The system, called Col-Ovo, is an artificial intelligence tool that automatically detects and counts Aedes aegypti eggs on photographs of oviposition substrates taken under real field conditions. Its development grew directly out of an operational surveillance program rather than a laboratory exercise. The researchers trained their model on 275 oviposition substrates, each a strip roughly 20.5 centimeters long, collected through routine monitoring activities run in partnership with the Vector-Borne Diseases Surveillance and Control Program of Santa Marta. That provenance matters. Laboratory image datasets are typically clean, well-lit, and carefully framed, which makes them poor proxies for what health workers actually encounter. Field substrates arrive stained, dirty, and inconsistently photographed, and any counting system that cannot cope with that messiness will fail the moment it leaves the bench.

The technical heart of Col-Ovo is a YOLOv11m object-detection model. YOLO, short for You Only Look Once, is a family of convolutional neural network architectures that performs object detection in a single forward pass through the network, rather than the multi-stage proposals used by older detection pipelines. That design choice makes YOLO models fast enough to run near real-time inference, a property the Colombian team exploited to reduce the processing time for a single substrate from approximately fifteen minutes of expert manual counting to less than three seconds of automated analysis. The model was trained on smartphone images captured in situ, with no preprocessing, no standardized lighting rig, and no laboratory-grade camera equipment. The images included substrates treated with bioattractants, specifically blackstrap molasses and the yeast Saccharomyces cerevisiae, which are used in the low-cost OviCol strategy to make ovitraps more attractive to egg-laying females. Those attractants stain the substrate darkly, and dark stains are precisely the kind of visual noise that confounds naive computer vision approaches, since mosquito eggs are themselves small, dark, elongated objects against a darkening background.

Col-Ovo was also deliberately engineered to tolerate a constraint that most academic computer vision papers ignore: image compression from messaging platforms. In community-based surveillance programs across Latin America, the practical workflow often involves local volunteers photographing substrates in their neighborhoods and sending the images to a central coordinator over WhatsApp or a similar application. Messaging services aggressively compress images to save bandwidth, destroying fine detail and introducing blocky artifacts that can obliterate the subtle texture differences between an egg and a speck of sand. By training and evaluating on images subjected to this kind of compression, the researchers built a system that fits the actual communication infrastructure of the communities it is meant to serve, rather than demanding a workflow the communities would have to build around it. The substrates in the dataset also contained sand and particulate debris, further stress-testing the detector’s ability to distinguish genuine eggs from lookalike clutter.

To judge whether the algorithm could be trusted, the researchers benchmarked it against two kinds of human evaluators. The first was expert manual counting, the traditional gold standard performed by trained technicians. The second was a virtual-human count conducted through the OviLab platform, in which evaluators examined images at greater than 200 percent magnification to resolve eggs that would be difficult to see at native resolution. Agreement between Col-Ovo and the human references was assessed using correlation and concordance analyses, the statistical tools of choice for asking whether two measurement methods agree closely enough to be interchangeable. The results were strong. Col-Ovo achieved a Spearman rank correlation of 0.91 with human evaluators, indicating that when the system reports a high count, humans report a high count too, and the ordering of substrates by infestation level is preserved. Lin’s concordance correlation coefficient, which penalizes deviations from the identity line and not merely correlation, came in at 0.89, showing that the automated counts were not just proportional to human counts but closely matched them in magnitude.

The absolute error figures tell a similarly encouraging story. The mean absolute error between Col-Ovo and human counts was 22.1 eggs per substrate. On substrates carrying hundreds of eggs, an error of that size is operationally negligible for surveillance purposes, where the goal is to rank neighborhoods and time periods by mosquito abundance rather than to certify exact totals. Perhaps most tellingly, when the researchers compared paired Col-Ovo counts against OviLab virtual-human counts, the difference between the two methods was not statistically significant, with a P value of 0.252. In practical terms, the algorithm behaved like another trained counter in the room, one that never fatigues, never needs retraining on visual recognition, and works three hundred times faster than its human colleagues.

