF and XGB showed the most robust performance under spatial cross-validation, supporting the feasibility of UAV-based approaches for localized F. hepatica risk mapping. Graphical


  • Open Access

    Genomic epidemiology of SARS-CoV-2 in Peru from 2020 to 2024

    by Benjamin Sobkowiak, Amy Langdon, Pedro E. Romero, Gabriel Carrasco-Escobar, Diego Villa, Renato Cava Miller, Víctor Cornejo Villanueva, Alejandra Dávila-Barclay, Diego Cuicapuza, Guillermo Salvatierra, et al.

    Background: Peru recorded one of the world’s highest COVID-19 mortality rates, with nearly 4.5 million reported cases and 220,000 deaths by March 2024. Understanding the emergence and spread of SARS-CoV-2 variants in this context is key to informing effective public health responses. This study describes the genomic diversity, transmission dynamics, and geographic spread of SARS-CoV-2 in Peru from 2020 to 2024. Methods: We analyzed nearly 50,000 high-quality public SARS-CoV-2 genome sequences collected nationwide between March 2020 and March 2024. Phylogeographic and mutational analyses were performed to identify variant lineages, trace their origins, and map viral movements within and beyond Peru. Results: We show that Peru’s epidemic waves were shaped by the emergence of locally evolved variants, including Lambda (C.37), Gamma (P.1.12), and Omicron (XBB.2.6 and DJ.1) sub-lineages. The city of Lima acted as the primary hub for inter-regional spread, accounting for 47.3% of inferred viral movements to other departments, notably Ancash, Cusco, and Piura. Peru was the source of various lineages that spread internationally, primarily to Chile, the USA, and Europe. Mutational analysis highlighted critical mutations in the spike protein, including L452Q and F490S in Lambda, associated with immune evasion and increased transmissibility. Conclusions: This work demonstrates the capacity of genomic surveillance in Peru to detect and track emerging SARS-CoV-2 variants, providing insights into regional and global transmission dynamics in a high-transmission, middle-income country setting. Sustained, cost-effective genomic monitoring, combined with strengthened bioinformatics and laboratory capacity, is essential for pandemic preparedness in resource-limited settings.


  • Open Access

    Urban and socioeconomic disparities in PM 2.5 exposure across 340 Latin American cities

    by Edson J Ascencio, Antony Barja, Jose Montes-Alvis, Josiah L Kephart, Nelson Gouveia, Daniel A Rodriguez, Tarik Benmarhnia, Ana V Diez Roux, Usama Bilal, J Jaime Miranda, et al.

    Background. Fine particulate matter (PM 2.5 ) is a leading global health risk. Latin American cities exhibit some of the world’s highest urban PM 2.5 levels, yet studies of neighborhood-level PM 2.5 exposure and associated disparities in the region are limited. Methods. We conducted a cross-sectional ecological analysis of 53 041 neighborhoods across 340 cities in eight Latin American countries, leveraging the Salud Urbana en America Latina study dataset. Annual PM 2.5 concentrations were derived from satellite data and linked to socioeconomic and urban characteristics. A multilevel model analyzed associations between neighborhood PM 2.5 levels and neighborhood- and city-level characteristics. Results. The median annual neighborhood PM 2.5 concentration was 18.49 µ g m −3 . Of the 256 million residents, all lived in neighborhoods with ambient PM 2.5 concentrations that exceeded the 2021 World Health Organization guidelines (5 µ g m −3 ). Variability was greatest between cities (54.3% of total variance), but substantial within-city variation (26% of variance) was observed. Higher neighborhood PM 2.5 levels were associated with higher neighborhood educational attainment (mean difference [MD] comparing top to bottom tertile = 0.17 µ g m −3 ), higher neighborhood intersection density (MD comparing top to bottom tertile = 0.17 µ g m −3 ), and older cities (MD comparing top to bottom tertile = 1.45 µ g m −3 ). Lower neighborhood PM 2.5 levels were related to higher neighborhood population density (MD comparing top to bottom tertile = − 0.55 µ g m −3 ), more greenness (MD comparing top to bottom tertile = − 0.76 µ g m −3 ), and larger distance from city centers (MD comparing top to bottom tertile = − 0.86 µ g m −3 ). Conclusions. Neighborhoods with higher PM 2.5 concentrations tended to have higher educational attainment, more intersections, and be located in older cities, while lower concentrations were associated with denser populations, more green space, and greater distance from city centers. Our findings reveal important within-city heterogeneity in PM 2.5 and the factors associated with it, suggesting strategies to mitigate air pollution within cities.


  • Open Access

    Rapid geographic expansion of local dengue community transmission in Peru

    by Edson J. Ascencio, Paloma M. Carcamo, Gabriel Carrasco-Escobar


  • Top Cited Papers
    Open Access

    Social disparities in neighborhood flood exposure in 44,698 urban neighborhoods in Latin America

    by Josiah L. Kephart, Usama Bilal, Nelson Gouveia, Olga Lucia Sarmiento, Emily Shingara, Karla Rangel Moreno, Maryia Bakhtsiyarava, Juan Pablo Rodriguez, Salvador Ayala, Gabriel Carrasco-Escobar, et al.

    Climate change is expected to greatly increase exposure to flooding, particularly among urban populations in low- and middle-income countries. Here we used daily flood data (2000–2018) to describe socioeconomic disparities in flooding on the basis of neighborhood educational attainment, comparing disparities across and within cities. We used multilevel models to examine disparities in area flooding by city- and neighborhood-level factors, including 44,698 neighborhoods in 276 cities from eight countries with a total of 223 million residents and 117 distinct flood events. One in 4 of residents of neighborhoods in the lowest quintile of educational attainment were exposed to flooding, compared with 1 in 20 residents of neighborhoods in the highest quintile of educational attainment. Neighborhoods experiencing more flooding included those with lower educational attainment, that were coastal, less dense, further from the city center and greener, and that had steeper slopes. We show large social disparities in neighborhood flooding within Latin American cities. Policymakers must prioritize flood adaptation and recovery efforts in neighborhoods with lower socioeconomic status.