The research team, led by Teena Mariam Thomas and John Antony Jude Prakash of the Department of Clinical Microbiology at Christian Medical College, Vellore, together with colleagues from the hospital’s Department of Medicine, set out to answer three linked questions: where cases cluster, whether those clusters are statistically meaningful, and whether the disease follows a predictable seasonal rhythm. To do so, they combined laboratory diagnostics with geographic information systems, spatial statistics and time-series modeling, an approach that treats disease data not as a flat list of numbers but as events anchored in both place and time.

Defining a case required laboratory confirmation rather than clinical suspicion alone. A patient presenting with fever and rash was counted as a spotted fever case only if serological testing, using either an enzyme-linked immunosorbent assay or an immunofluorescence assay specific for spotted fever group rickettsiae, returned a positive result, or if molecular assays such as nested polymerase chain reaction or quantitative PCR detected the pathogen’s genetic material. This dual diagnostic strategy matters because rickettsial infections are notoriously difficult to distinguish clinically, and antibody-based tests can be influenced by prior exposure in endemic areas. Demographic details, including age, sex and residential location, were extracted from the hospital’s electronic medical records, allowing each confirmed case to be geocoded to its district of origin.

The spatial analysis relied on two complementary software platforms. Maps were generated using QGIS version 3.34.3, an open-source geographic information system, while spatial autocorrelation and hotspot detection were carried out in GeoDa version 1.22.0.21. Spatial autocorrelation asks a deceptively simple question: are cases located near other cases more often than chance would predict? The team employed Local Indicators of Spatial Association, or LISA, a statistical technique that classifies each area as a cluster of high values, a cluster of low values, or a spatial outlier. Hotspot analysis then refined this picture by identifying districts where case concentrations were statistically significant rather than random fluctuations. Seasonal patterns were quantified with R software version 4.5.2, which allowed the researchers to model monthly case counts across nearly two decades and pinpoint the peak period of transmission.

The geographic findings were striking in their concentration. Vellore district accounted for 39.9 percent of all confirmed cases, while neighboring Chittoor district, just across the state border in Andhra Pradesh, contributed 38.8 percent. Together, these two adjacent districts produced nearly four out of every five cases in the dataset. Tirupattur added 12.5 percent, and the remaining cases were distributed between Ranipet at 4.5 percent and Tiruvannamalai at 4.3 percent. This pattern suggests that the ecological conditions favoring the tick and mite vectors of spotted fever group rickettsiae are concentrated in a relatively compact zone straddling the Tamil Nadu–Andhra Pradesh boundary, rather than being evenly spread across the hospital’s catchment area.

The temporal analysis revealed an equally clear signal. Cases rose sharply between October and March, the cooler and drier months of the year in this part of peninsular India, and peaked in January. This seasonality is consistent with the ecology of the arthropod vectors responsible for transmission. Tick survival, abundance and host-seeking activity are all sensitive to temperature and humidity, and the post-monsoon cool season appears to create conditions in which both the vectors and the humans who encounter them, often through agricultural work, outdoor labor or domestic activities, come into contact more frequently. The finding that the disease peaks in the coolest months rather than the monsoon itself is a practical clue for clinicians, who might otherwise lower their guard for tick-borne illness once the rains subside.

Demographically, the study identified two groups at elevated risk: children under ten years of age and housewives. Both findings carry interpretive weight. Young children may be exposed through outdoor play, contact with domestic animals, or peridomestic environments where ticks and mites thrive, and their developing immune systems may also influence how infection presents. Housewives, meanwhile, may acquire infection close to home, through contact with vegetation, livestock, pets or infested structures in the domestic environment, rather than through occupational exposure in fields or forests. The pattern implies that spotted fever in this region is not confined to rural laborers or forest workers but is a genuinely household-level threat, one that pediatricians and primary care providers should keep firmly in view.

The clinical implications of the study center on a category of illness known as acute undifferentiated febrile illness, or AUFI, the diagnostic puzzle that occupies much of tropical medicine. Patients with AUFI present with fever but no obvious localizing signs, and the differential diagnosis spans dengue, chikungunya, malaria, leptospirosis, typhoid and rickettsial infections, among others. Because spotted fever typically announces itself with fever and an eschar or rash, and because it responds well to doxycycline when treated early, missing the diagnosis can mean prolonged illness, unnecessary investigations and, in severe cases, complications affecting the kidneys, lungs or central nervous system. The authors argue that clinicians evaluating AUFI patients with rash, particularly children under ten who live in or near the identified hotspot districts, should consider spotted fever a serious possibility during the October-to-March window, and that diagnostic capacity for rickettsial testing should be strengthened accordingly.