The performance held up under the harshest conditions the field could supply. Substrates that were heavily stained by the molasses and yeast bioattractants, and those contaminated with sand and debris, were exactly the samples where manual counting is slowest and most error-prone, and they were precisely the samples Col-Ovo was built to handle. The system successfully detected and quantified eggs across this spectrum of difficulty, and its accuracy remained robust when the input images had been compressed by messaging applications. That combination of robustness and speed transforms the economics of ovitrap surveillance. A program processing a thousand substrates would previously have faced roughly 250 hours of expert counting labor; with Col-Ovo, the same workload requires under an hour of machine time, freeing technicians for the control interventions that actually reduce mosquito populations.

The broader significance of the work lies in how it fits into an emerging digital infrastructure for community-led vector control. Col-Ovo is designed to integrate with OviLab, the platform through which virtual-human counts are performed, creating a flexible pipeline in which routine smartphone photographs flow into automated analysis, with human review reserved for ambiguous or high-stakes cases. The OviCol strategy that underpins the surveillance effort combines monitoring with control in a single low-cost ovitrap design, making it feasible for neighborhoods to run their own mosquito programs. What has been missing is a way to process the resulting flood of substrates without a small army of counters. By eliminating the need for specialized equipment, standardized imaging setups, or any image preprocessing, Col-Ovo addresses what the authors identify as a critical barrier to scaling ovitrap surveillance to the thousands of deployment sites a city-wide program requires.

The study was conducted within the legal mandates and public health competencies of the Santa Marta vector control authorities, and the authors disclose that they were involved in developing both Col-Ovo and OviLab, a relationship declared for transparency. The work was supported by Fundación Chilloa and by a postdoctoral grant from MinCiencias-Colombia. Those details matter because the tool’s credibility rests on having been forged in operational conditions rather than demonstrated in a curated dataset. If systems like Col-Ovo can be validated and deployed across other endemic regions, the implications extend beyond dengue to Zika, chikungunya, and yellow fever, all of which ride on the same vector. Rapid, accurate, and scalable egg counting turns ovitraps from a labor-intensive research instrument into a genuine early-warning network, one in which a volunteer with a phone in a Santa Marta backyard can contribute a data point to a city-wide picture of mosquito activity within seconds of pressing the shutter. In the ongoing contest between urban humans and Aedes aegypti, that kind of speed may prove decisive.

Subject of Research: Smartphone-based artificial intelligence for automated counting of Aedes aegypti eggs in ovitrap-based vector surveillance

Article Title: Col-Ovo: Smartphone-based artificial intelligence for rapid counting of Aedes mosquito eggs under field conditions

Article References: Col-Ovo: Smartphone-based artificial intelligence for rapid counting of Aedes mosquito eggs under field conditions. (n.d.). https://doi.org/10.1186/s13071-026-07734-1

Image Credits: AI Generated

DOI: 10.1186/s13071-026-07734-1

Keywords: Aedes aegypti, Col-Ovo, artificial intelligence, object detection, YOLOv11m, ovitraps, dengue, Zika, chikungunya, vector surveillance, computer vision, smartphone imaging

The research, led by Jaqueline S. Cruz, Hussein Khalil, and colleagues working with long-standing infectious disease programs in Salvador, set out to answer a deceptively simple question: which socioeconomic and environmental factors determine both where Aedes aegypti mosquitoes flourish and which residents show evidence of past chikungunya virus infection? To do so, the team spent months embedded in a vulnerable informal community, deploying ovitraps and drawing blood samples from residents between October 2021 and June 2022. The result is one of the most granular portraits yet of how urban poverty, ecology, and viral transmission intersect at the scale of individual households.

Chikungunya virus, or CHIKV, is an arbovirus transmitted primarily by Aedes mosquitoes, the same vectors responsible for dengue and Zika. First identified in Tanzania in the 1950s, the virus has swept through tropical and subtropical regions in successive waves, causing explosive epidemics characterized by fever and often debilitating joint pain that can persist for months or years. In Brazil, incidence has risen sharply in recent years, with the burden falling disproportionately on socioeconomically marginalized neighborhoods where irregular water supply, poor waste collection, and unpaved surfaces create abundant breeding habitat. Despite this clear ecological logic, the microscale drivers of transmission within informal settlements have remained poorly quantified, largely because most surveillance operates at municipal or district levels too coarse to capture household-level heterogeneity.

