As late-season storms fuel surges in dengue and leptospirosis, the Philippines is turning to AI-powered climate–health forecasting tools to predict disease hotspots and prepare hospitals before outbreaks strike.
November continues to pose elevated public-health risks in the Philippines as climate-sensitive diseases rise following weeks of heavy rains and flooding from late-season storms.
According to advisories reported on November 22–23, the Department of Health (DOH) recorded 14,038 dengue cases from October 12 to 25, reflecting sustained transmission despite a slight decline from the previous two-week period.
DOH spokesperson Dr. Albert Domingo also confirmed that 2,396 leptospirosis cases occurred between June 8 and August 7 during prolonged monsoon flooding— a trend consistent with the disease’s known one-to-two-week lag after inundation.
This pattern is reinforced by a peer-reviewed study published in Pathogens in 2025, which found a 43.4 percent rise in nationwide leptospirosis cases from January to July 2025 compared with the same period in 2024.
As climate-linked infections intensify toward year’s end, the DOH has increasingly aligned its health advisories with rainfall and flood bulletins from the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA), enabling more climate-informed monitoring during high-risk periods.
Meanwhile, global momentum surrounding climate and health—amplified by COP28’s first-ever Health Day, where 123 countries signed the UAE Declaration on Climate and Health, and COP29’s adaptation-focused negotiations in Baku— is accelerating Philippine efforts to integrate artificial intelligence into disease surveillance.
AI Steps Into Disease Forecasting
As the country moves into the tail end of the typhoon season, health authorities warn that climate-sensitive diseases remain elevated well into the last quarter of the year.
DOH surveillance showed that dengue cases from January 1 to March 1 reached 62,313, or about 73 percent higher than in the same period in 2024, signaling a persistently high baseline even before the arrival of late-season storms.
Leptospirosis has likewise emerged as a leading rainy-season illness, with DOH officials noting heightened risks in urban communities where residents wade through contaminated floodwaters after typhoons.
Global risk analyses mirror these concerns. A 2024 anticipatory note by ACAPS, a humanitarian analysis organization, concluded that Filipinos face high vulnerability to waterborne and vector-borne diseases after typhoons because flooding expands mosquito habitats and exposes communities to polluted water.
PAGASA likewise describes the Philippine climate as tropical maritime—characterized by high humidity and abundant rainfall that sustain mosquito breeding practically year-round.
Scientific evidence continues to strengthen the link between extreme weather and disease outbreaks.
A 2025 article in Frontiers in Climate summarized global studies showing how floods, typhoons, and rising temperatures increase the risk of leptospirosis, diarrhea, and other infections.
The same paper referenced Philippine research demonstrating that heavy rainfall and floods in Manila correlate with spikes in leptospirosis hospitalizations.
The Pathogens study focusing on the Philippines reaffirmed that leptospirosis outbreaks are strongly associated with intense rainfall and flood events, particularly in densely populated, low-lying communities.
Although the traditional peak of the typhoon season spans July through October, PAGASA’s climatological records show that tropical cyclones continue to affect the archipelago well into November and December.
Climate-attribution scientists warn that late-season storms are becoming more dangerous as sea-surface temperatures rise.
A 2024 analysis by the World Weather Attribution group, reported by Reuters, found that an unusual cluster of four tropical cyclones that formed around the Philippines in November 2024 was about 70 percent more likely due to warming of 1.3°C above pre-industrial levels.
The assessment concluded that climate change has nearly doubled the risk of severe typhoons hitting the country.
DOH and PAGASA now increasingly describe these trends as climate-driven health threats. In a November 2025 report by The Chronicle, Health Secretary Teodoro Herbosa warned that climate change is “intensifying public-health risks,” noting that hotter temperatures accelerate mosquito reproduction while stronger storms and heavier rainfall create more frequent and prolonged flooding that elevates leptospirosis hazards.
DOH’s rainy-season advisories have consistently urged the public to avoid wading in floodwater and remain vigilant for symptoms of dengue, leptospirosis, and waterborne diseases.
PAGASA, meanwhile, is rolling out its 2022–2027 multi-hazard impact-based forecasting system, which shifts alerts from purely meteorological bulletins to warnings about the specific impacts expected in communities.