The methodological framework of the study also demonstrates the growing value of spatial epidemiology in resource-limited settings. By pairing routine hospital laboratory data with freely available geographic and statistical tools, the researchers converted eighteen years of clinical records into actionable public health intelligence without the expense of a dedicated surveillance network. The identification of statistically significant spatial clusters provides a rational basis for targeting interventions, whether those involve vector control campaigns, community education about tick avoidance, or the strategic placement of diagnostic services. It also generates hypotheses for future field studies: what vegetation types, land-use patterns, animal reservoirs or microclimates sustain the vector populations in Vellore and Chittoor, and why do the surrounding districts show lower transmission despite their proximity?

The authors are careful to frame their conclusions within the limits of a retrospective, single-center design. The dataset reflects patients who reached a tertiary care hospital, which means milder cases managed elsewhere are absent, and the catchment geography may partly reflect healthcare-seeking behavior as well as true disease distribution. They call for prospective, preferably multi-centric studies to confirm the hotspots, quantify the environmental drivers of transmission, and elaborate the full range of risk factors. Funded by the Indian Council of Medical Research and approved by the institutional ethics committee, the study nonetheless delivers an immediate and practical message: in the cooler months of southern India, when a child under ten arrives at a clinic with fever and a rash from one of the identified hotspot districts, spotted fever deserves a prominent place on the differential diagnosis list, and timely treatment with appropriate antibiotics could spare families the consequences of a missed diagnosis.

Subject of Research: Spatio-temporal epidemiology of tick-borne spotted fever group rickettsioses in southern India

Article Title: Spatio-temporal analysis of spotted fever cases reported to a tertiary care hospital in Southern India

Article References: Thomas, T. M., D’Cruz, S., Perumalla, S. K., Gunasekaran, K., & Prakash, J. A. J. (2026). Spatio-temporal analysis of spotted fever cases reported to a tertiary care hospital in Southern India. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14466-1

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14466-1

Keywords: spotted fever, rickettsia, tick-borne disease, spatio-temporal analysis, seasonality, India, Vellore, fever and rash, acute undifferentiated febrile illness, hotspot analysis, children, public health

The research, led by Kim Dianne Ligue-Sabio and colleagues at the University of the Philippines Mindanao, is notable not just for what it found but for how it found it. Over three fieldwork sessions conducted between May 2021 and May 2022, the team logged 270.5 hours of direct observation across 29 days, recording the behavior of visible troop members in 1,082 scan samples. Each scan lasted five minutes and was followed by a ten-minute pause, creating a fifteen-minute sampling cycle that ran from 6:00 AM to 6:00 PM. At every scan, the observers noted the primary behavior of each visible animal and recorded the GPS coordinates of the troop’s approximate center, generating a rich spatiotemporal dataset that could be interrogated with formal spatial statistics.

The five most frequently observed behaviors were moving, recorded in 99.26 percent of scans, resting at 92.42 percent, grooming at 68.48 percent, feeding at 67.74 percent, and foraging at 57.12 percent. To analyze these, the researchers turned to the Getis-Ord Gi* statistic, a hotspot detection method that identifies statistically significant clusters of activity, and to a more sophisticated technique called Emerging Hotspot Analysis. The latter builds a space-time cube, aggregating the fifteen-minute scans into hourly bins across a fishnet grid, then applies both the Getis-Ord statistic and the Mann-Kendall trend test to classify each bin as a new, consecutive, intensifying, persistent, diminishing, sporadic, oscillating, or historical hotspot or coldspot. The result is a map that shows not just where behaviors cluster, but how those clusters rise and fall through the day.

The temporal findings were strikingly clean. Foraging peaked between 8:00 and 9:00 in the morning and then declined steadily, a trend confirmed by the Mann-Kendall statistic. Grooming showed the opposite trajectory, climbing through the afternoon to peak between 4:00 and 5:00 PM. Feeding, resting, and moving showed no significant monotonic trends across the day. Read together, the data describe a troop that prioritizes energy acquisition early, exploiting both provisioned food and wild resources, and then shifts its attention to social bonding and rest as the day winds down. It is a familiar structure of primate time budgets, but here it plays out across a landscape divided between houses and forest.