To overcome this limitation, the researchers combined two complementary strands of evidence. The first was entomological: 194 ovitraps, simple containers holding water and a wooden paddle where female Aedes aegypti deposit their eggs, were distributed across the study area and monitored regularly. Egg counts per trap served as a proxy for vector abundance, capturing the spatial texture of mosquito proliferation at resolutions far finer than conventional indices. The second was serological: the team enrolled 888 residents aged five years and older, collecting blood samples to test for antibodies against chikungunya virus, alongside detailed questionnaires covering sociodemographic characteristics, housing conditions, and environmental exposures.

Analyzing these data required statistical machinery capable of handling both spatial autocorrelation and multiple sources of uncertainty. The team employed Integrated Nested Laplace Approximation, or INLA, a Bayesian computational framework well suited to spatial modeling, to fit models linking ovitrap egg counts and individual seropositivity to their candidate predictors. Because the models report credible intervals rather than single point estimates, the findings convey a calibrated picture of how confident one can be in each association, an important safeguard when working with complex, noisy community data.

The entomological results were striking. Ovitraps placed in peridomicile locations, the spaces immediately surrounding homes such as yards and alleys, recorded substantially higher egg counts than traps in other settings, with an incidence rate ratio of 1.99 and a 95 percent credible interval spanning 1.31 to 3.06. In other words, simply being near a house nearly doubled the expected mosquito egg density, implicating the shaded, cluttered, and water-holding niches around dwellings as primary breeding grounds. Households reporting food insecurity, a marker of deep socioeconomic deprivation, also showed higher egg counts around their properties, as did homes with unpaved yards and locations surrounded by greater garbage density, which carried a risk elevation of 25 percent per unit increase.

The serological analysis told a parallel story about human exposure. Among the 888 participants, several factors were independently associated with testing positive for chikungunya antibodies. Elderly residents faced roughly double the odds of seropositivity compared with younger adults, a pattern consistent with cumulative lifetime exposure and possibly with reduced access to protective behaviors such as air conditioning or intact window screens. Housing quality emerged as a powerful signal: living in a home with uncovered brick walls, a hallmark of unfinished informal construction, doubled the odds of prior infection, while having a yard, whether paved or unpaved, was associated with twofold to nearly threefold increases in seropositivity. Residents who reported personally seeing mosquitoes in or around their homes showed 94 percent higher odds of seropositivity, suggesting that individual perception tracks real exposure reasonably well.

Intriguingly, the model also incorporated remotely sensed satellite data, specifically Land Surface Temperature, and the spatially smoothed ovitrap egg counts as continuous covariates of seropositivity. Both variables showed positive associations with point estimates that did not exclude the null in the credible intervals, but their inclusion demonstrates a methodological advance: temperature measured from orbit at microscale resolution, and local mosquito abundance itself, can be layered into risk models alongside household characteristics. Warmer microclimates accelerate mosquito development and viral replication, and the authors argue that such fine-scale thermal variation deserves attention in future control planning, particularly as urban heat islands intensify.

Taken together, the findings paint a coherent picture of chikungunya risk as a product of converging vulnerabilities. The same environmental degradations that concentrate mosquito breeding, unpaved ground, accumulated refuse, and precarious construction, also coincide with the socioeconomic conditions, food insecurity, elderly residents, and unfinished housing, that heighten human exposure. Crucially, the spatial overlap between ovitrap hotspots and seropositive clusters suggests that where mosquitoes are most abundant, residents are indeed being infected, validating ovitraps as a cheap and actionable surveillance tool that could direct interventions before outbreaks ignite.

The authors conclude that effective chikungunya control in informal settlements cannot rely on generic campaigns alone. Instead, strategies must be tailored to the neighborhood scale: upgrading basic urban infrastructure, clearing garbage, paving and managing yards, targeting breeding sites around homes where egg densities peak, engaging communities in mosquito-source reduction, and incorporating microscale temperature data into risk maps. In an era of expanding arboviral threats across the tropics, this study offers a blueprint for precision public health, showing that with careful spatial epidemiology, the invisible geography of risk in a vulnerable community can be made visible, and ultimately, actionable.

