Forensic AI: New Metric Aims to Prove Pollution’s Exact Link to Hospital Admissions Amid Severe Air Crisis
To successfully validate the model, Aether 360 is seeking institutional partnerships with organizations such as the WHO, UNEP, or AIIMS, using real-world, anonymized patient data.
By Rakesh Raman
New Delhi | December 18, 2025
The air quality crisis faced by national capitals like Delhi, where the sky darkens into a grey sheet during the winter, has been recognized as an annual emergency. During this “pollution season,” the air, especially in Delhi, becomes poisonous. The region frequently records Air Quality Index (AQI) levels crossing the “Severe” category (401–500) on many winter days, far exceeding the safe AQI level of below 50.
The primary hazard is fine particulate matter (PM2.5), which is extremely small, highly toxic, and can travel deep into the lungs or mix with the bloodstream. These toxic particles stem from various combined sources, including vehicular exhaust, industrial emissions, construction dust, and stubble burning.
This contamination has resulted in a critical public health crisis, leading to continuous exposure that causes severe health problems such as asthma, heart attacks, strokes, lung infections, and premature death. Hospitals consistently report higher admissions during peak pollution months. While government bodies track pollution levels and propose action plans, the issue often remains unsolved due to a lack of consistent action and enforcement of existing rules.
To bridge the gap between environmental data and these direct health consequences, researchers are developing a new AI model called Aether 360. The goal of this project is to develop the world’s first AI model capable of calculating the precise probability that a specific patient’s acute cardiac or respiratory hospital admission was directly caused by a recent spike in air pollution.
Aether 360 establishes this crucial causal connection using Explainable AI (XAI). The tool utilizes granular air quality readings alongside de-identified patient data to generate a “Pollution Probability Link” (PPL). From this data, the model calculates the core metric: the “Attribution Rate” (A-Rate), which is designed specifically to quantify this causal probability.
The developers anticipate that providing this quantifiable, localized evidence will force a profound change in policy and justify robust public health interventions, thereby breaking the current policy stalemate. The project aims to function as a high-tech forensic detective, using AI to identify and precisely measure the link between pollution spikes (the “assailant”) and specific hospital admissions (the “victims”).
To successfully validate the model, Aether 360 is seeking institutional partnerships with organizations such as the WHO, UNEP, or AIIMS, using real-world, anonymized patient data. This tool could provide the clear, undeniable evidence needed to treat toxic air as the slow-moving public health emergency it is.
By Rakesh Raman, who is a national award-winning journalist and social activist. He is the founder of a humanitarian organization RMN Foundation which is working in diverse areas to help the disadvantaged and distressed people in the society.
Donation to RMN Foundation