The spatial analysis added a second layer of insight. Feeding, grooming, and resting clustered tightly, with significant spatial autocorrelation emerging at distance bands as short as two meters and never exceeding seventeen meters. This tight clustering suggests repeated use of specific, localized spots for social and resting activities, likely anchored to the fixed provisioning site and a consistent sleeping area within the village. Foraging, by contrast, only clustered at a thirty-meter bandwidth, and its hotspots extended deep into the forest-agricultural mosaic, well beyond the built-up zone. The researchers interpret this broader dispersion as a strategy for locating diverse food resources and reducing within-group competition by avoiding the repeated exploitation of the same patches.

Perhaps the most consequential finding concerns diet. Despite being fed regularly by the Moncadista community, a local religious sect that regards the macaques as spiritually connected to their ancestors and strictly prohibits hunting them, the animals continued to forage intensively for natural foods. Observers recorded them eating Chinese bamboo leaves, fruits, grass, and insects in the forest, alongside provisioned bananas, eggs, peanuts, and even junk food in the village. The persistence of foraging hotspots in the forest suggests that human provisioning alone does not meet the troop’s nutritional needs, particularly for a group of at least fifty individuals. Primates actively forage to maintain an optimal nutritional balance, and insects in particular provide protein that cultivated crops may lack. The implication for conservation is direct: even provisioned macaque populations depend on intact forest habitat.

The study also offers a window into why the macaques prefer the village for their downtime. Although there is no hunting pressure anywhere in the troop’s range, grooming and resting hotspots were concentrated in built-up areas while coldspots for these behaviors appeared throughout the forest. The researchers observed a telling incident in the forest: the appearance of a large red-tailed green rat snake triggered synchronized alarm calls and mobbing behavior by more than twenty macaques, a response never seen in the village. Farm deterrents such as dogs and slingshots are present in agricultural areas, while the macaques appear habituated to the domestic dogs within the settlement. This pattern fits the predator avoidance and human-shield hypotheses, in which prey animals use proximity to humans as a buffer against predators, combined with the proximity-to-food-resource hypothesis that favors resting sites near reliable food.

Methodologically, the paper makes a case for pairing static and dynamic spatial tools. The conventional hotspot analysis pinpointed key areas of activity that could serve as targets for management interventions, such as preserving foraging zones or timing tourism activities. The emerging hotspot analysis then revealed how those areas function differently across the day: feeding and foraging in the village produced oscillating hotspots, meaning the animals engaged in those behaviors elsewhere earlier, while grooming and resting produced consecutive hotspots that only intensified toward the end of the day. The two approaches validated each other where patterns agreed and exposed new structure where they diverged, though the researchers caution that aggregating scans into hourly bins may have smoothed away finer behavioral fluctuations.

The setting itself is unusual and worth savoring. New Israel sits within Mount Apo Natural Park, a protected area whose original tropical lowland dipterocarp forest has been transformed into a patchwork of residences, secondary forest, and small farms growing taro, cassava, sweet potato, and fruit trees. The community has built a small tourism industry around the macaques, complete with parks and ziplines. During the first fieldwork session, conducted amid COVID-19 pandemic restrictions when tourist visits were limited, the troop’s ranging area in the forest was markedly larger, at roughly 184,000 square meters, than in the later sessions, which shrank to under 95,000 square meters. The authors suggest that reduced tourist provisioning may have pushed the macaques to forage more widely, though visitor data were not systematically collected.

For a species whose fortunes have turned grim, the study carries real weight. In 2024, the International Union for Conservation of Nature reclassified the long-tailed macaque as Endangered, citing a 50 to 70 percent decline in global populations over three decades driven by hunting, biomedical exploitation, the pet trade, and habitat loss. Understanding how the species navigates human-modified landscapes is therefore no longer academic. The New Israel macaques demonstrate remarkable behavioral plasticity, running a daily circuit that treats village and forest as complementary resources rather than alternatives. The authors argue that conservation planning must account for both, protecting forest foraging grounds while managing the built-up areas where interactions occur. Future work, they note, should quantify provisioning rates, predator presence, and food distribution to build fuller models of what draws these adaptable primates back and forth across the boundary between two worlds.