Subject of Research: Socioeconomic and environmental drivers of Aedes aegypti distribution and chikungunya transmission risk in a low-income Brazilian community

Article Title: Socio-environmental assessment of Aedes aegypti distribution and risk of chikungunya in vulnerable community

Article References: Cruz, J. S., Argibay, H. D., Nery, N., Mogaji, H., Barreto, T. D. M., Batista, L. E. S., Silva, L. R. S., Nery, V. K., Ferreira, V. C. D., Carneiro, I. D. O., López, Y. A. A., Reis, M. G., Ko, A. I., Diggle, P., Begon, M., Costa, F., & Khalil, H. (2026). Socio-environmental assessment of Aedes aegypti distribution and risk of chikungunya in vulnerable community. PLOS Neglected Tropical Diseases, 20(10), e0014776. https://doi.org/10.1371/journal.pntd.0014776

Image Credits: AI Generated

DOI: 10.1371/journal.pntd.0014776

Keywords: chikungunya, Aedes aegypti, Brazil, spatial epidemiology, ovitraps, seroprevalence, Bayesian modeling, informal settlements, vector-borne disease, public health, land surface temperature, PLOS Neglected Tropical Diseases

The research, led by Eimear Griffin and colleagues, combined entomological fieldwork, household surveys, and satellite-based remote sensing to build a detailed picture of mosquito infestation in two disadvantaged communities of Salvador. Between March and December 2023, the team deployed ovitraps—small containers designed to attract egg-laying female mosquitoes—in 174 households, while conducting visual inspection surveys in 119 of them to catalogue every potential breeding site in the peri-domestic environment. The result was one of the most granular datasets yet assembled on how environmental and behavioral factors shape mosquito density in tropical urban settings.

The numbers tell a striking story. Of the 303 ovitraps recovered at the end of the study period, 55.6 percent of those placed indoors and a remarkable 82.0 percent of those placed outdoors tested positive for Aedes aegypti eggs. In total, the researchers recorded 11,211 eggs across the study area. That outdoor positivity rate, far exceeding the indoor figure, underscores a central theme of the work: the peri-domestic zone—the patios, alleys, and yards immediately surrounding dwellings—serves as the principal engine of mosquito production in these communities.

Using multivariate statistical modeling, the team identified a set of environmental factors strongly associated with egg abundance. Ovitrap location was itself a key predictor, confirming the indoor-outdoor gradient. Beyond location, three household and neighborhood characteristics stood out: garbage deposited in front of the house, proximity within ten meters of an open sewer, and the presence of open plastic containers in the peri-domestic environment. Each of these features creates or sustains the standing water that female mosquitoes require for oviposition, and each is amenable to targeted intervention.

The open sewer finding deserves particular attention. In many rapidly growing tropical cities, informal drainage channels carry untreated wastewater through residential streets, and households often live within a few meters of them. The study demonstrates that this proximity is not merely an aesthetic or olfactory nuisance but a measurable driver of vector density. Standing water in and around poorly maintained sewer infrastructure, combined with the organic richness of such sites, appears to create productive larval habitats that conventional indoor-focused control programs never reach.

Discarded and open plastic containers represent another modifiable risk factor with clear behavioral roots. Plastic waste accumulates quickly in neighborhoods with limited collection services, and even a small bottle cap or food container can hold enough rainwater for a complete mosquito life cycle. The visual inspection surveys conducted by the research team systematically documented these potential breeding sources, linking them directly to the egg counts recorded in nearby ovitraps. This connection between observed household practices and measured vector density is what gives the study its practical force: it does not simply describe where mosquitoes are abundant, but identifies the specific, changeable conditions that produce that abundance.

