1. Context and core idea

0.1 AI-powered tools are being used by frontline health workers to detect diseases early and reshape public health delivery.
0.2 The focus is on using simple signals such as cough sounds, X-rays, and fundus images to support diagnosis at the primary-care level.

2. AI-based cough detection for tuberculosis

0.3 In Kurukshetra, Haryana, a healthcare worker records cough sounds using a smartphone app.
0.4 The AI analyses cough patterns to indicate whether a person is likely TB-positive, flagging cases for confirmatory testing.
0.5 The Cough Against TB app screened 1.62 lakh patients during Haryana’s 100-day campaign and in Mumbai and Mizoram (2023–24).
0.6 These efforts led to a 13% increase in TB diagnoses compared to normal screening under the national programme.
0.7 The tool is still being validated by the Indian Council of Medical Research (ICMR).

3. AI-enabled X-ray devices

0.8 AI-enabled handheld X-ray devices can take images outside hospital settings, including in the community.
0.9 These devices identify lung patches indicative of TB, even before symptoms like coughing appear.
0.10 At least 473 devices are in use, with 1,500 more approved by ICMR, including indigenously developed systems.
0.11 These tools have helped detect nearly 2.85 additional asymptomatic TB cases.

4. Role in strengthening public health infrastructure

0.12 AI tools are being leveraged to speed up diagnoses and improve treatment outcomes.
0.13 Government hospitals and national institutes are collaborating with startups and research bodies to integrate AI.
0.14 This approach is crucial for last-mile access, given the reach of ASHA and ANM workers across India.

5. AI as a clinical decision support system

0.15 The government’s telemedicine platform eSanjeevani now functions as an AI-assisted decision-support guide.
0.16 The AI assists doctors by asking structured questions, recording symptom details, and suggesting probable diagnoses.
0.17 It has been trained using data from 1.36 lakh primary health centres and 18,000 hub hospitals.
0.18 A portion of consultation data feeds into the AI-enabled Clinical Decision Support System (CDSS).

6. Impact on doctors and patient flow

0.19 The AI acts as a filter, reducing doctors’ workload by improving triage and referral decisions.
0.20 It helps guide patients to appropriate facilities, improving health system efficiency.
0.21 The system enhances clinical decision-making by refining doctors’ judgments through feedback.

7. AI in disease surveillance

0.22 Disease surveillance earlier relied on field health workers and paper-based data.
0.23 An AI model now scans news articles in 13 languages to detect early signs of outbreaks.
0.24 The model tracks uncommon symptoms and patterns, including deaths linked to natural disasters.
0.25 Over time, it has been trained to identify gastroenteritis, respiratory infections, and dengue.
0.26 The system has resulted in a 150% increase in alerts, which are then investigated.

8. AI for prevention of blindness

0.27 AI-based fundus cameras detect diabetic retinopathy at an early stage.
0.28 High blood sugar damages retinal blood vessels, potentially leading to blindness.
0.29 Many patients remain unaware of retinal damage until the condition is advanced.
0.30 Early detection enables timely intervention to prevent progression.
0.31 The AI model can be integrated into existing fundus cameras at primary health centres.
0.32 This allows screening at the same location where diabetes testing is conducted.

9. Broader disease applications

0.33 Similar AI tools are being developed for detecting glaucoma and other diseases.
0.34 These initiatives aim to make preventive screening accessible at the primary healthcare level, reducing dependence on specialists.