AquaPulse-OneHealth

Urban Freshwater Telemetry & Vector Surveillance Platform

Track

Track 1: AI-Supported Assessment of Urban Aquatic Ecosystems AquaPulse OneHealth integrates multimodal artificial intelligence to evaluate stream health from citizen field notes, visual imagery, and physical-chemical measurements.
Track 2: Resilience Informatics for Public and Environmental Health The platform processes ecological indicators to quantify vector-borne disease risks and generate operational action plans for both municipal planners and local communities.

main

Problem

Urban aquatic ecosystems suffer from fragmented data collection, non-point source pollution, and rapid vector-borne disease proliferation. Traditional laboratory sampling is costly, periodic, and slow to inform public health officials. Consequently, local governments lack actionable telemetry, while citizen scientists remain disconnected from municipal decision-making processes.

Solution

AquaPulse OneHealth provides an end-to-end telemetry ingestion and diagnostic engine. It converts physical observations, bio-indicator counts (such as macroinvertebrates and mosquito larvae), chemical readings, and field photos into standardized health metrics.

Using a combined analytical approach—deterministic scoring models paired with multimodal LLM diagnostics—the platform produces immediate public health alerts, dual-tiered intervention strategies, and geospatial risk mapping.

Target Users

Municipal Health & Environmental Officers: Receive structured policy recommendations, pollution discharge warnings, and targeted intervention strategies.
Urban Planners & Hydrologists: Monitor watershed integrity and spatial trends via live GIS mapping.
Citizen Scientists & Stream Stewards: Submit field telemetry, verify site conditions using photo submissions, and track community engagement via points and leaderboards.

Impact

Accelerated Hazard Detection: Reduces the delay between field observations and municipal alerts from weeks to seconds.
Dual-Action Guidance: Automatically generates distinct, actionable steps for city authorities (e.g., stormwater channel inspections) and citizens (e.g., removing standing water).
Data Standardization: Aligns raw telemetry with One Health principles and FHIR-compatible data models for interoperability with broader public health software.

Alignment with Evaluation Criteria

1. How Your Solution Aligns with OneAquaHealth

AquaPulse-OneHealth directly operationalizes the OneAquaHealth framework by linking urban aquatic ecosystem health with human public health and environmental resilience:

Holistic Interconnection: It connects physical/chemical water quality (turbidity, pH) and ecological bio-indicators (macroinvertebrate populations) to vector disease risks (mosquito breeding) and human well-being metrics.
Dual-Tiered Actionability: Translates complex ecological telemetry into immediate, complementary operational plans—policy-level interventions for municipal authorities and community-level mitigation steps for citizens.
Standards-Based Interoperability: Implements HL7 FHIR-aligned data schemas to ensure ecological and vector telemetry can directly feed into broader healthcare and municipal health systems.

2. Innovation and Practical Value

Hybrid Diagnostic Engine: Blends deterministic environmental scoring (WQI, EII, Vector Risk) with generative multimodal AI to deliver fast, reproducible, and context-aware risk evaluations.
Closing the Citizen-to-Government Gap: Replaces slow, weeks-long laboratory workflows with instant, photo-verified field assessments powered by citizen scientists.
Gamified Community Engagement: Features a built-in incentive system with points and leaderboards to maintain long-term user participation and high-density spatial data collection.
Actionable Spatial Intelligence: Interactive GIS mapping transforms raw data points into actionable hotspot visualizations for rapid emergency and municipal planning.

3. Effective Use of Data, Technology, AI, APIs, & Standards

Multimodal Visual AI (Google Gemini 2.5 Flash): Analyzes user-submitted field photos in real time to verify self-reported turbidity, stream degradation, and discharge presence against field notes.
Vector Risk Predictive ML (Hugging Face Service): Utilizes specialized machine learning microservices to infer vector-borne disease transmission potential from bio-indicator counts.
Robust Enterprise Stack (FastAPI & Pydantic v2): Implements an asynchronous, highly scalable Python backend using Pydantic v2 for strict data validation and instant API responsiveness.
Geospatial & Interoperability Standards: Leverages Leaflet.js and OpenStreetMap APIs for frontend GIS rendering, alongside HL7 FHIR R4 schemas for standardizing environmental health records.

Key Results & Capabilities

Deterministic Index Computation: Calculates Water Quality Index (WQI), Ecological Integrity Index (EII), Vector Disease Risk Score, and Human Wellbeing Impact.
Multimodal Image Verification: Decodes field photographs via Gemini Vision to cross-verify reported turbidity, vegetation, and discharge source presence.
Geospatial GIS Tracking: Maps telemetry collection points on an interactive layer to highlight disease risk hotspots and stream degradation zones.
Community Incentive System: Tracks user submissions, awards points, and calculates steward ranks to maintain high engagement in citizen science programs.