Subject of Research: Spatiotemporal behavioral patterns of long-tailed macaques in a human-modified Philippine landscape

Article Title: Spatiotemporal patterns of behavior by long-tailed macaques Macaca fascicularis (Raffles 1821) in a human-modified environment

Article References: Ligue-Sabio, K. D., Plaza, A. R., Sabanal, B., Oguis, G. F., Torrefiel, J., & Gamalo, L. E. (2026). Spatiotemporal patterns of behavior by long-tailed macaques Macaca fascicularis (Raffles 1821) in a human-modified environment. Discover Conservation, 3(1), Article 3. https://doi.org/10.1007/s44353-025-00071-x

Image Credits: AI Generated

DOI: 10.1007/s44353-025-00071-x

Keywords: long-tailed macaques, Macaca fascicularis, hotspot analysis, emerging hotspot analysis, human-wildlife coexistence, behavioral ecology, foraging behavior, food provisioning, Mount Apo, conservation planning, primates, Philippines

The stakes could hardly be higher. The Pra, Ankobra, and Tano basins drain the Birimian greenstone belts, geological formations rich not only in gold but also in arsenic-bearing sulfides such as arsenopyrite. Over the past decade, the explosive growth of artisanal and small-scale gold mining—known locally as galamsey—has torn apart channel and floodplain sediments, accelerating the oxidation of these sulfides and flushing arsenic into rivers and adjacent alluvium. The people living along these corridors depend on the same waters for drinking, smallholder irrigation, and fisheries, which turns a hydrological problem into a direct public health and livelihood crisis. Routine monitoring campaigns have long reported frequent guideline exceedances, but the spatial pattern of that risk has been difficult to pin down with confidence.

The research team established 88 fixed monitoring sites across the three basins—42 on the Pra, 24 on the Tano, and 22 on the Ankobra—positioned to capture upstream-to-downstream gradients, confluences, municipal water abstraction points, and zones of mining disturbance. Each site was visited monthly for roughly 20 months, producing 1,752 observations that covered at least one full hydrological cycle of wet and dry seasons. Field teams collected mid-channel grab samples in acid-washed bottles, filtered and acid-preserved them on ice, and shipped them to the laboratory within 24 hours, where arsenic was quantified by inductively coupled plasma mass spectrometry or graphite furnace atomic absorption spectroscopy. In situ probes recorded temperature, pH, and specific conductance, while GPS units logged precise coordinates to within a few meters.

Before any modeling began, the researchers audited their data with unusual rigor. The dataset showed zero percent missingness, so no imputation was needed. Screening for extreme values flagged only longitude, where stations strung along narrow river corridors produced leverage-prone outliers; the team winsorized that single predictor at the 2.5th and 97.5th percentiles while deliberately leaving the arsenic measurements untouched, so that the public health burden reflected in exceedance statistics would not be artificially softened. The raw numbers told a sobering story: mean arsenic concentration was 11.95 micrograms per liter, the median 11.12, and the overall exceedance rate stood at 63 percent. By basin, Ankobra fared worst with 76.5 percent of samples over the guideline, followed by Pra at 61.3 percent and Tano at 51.4 percent.

The modeling comparison pitted a random forest against a support vector machine with a radial basis kernel and a feed-forward neural network. To eliminate so-called lucky-split bias, the team used five-times-repeated five-fold nested cross-validation, in which outer folds estimated generalization performance while inner folds tuned hyperparameters. They also tested four different train-validation-test partitions with 30 repeated stratified splits each. The random forest emerged as the most accurate or statistically tied-for-best model across nearly every configuration. At the 70-20-10 partition, it achieved a median mean absolute error of 3.86 micrograms per liter and a root mean square error of 4.93, with an area under the receiver operating curve of 0.73 for classifying exceedances. The support vector machine closed the gap at larger training sizes, but the neural network lagged consistently, its discrimination actually deteriorating as training data grew—likely a symptom of limited sample size and a sparse predictor set rather than a fundamental flaw of deep learning itself.