These developments shape the DOH’s Early Warning, Alert and Response System (EWARS), which uses PAGASA forecasts to anticipate hospital surges during prolonged rainfall.
The DOH says climate-driven surveillance allows facilities to prepare for sudden increases in dengue and leptospirosis cases that typically follow successive storms.
READ: How Innovation Is Transforming Filipino Wellness
Filipino-Led AI Models Identify Hotspots Before Cases Spike
One of the most prominent innovations this season is Project AEDES, developed by Manila-based data science firm CirroLytix.
According to the Digital Public Goods Alliance and UNICEF Venture Fund, the platform uses climate data, satellite imagery, population density, and historical DOH case reports to predict dengue hotspots.
The tool, certified as a Digital Public Good, generates heatmaps and risk scores that allow local governments to prioritize clean-up drives and hospital readiness.
UNICEF notes that Project AEDES provides “test, train, and deploy” capabilities for dengue-hotspot prediction and is designed to support scalable, open-source public-health applications.
Academics are complementing this work. In 2024, researchers from the University of the Philippines Diliman National Institute of Physics developed a perceptron-based dengue risk mapping system for Cagayan de Oro.
In a study presented at the Proceedings of the Samahang Pisika ng Pilipinas, the team reported that their model analyzed population density, built-up areas, historical dengue incidence, and environmental variables to identify vulnerable neighborhoods.
The authors said the machine-learning approach generated more detailed and sensitive risk maps than traditional statistical methods, allowing earlier interventions.
In the Davao Region, researchers at UP Mindanao are testing the PASYENTE project, which uses supervised machine learning to classify dengue-symptom severity based on patient-reported data.
Findings published in the Mindanao Journal of Science and Technology described how the prototype application recommends nearby health facilities based on waiting time and cost, and how anonymized data collected from users may help detect symptom clusters before formal case numbers rise.
These efforts reflect broader evidence on AI-assisted outbreak forecasting. Reviews in The Lancet Planetary Health and Nature Digital Medicine highlight that machine-learning models consistently outperform traditional statistical approaches in predicting dengue trends in tropical countries.
Philippine health officials say these tools support their goal of anticipating, rather than merely reacting to, climate-driven outbreaks.
Private Sector Innovations Strengthen Nationwide Forecasting
Beyond academia, private companies and multilateral initiatives are building tools to reinforce surveillance capacity. CirroLytix’s Project AEDES remains the most visible Philippine contribution to dengue forecasting, but other developments are emerging.
In 2023, Nikkei Asia reported that Japanese firm SORA Technology launched a pilot project in the Philippines using AI-equipped drones to identify stagnant water and potential mosquito habitats in hard-to-reach areas.
The system analyzes drone-captured images to detect water pools and vegetation patterns conducive to mosquito breeding, helping vector-control teams deliver larvicides more precisely.
Meanwhile, the AI for Global Health Initiative (AI4GHI) is working with Philippine partners in Iloilo to merge DOH surveillance data with satellite-derived rainfall and temperature monitoring.
According to the initiative’s project summary, machine-learning models generate barangay-level dengue risk micro-forecasts, which are then translated into simplified public advisories through large language models.
The group says the goal is to equip frontline health workers with “actionable, AI-generated insights” during typhoon-driven outbreak windows.
AI for Leptospirosis: A Developing Frontier
Compared with dengue, AI applications for leptospirosis remain in early stages in the Philippines, but global studies indicate significant potential.
A 2025 nationwide review published in Pathogens reported a 43.4 percent increase in leptospirosis cases between January and July 2025 compared with the year prior, underscoring the urgency of developing predictive systems.
A time-series analysis in PLOS Neglected Tropical Diseases found that leptospirosis hospital admissions in Manila peaked around two weeks after periods of heavy rainfall, confirming the disease’s strong link with flooding.
International researchers have begun applying machine learning to leptospirosis outcomes. A 2024 study in Scientific Reports showed that decision-tree, random-forest, and AdaBoost algorithms predicted clinical outcomes for leptospirosis patients in Brazil with accuracy approaching 90 percent.
In 2025, a Thai study in PLOS Neglected Tropical Diseases used an XGBoost algorithm to classify provincial leptospirosis risk with an area-under-the-curve score of about 0.95.