Perhaps the most surprising result, however, is what the researchers did not find. Despite the intense transmission potential of Aedes aegypti in urban Brazil, the spatial analysis revealed no significant clustering of mosquito egg density across the study area. Infestation was not concentrated in particular blocks or hotspots that could be targeted with area-wide spraying. Instead, the drivers of proliferation were distributed at the scale of individual households and their immediate surroundings. This has profound implications for control strategy, suggesting that broad, undifferentiated interventions may be less effective than fine-grained, community-led actions tailored to the conditions of each home and street.

To contextualize these household-level findings, the team also performed landcover classification of the study area using remote sensing data, quantifying vegetation coverage and other landscape features. While the multivariate models ultimately pointed to proximate environmental factors—waste, sewers, containers—as the dominant predictors, the integration of satellite imagery with ground-based entomology reflects a growing methodological trend in vector surveillance. Combining these data streams allows public health authorities to characterize urban environments at scale and to prioritize neighborhoods where disadvantaged conditions intersect with favorable mosquito habitat.

The study’s authors argue that their evidence supports a shift toward integrated strategies that combine improved waste disposal, better water storage practices, and upgraded sanitation infrastructure. In disadvantaged communities characterized by overcrowding, poor sanitation, and limited access to formal vector control, these structural improvements could amplify the effectiveness of conventional measures such as insecticide application and larval source management. Crucially, because the identified risk factors are visible and actionable at the household level, residents themselves can play a direct role—clearing garbage from front areas, covering or removing open plastic containers, and advocating for sewer maintenance—making community-led intervention a realistic complement to municipal programs.

As dengue and other arboviruses continue to expand their footprint across tropical cities, driven by urbanization, climate change, and interconnected travel, studies of this kind offer a template for precision public health. By identifying the peri-domestic environment as the decisive battleground and by pinpointing garbage, open sewers, and unmanaged containers as the key modifiable drivers, the Salvador study converts a diffuse and daunting problem into a set of concrete, local actions. For the millions of people living in similar conditions worldwide, the message is both sobering and empowering: the mosquitoes that threaten them are bred in their own backyards, and the tools to stop that breeding are, to a remarkable degree, already in their hands.

Subject of Research: Environmental and behavioral drivers of Aedes aegypti mosquito infestation in disadvantaged tropical urban communities

Article Title: Household and community drivers of Aedes aegypti infestation in tropical cities: Guiding community-led interventions

Article References: Griffin, E., da Mata Barreto, T., Ward, Z., Santana, J. O., Khalil, H., Begon, M., Reis, M. G. G., Cremonese, C., Costa, F., Argibay, H. D., & Howard, G. (2026). Household and community drivers of Aedes aegypti infestation in tropical cities: Guiding community-led interventions. PLOS Neglected Tropical Diseases, 20(10), e0014728. https://doi.org/10.1371/journal.pntd.0014728

Image Credits: AI Generated

DOI: 10.1371/journal.pntd.0014728

Keywords: Aedes aegypti, arboviruses, dengue, vector control, ovitraps, urban sanitation, peri-domestic environment, open sewers, plastic waste, Salvador Brazil, public health, remote sensing

The invasive mosquito has expanded its range across German cities in recent years, creating a nuisance for residents and a genuine public health concern because the species can transmit pathogens such as dengue, chikungunya and Zika viruses under suitable conditions. While intensive control and monitoring programs have demonstrated that local populations can be suppressed, these efforts are labor-intensive and expensive, prompting scientists to search for ways to target limited resources more efficiently.

In the new research, published in the journal Parasites & Vectors, the team focused on a 121-hectare suburban area in Pfaffengrund where active mosquito control was already underway. As part of routine control measures, breeding sites on 1,455 land parcels were repeatedly recorded between May and September 2023, offering an unusually detailed census of where water-holding containers suitable for mosquito larvae could be found. In parallel, egg counts from 39 ovitraps, standard surveillance devices that attract gravid females to lay eggs on wooden paddles or similar substrates, provided a direct measure of egg-laying activity across the neighborhood.