Crucially, the winning model was not a black box. Using SHAP analysis, a model-agnostic technique that quantifies each feature’s contribution to individual predictions, the researchers found that water temperature dominated the model’s logic by a wide margin, followed by pH and latitude. This hierarchy is chemically plausible: temperature-linked hydrochemistry is known to modulate arsenic mobility, and the strong latitudinal signal reflects broad geological and land-use gradients across the basins. Site-level location effects captured residual heterogeneity that coordinates alone could not explain, while calendar month contributed only modestly. Seasonal-trend decomposition of the 2020 to 2026 monthly series confirmed that picture, revealing weak seasonality with a peak-to-trough amplitude of roughly half a microgram per liter and only a slight downward drift of about 0.1 micrograms over the period. In other words, arsenic risk in these rivers is governed far more by spatial position and water chemistry than by the calendar.

Perhaps the study’s most methodologically important contribution lies in how it corrected for a subtle but pervasive bias in environmental monitoring. The Pra basin hosts nearly half of all stations, so naive hotspot mapping would inevitably paint broad swaths of the Pra corridor as dangerous simply because more measurements were taken there. The team countered this with three complementary techniques: an effort-adjusted exposure index that multiplies each site’s arsenic level by the inverse of local station density, a density-weighted and row-standardized Getis-Ord Gi* statistic with a 25-kilometer neighborhood that downweights oversampled areas, and a relative-risk surface that divides the kernel density of exceedances by the kernel density of sampling effort. After these corrections, the sprawling hot corridors largely vanished, leaving only sparse, localized hotspots—precise targets for field audits and enforcement rather than blanket crackdowns.

The corrected maps reshuffled the policy priorities in an instructive way. Once monitoring effort was accounted for, the Pra basin carried the largest aggregate exposure burden, with a median effort-adjusted exposure index of 321.45, reflecting the broad spatial extent of contamination across its heavily settled corridor. The Ankobra, by contrast, retained the highest probability that any single measurement would exceed the guideline, at 76.5 percent. The Tano ranked lower on both metrics but was emphatically not risk-free. The authors therefore recommend a dual-track strategy: invest in basin-scale mitigation and source control—riparian buffers, tailings containment, treatment upgrades—in the Pra, while running high-frequency compliance surveillance and rapid-response protocols in the Ankobra, where unsafe readings are most likely at any given visit. The Tano warrants sustained but proportionate monitoring as resources allow.

The study is candid about its limits. The predictor set was intentionally parsimonious—temperature, pH, discharge where available, coordinates, and time—because hydrochemical variables such as iron, dissolved organic carbon, alkalinity, and redox potential were not consistently measured across sites. The SHAP diagnostics suggest those missing chemistry variables, especially redox proxies, would add real explanatory power, and the authors call for their routine collection. Roughly 20 months of data is also too short to resolve interannual or climate-driven trends with confidence, and population-weighted exposure estimates built on 100-meter raster products can misrepresent the linear villages that hug riverbanks. The team recommends spatially balanced sampling designs, sentinel stations in sparse reaches, and spatial block cross-validation as the path forward.

What makes this work resonate beyond Ghana is its template. By pairing rigorous, repeated nested validation with transparent model explanation and bias-aware spatial statistics, the researchers have produced something rare in environmental machine learning: predictions that are accurate, defensible, and directly actionable. The framework aligns with several United Nations Sustainable Development Goals, from clean water and sanitation to health and climate action, and it transfers readily to other contaminants and other data-limited regions where monitoring stations cluster unevenly. For the millions of people who draw their daily water from the Pra, Ankobra, and Tano, the message is concrete: the danger is real, it is mappable, and the tools now exist to target remediation where it will save the most health—provided the monitoring network itself becomes as balanced as the science it supports.

Subject of Research: Machine learning prediction of arsenic exposure risk in Ghana's riverine ecosystems

Article Title: Comparative performance of machine learning models in predicting arsenic exposure risk in Ghana’s riverine ecosystems

Article References: Bosson-Amedenu, S., Boakye, E., Ayitey, E., & Addor, J. A. (2025). Comparative performance of machine learning models in predicting arsenic exposure risk in Ghana’s riverine ecosystems. BMC Environmental Science, 2(1), Article 26. https://doi.org/10.1186/s44329-025-00040-8

Image Credits: AI Generated

DOI: 10.1186/s44329-025-00040-8

Keywords: arsenic contamination, machine learning, random forest, Ghana, galamsey, river basins, water quality, SHAP interpretability, hotspot analysis, WHO guideline, Pra River, Ankobra River