Although no Philippine pilot integrating AI with leptospirosis diagnostics or early-warning systems has yet been documented in peer-reviewed literature, experts say these global models offer a roadmap for future national applications.
Integrating Climate and Health Models
The Philippines has been strengthening its climate-sensitive disease surveillance by integrating meteorological and health data.
According to DOH epidemiologists, dengue and leptospirosis consistently surge after prolonged rainfall, prompting closer coordination with PAGASA. Climatological data show that heavy rains and storms continue late into the year, extending exposure to stagnant water that promotes disease transmission.
The collaboration intensified in line with global recommendations. During COP28 and COP30, WHO urged countries to link meteorological forecasts with public-health data to support early-warning systems.
Likewise, the Asia Pacific Strategy for Emerging Diseases (APSED III) recommends multisectoral frameworks that connect real-time case reporting, weather monitoring, and risk communication.
PAGASA confirms that rainfall outlooks, monsoon advisories, and flood-susceptibility maps are regularly shared with the DOH through national disaster and health-coordination platforms.
The DOH, in turn, overlays these climate inputs with weekly case reports from the Philippine Integrated Disease Surveillance and Response (PIDSR) system, along with entomological surveillance and hospital-admission monitoring.
These combined inputs feed into the DOH’s Early Warning, Alert and Response System, which supports short-term outbreak forecasts.
According to WHO’s Philippine office, the system allows early mobilization of vector-control teams and pre-positioning of supplies such as intravenous fluids, diagnostic kits, and antibiotics.
During intense rainfall periods, DOH regional units issue barangay-level advisories aligned with PAGASA data, encouraging communities to avoid floodwater exposure and conduct clean-up campaigns.
WHO stresses that early warning must be paired with anticipatory action. APSED guidance calls for rapid deployment of response teams, community alerts, and environmental sanitation during elevated disease-risk periods. Philippine officials say the model remains a work in progress but represents a shift toward predictive public-health planning as climate extremes intensify.
Challenges: Data Gaps, Funding, and Validation
Despite progress, significant challenges remain. A 2025 review titled Leptospirosis in the Philippines: Confronting the Structural Challenges notes that the country lacks a centralized, accessible national registry for climate-sensitive diseases, resulting in delayed reporting from many local government units. This fragmentation means AI models often rely on inconsistent data.
Model validation is another hurdle. A 2023 systematic review in Infectious Disease Informatics stressed that machine-learning models face barriers such as incomplete data, risk of bias, and overfitting, underscoring that predictive tools require rigorous testing before adoption.
A 2023 article in the Journal of Medical Science and Technology Philippines likewise urged Philippine authorities to develop validated datasets to ensure AI safety and quality.
Financing also poses challenges. WHO estimates that climate-proofing health systems will require US$2–4 billion annually by 2030 globally.
A WHO Western Pacific feature published in October 2025 noted that the Philippines is exploring new mechanisms to sustain climate-health investments, warning that without steady funding, AI-for-health innovations risk remaining isolated pilot projects.
What’s Next: AI as a National Public-Health Tool
Philippine health authorities say they are moving toward more anticipatory disease-management models as climate pressures intensify.
In a joint statement with the Climate Change Commission, the DOH emphasized that health considerations are now integrated into the country’s climate-action framework.
A 2025 scoping review in Frontiers in Climate also found a sharp rise in Philippine climate-health studies between 2018 and 2023, though gaps remain in long-term surveillance.
In addition, the Department of Science and Technology reported in May 2025 that artificial intelligence and digital-health tools are being promoted to modernize health systems, including remote diagnostics and real-time analytics.
Officials note that while the groundwork for more advanced early-warning systems is expanding through AI adoption and cross-agency collaboration, a fully operational national climate-health early-warning center has yet to be formally defined or scheduled.
Nevertheless, experts say the Philippines is laying crucial foundations for AI-supported climate–health forecasting, an approach increasingly necessary as late-season storms intensify and communities face recurring cycles of flooding, stagnant water, and disease transmission.
As climate change reshapes public-health risks, the country’s growing commitment to predictive analytics may determine how effectively it protects vulnerable populations in the years ahead.
Photo by Daniel Frank on Unsplash
