Telangana deploys AI-driven handheld X-ray devices to improve TB diagnosis

The Union Health Ministry has introduced portable AI-powered X-ray devices under a pilot project in Telangana to improve tuberculosis detection. The technology provides instant chest X-ray analysis, helps identify missed TB cases, and enables health workers to monitor treatment progress in remote communities.

Hyderabad: A new wave of portable, Artificial Intelligence (AI)-driven diagnostic technology is showing a lot of promise in improving tuberculosis (TB) detection and long-term patient adherence to treatment.

As a part of the National TB Elimination Programme, the Union Health Ministry has deployed 52 portable handheld AI-driven X-ray devices on a trial basis in tribal belts of Adilabad, Kumuram Bheem Asifabad, and Bhadradri Kothagudem, as well as urban and peri-urban clusters in Hyderabad, including Uppal, Peerzadiguda, and Pocharam.

---

The initiative is designed to aggressively track missing cases and monitor patient recovery directly within local communities. By taking the diagnostic process straight to the patient’s bedside, the technology aims to significantly reduce patient drop-outs.

Traditionally, digital films taken during routine health camps are sent to district hospitals for specialist radiologist review, causing delays of days or weeks. As a result, the low-income patients have to travel far to collect results or drop out. Thanks to the AI-backed mobile device, X-ray results are now delivered on the spot.

The handheld units are integrated with DeepCXR, an indigenously developed deep-learning software tool created by ICMR. The software functions as an automated, on-the-spot ‘X-ray reader’ that runs locally on basic field hardware without relying on clinical metadata or external internet connectivity.

The field-level health care worker carries the camera-sized, battery-operated device directly into a household to capture a digital chest exposure of the patient. Within 30 seconds, the radiograph is transmitted wirelessly via local Wi-Fi or Bluetooth to a paired tablet, eliminating manual chemical film development.

Even in deep forests or narrow lanes with zero cellular network, the offline AI scans the image to identify structural lung abnormalities like consolidations, cavities, and fluid build-ups, displaying a triage probability score in under a minute.

Officials note that beyond initial detection, field trials have shown a lot of promise and have indicated that the technology improves treatment adherence. Health workers can now conduct regular monthly follow-ups, visually showing patients their lung lesions healing in real time on the tablet screen.

• AI-driven mobile diagnostic tools are replacing X-ray for TB diagnosis
• No need to develop the X-ray film or for a radiologist to check
• AI software works as an on-the-spot X-ray reader
• AI scans images to identify lung abnormalities, lesions, and fluid build-ups
• The device provides a triage score, indicating how likely the chest X-ray shows signs of active
• Score of less than 30, lungs are normal, and TB is clear
• If the score is more than 70, probable structural damage to lungs