AIS comprehensive radon intelligence platform: a connected ecosystem for radon awareness, decision support, and mitigation

The National Institute of Environmental Health Sciences awarded AI Solutions 87 LLC $259,415 on July 1, 2026, under the Environmental Health program (CFDA 93.113) to develop a comprehensive radon intelligence platform integrating real-time IoT sensor data, geological and meteorological information, and AI models into an interactive, high-resolution map for radon risk prediction and mitigation.

The platform addresses the public health burden of radon-induced lung cancer, estimated at 21,000 American deaths annually, by providing neighborhood-level radon predictions to homeowners, real estate agents, construction companies, policymakers, and state and local health and environmental departments. Current static county and zip-code level maps fail to capture radon variability driven by weather and building characteristics. AI Solutions 87's Phase I effort will build a secure application enabling users to upload static radon measurements and share real-time sensor data on an interactive map that integrates environmental and geological factors.

AI Solutions 87 LLC, a Wisconsin-based small business, is the recipient. Performance is based in West Bend, Wisconsin. The award period runs through June 30, 2027. This is an SBIR Phase I project grant, evaluated on technological innovation, commercial potential, feasibility of approach, and qualifications of the principal investigator and staff.

Abstract

/ summary an estimated 21,000 americans DIE from radon-induced lung cancer every year, making radon gas the second leading cause of lung cancer among smokers and the primary risk factor for non-smokers. Lack of awareness coupled with outdated maps, and gaps in testing areas leave many, especially in high-risk areas, unaware of the danger. A significant challenge in radon risk management is obtaining real-time data on radon gas variability as a function of geology, meteorology, and structure factors. Traditionally used static, low-resolution county and ZIP-code level radon potential zone maps make determining radon levels at the neighborhood-level impossible to interpret, particularly since radon concentrations are known to fluctuate wildly due to weather and dwelling characteristics. Homeowners, real-estate agents, construction companies, policymakers, and state and local health and environmental departments need an easily accessible high-resolution, interactive map that provides accurate, real-time radon predictions and solutions for mitigating radon health risks. Although attempts have been made to gather measurement data and utilize advanced computing for predicting radon risk zones, none to date have integrated all known contributing factors with AI models and been made commercially available. AI-solutions 87 (AIS), a wisconsin based small business, is developing an end-to-end software development and real-time data platform for people, businesses and elected officials concerned about radon risks for their families, customers, and constituents. Preliminary studies have shown the AIS's ability to provide a secure application with the ability to upload static measurements and share real-time IOT sensor data onto an interactive map that integrates geological, and meteorological data. In phase 1, the PI, AIS staff, and collaborating parties will determine the feasibility and user acceptability of the platform as well as AIS's ability to generate predictive models. Aim 1: engineer software that aggregates and analyzes radon measurements, environmental, geological, atmospheric and soil data, enables role-based social collaboration, and provides advanced computational resources for a web and mobile application. Aim 2: construct an ultra-high resolution radon prediction map for bozeman, MT using available indoor radon measurements from national laboratories, state and local environmental departments, and local radon testing specialists. Aim 3: establish the blueprint for a dynamic radon prediction map by enabling real-time radon exposure data integration using progressively interconnected IOT radon monitors. By the end of phase 1, AIS will have established a robust framework for data integration, manufactured a data pipeline, implemented AI/ML techniques, and deployed scalable cloud solutions to enhance dynamic radon prediction mapping. Phase II efforts include expanding coverage to the state and then national level. A significant market opportunity exists for AIS's AI-powered technology to determine current and future radon exposure risk. This preemptive intervention protects families and saves lives.