Live PLATFORM

Global zoonotic spillover surveillance and early warning system.

nipahwatch.com aggregates Nipah virus outbreak signals and other zoonotic spillover events from WHO alerts, ProMED Mail, and local health ministry reports. AI-parsed for relevance, manually verified for accuracy, geo-mapped for analytical context. Built for epidemiologists, public health analysts, and OSINT researchers tracking emerging viral threats.

Live URL nipahwatch.com
Data sources WHO + ProMED + ministries
Verification AI parse + manual confirm
Coverage Global
Status Active surveillance
WHAT WE BUILT

Independent surveillance with disciplined verification.

nipahwatch ingests zoonotic disease reports from World Health Organization alerts, ProMED Mail dispatches, and official local health ministry releases. The AI parsing layer extracts case counts, geographic location, fatality data, and confirmation status. Every entry is then manually reviewed before publication. The platform makes a clear distinction between confirmed laboratory-verified cases and suspected clusters pending confirmation, and it links every entry to its primary source.

The platform is positioned explicitly as an independent OSINT tool, not a substitute for official health authority guidance. Methodology and source attribution are exposed on every page. Raw surveillance data is available as a downloadable CSV for researchers who need it in their own analytical pipelines.

TECH STACK
Cloudflare PagesGoogle Sheets backendPapaParseLeaflet.jsEmail subscription
OPERATIONAL METRICS

By the numbers.

3
Authoritative source tiers
AI+manual
Two-stage verification
Global
Geographic coverage
CSV
Raw data export
Lab-verified
Confirmed case standard
Schema.org
Structured citations
CAPABILITIES

What the dashboard provides.

Three-tier source taxonomy

WHO Disease Outbreak News for top-tier alerts, ProMED Mail for early epidemiological signals, official local health ministries for in-country confirmation. Each signal is tagged with its source tier.

AI-parsed plus human-verified

Initial parsing extracts case counts and geographic data automatically. Manual verification confirms accuracy before publication. The methodology is documented openly so users understand the verification chain.

Confirmed vs suspected case separation

Lab-verified cases and suspected clusters are tracked as distinct categories. Case Fatality Rate is computed per confirmed cohort, not blended. Distinction matters for epidemiological analysis.

Geo-mapped outbreak visualization

Leaflet-powered map with active outbreak, suspected cluster, and historical monitoring layers. Visual scan reveals geographic clustering and spread patterns.

Regional statistics table

Country-level breakdown with confirmed cases, suspected cases, deaths, CFR, and current status. Sortable, exportable, structured for downstream use.

Raw CSV export

The full surveillance dataset is published as a downloadable CSV via Google Sheets. Researchers can pull the raw data into their own analytical environments without scraping the site.

Email alert subscription

Subscribers receive notifications when a confirmed Nipah cluster is identified. Notification scope is intentionally narrow to avoid alert fatigue.

Independent OSINT positioning

The platform documents itself clearly as an independent OSINT tool, not an official health authority. Users are directed to defer to local health authorities for clinical decisions.

Resources and citations layer

Every page links back to authoritative sources (WHO, ProMED) so users can trace any claim to its origin. No information appears without a verifiable provenance trail.

WHY THIS PROJECT MATTERS

Why nipahwatch is the right shape for emerging threats.

Public health surveillance is an underserved category for independent platforms. WHO publishes alerts at official cadence. ProMED runs subscriber-only dispatches. Local ministries publish in their own languages and formats. nipahwatch sits at the intersection: aggregating, translating, geo-mapping, and standardizing this fragmented signal layer for analysts who need a single defensible view.

The architecture is deliberately conservative. Two-stage verification (AI parse plus manual review), explicit source attribution on every entry, clear methodology documentation, and a stated boundary that the platform is not a substitute for official guidance. The discipline is what makes the data trustworthy enough to be useful in a real epidemiological workflow.

RELATED WORK

More from the portfolio.

READY WHEN YOU ARE

Want a surveillance dashboard like this?

Send a brief. Tileterra Systems responds within 24 hours with scope and pricing scaled to your specific needs.

Start a project