A Pioneering Step Toward Zero TB Deaths
Tamil Nadu has taken a historic step. It is now the first state in India to integrate a TB death prediction model into its existing tuberculosis elimination programme.
This AI-powered model helps identify patients most at risk, ensuring faster diagnosis and quicker access to care.
What the AI Model Does
The model, developed by the ICMR-National Institute of Epidemiology (ICMR-NIE), uses real patient data to predict the probability of death from TB.
It was trained using information from 56,000 TB patients across public health facilities in Tamil Nadu between July 2022 and June 2023.
“This model alerts frontline staff about severely ill TB patients using five key indicators,” said Dr. Asha Frederick, State TB Officer.
Built into Existing System: TB SeWA
This new feature is now a part of TB SeWA (Severe TB Web Application). Tamil Nadu launched TB SeWA in 2022 under the Tamil Nadu-Kasanoi Erappila Thittam (TN-KET).
TB SeWA helps frontline health workers triage patients using both digital and paper tools at all 2,800 public health facilities in the state.
Why Prediction Matters
Time is critical. According to WHO, over 70% of TB deaths happen within the first two months of treatment.
In India, two people die of TB every three minutes. Yet, these deaths are largely preventable with timely care.
The model can predict a death probability of 10% to 50% based on indicators such as body weight and inability to stand without support. For others, the risk drops to just 1–4%.
Current Challenges in Care
Though the average time from diagnosis to admission is just one day, about 25% of severely ill TB patients still wait up to six days for admission.
“This delay can mean the difference between life and death,” explained Dr. Hemant Shewade, senior scientist at ICMR-NIE.
Global and Local Context
TB remains a major global killer. It is among the top causes of morbidity and mortality worldwide.
A study from Ethiopia titled “Time to Death and Associated Factors among Tuberculosis Patients” noted that old age, HIV co-infection, and low body weight significantly increase TB mortality.
Similar findings are seen in India. That’s why predictive tools that identify high-risk patients can be game changers.
AI in Healthcare: A New Era
By using artificial intelligence to flag severe TB cases, Tamil Nadu is setting an example. It proves that data-driven decisions can improve outcomes in public health.
The model also enables better resource allocation, ensuring that the sickest patients receive the fastest care.
Public Health Facilities Embrace Innovation
The integration covers every corner of the state. From Primary Health Centres to medical colleges, all public health institutions now use the TB SeWA system.
This broad implementation ensures that no patient—no matter how remote—slips through the cracks.
A Model for the Nation
With India aiming to eliminate TB by 2025, Tamil Nadu’s move is timely. If replicated nationwide, this model could save thousands of lives.
More states are likely to adopt similar AI-powered solutions. The fusion of technology and healthcare is no longer the future—it’s the present.
Conclusion: Saving Lives with Speed and Data
Tamil Nadu’s integration of AI into TB care marks a turning point. It blends compassion with science. It replaces delay with speed.
Most importantly, it shows that technology—when used right—can save lives. And in the fight against TB, every second counts.