The statistical analysis hinged on generalized linear models and generalized linear mixed models, tools that allowed the researchers to test how building type moderated the effect of control measures on breeding sites, and how the composition of the surrounding urban landscape influenced trap counts. For each ovitrap, the team examined circular buffers of 100, 200 and 300 meters, calculating the proportion of parcels occupied by single-family houses, terraced houses and multi-family buildings, along with breeding site availability and the presence of parcels that were not included in control activities. Separate models were fitted for each buffer size and their performance was compared to identify the spatial scale that best explained the observed patterns.

The results were striking in their consistency. Terraced building plots contained the largest number of breeding sites, while multi-family house plots contained the fewest. This pattern likely reflects the structural features of terraced housing, where small private gardens, narrow courtyards and dense rows of interconnected properties offer numerous sheltered niches for water-collecting containers, from flower pot saucers to rain barrels, while shared access can complicate inspection and control efforts.

Building type also shaped the landscape-level signal in the ovitrap data. A higher share of terraced building plots within the surrounding area was associated with higher maximum egg counts in nearby traps, indicating that females were finding more favorable oviposition opportunities in these microhabitats. The researchers interpreted maximum egg counts as a measure of peak reproductive pressure at a location, making it a sensitive indicator of where mosquito populations were concentrating their egg-laying effort.

Perhaps counterintuitively, maximum egg counts were lower when the share of uncontrolled building plots in the buffer was higher and when more large breeding sites were present. The authors suggest that when untreated parcels and substantial breeding habitats dominate an area, female mosquitoes may disperse their eggs across those natural and semi-natural containers rather than concentrating them in the artificial attractants of ovitraps, diluting the apparent trap signal even where actual mosquito abundance may be high. This underscores a key lesson of surveillance ecology: ovitrap counts reflect both mosquito activity and the availability of competing breeding options in the environment.

Scale mattered as well. The best model performance emerged at a buffer radius of 200 meters, suggesting that the relevant neighborhood for predicting egg-laying behavior extends a few hundred meters around any given trap. However, when the temporal variability of the egg count data was incorporated into the analysis, no single optimal buffer size could be identified, and the effect of breeding site abundance lost statistical significance. The researchers caution that the noisy, seasonally fluctuating nature of ovitrap data limits how precisely the spatial drivers of egg laying can be pinned down, and that the maximum-count approach, while informative, should be complemented by temporal modeling in future studies.

For mosquito control programs, the practical implications are considerable. Because terraced house plots are positively associated with both breeding site abundance and ovitrap egg counts, control measures could be prioritized in neighborhoods dominated by this building type to increase efficiency. In an era when municipal budgets for invasive species management are under pressure, being able to direct inspections, larvicide applications and community engagement toward the most productive habitats represents a meaningful gain in cost effectiveness. Rather than treating every parcel equally, authorities can use building structure, a feature already mapped in most European cities, as a rapid proxy for risk.

More broadly, the study highlights how the built environment itself functions as ecological infrastructure for invasive species. As Aedes albopictus continues its northward expansion in a warming climate, the interplay between housing form, human behavior and vector biology will increasingly determine which streets and courtyards become mosquito strongholds. The Heidelberg findings provide a template that other cities in Germany and beyond could adapt, combining parcel-level breeding site records with trap surveillance and spatial modeling to design smarter, more targeted control campaigns against one of the world’s most successful urban invaders.

Subject of Research: Effects of urban building structure on breeding site distribution and ovitrap egg counts of the invasive mosquito Aedes albopictus in Heidelberg, Germany

Article Title: Effects of urban building structures on the distribution of breeding sites and ovitrap egg counts of Aedes albopictus in Pfaffengrund, Heidelberg, Germany

Article References: Wahl, C., Lautenbach, S., Tokatlian Rodríguez, A., & Becker, N. (2026). Effects of urban building structures on the distribution of breeding sites and ovitrap egg counts of Aedes albopictus in Pfaffengrund, Heidelberg, Germany. Parasites & Vectors, 19(1), Article 393. https://doi.org/10.1186/s13071-026-07655-z

Image Credits: AI Generated

DOI: 10.1186/s13071-026-07655-z

Keywords: Aedes albopictus, Asian tiger mosquito, invasive species, mosquito control, urban ecology, breeding sites, ovitraps, terraced houses, vector-borne disease, Germany, urban structure, egg counts