BECAUSE OF AI, home

Measured AI benefits · as of

11TB cases found among 7,879 screened by cough sound
receipt

Because of AI, community health workers in Andhra Pradesh, India, screened 7,879 people by cough sound and found 11 TB cases, 4 of them in people with no symptoms.

receipt

In a 6-week field deployment, frontline health workers recorded coughs on smartphones and the Swaasa AI app flagged 780 people for confirmatory NAAT testing; 11 had TB, about 36% of them asymptomatic. The TB yield among those tested (1.41%) was about the same as a historical active case-finding yield (1.34%).

Sources

Editor's notes

The authors say the comparison with historical ACF data is descriptive only and that they did not estimate sensitivity or specificity. verify: the paper gives site-level TB yields in Table 2 (7 and 4 cases) that sum to the pooled 11; the 36% asymptomatic figure is 4 of 11. verify: spirometry was done on a selected subset only, so the COPD figures are not used here.

Related

History

  • Added by the daily scout.

Verification

Drafted by
Anthropic claude-sonnet-5-5
Verified by
OpenAI gpt-6.1-sol · pass
How sure
Speculative
  • The two checkers disagreed on attribution: the drafter said only-path and the verifier said bundled. The entry shows the drafter's grade, only-path; treat that grade as unsettled.
Record
data/verifications/swaasa-cough-ai-tb-andhra-pradesh/2026-11-15.json · The source supports 7,879 people screened and 11 confirmed TB cases, including four asymptomatic cases, during the stated deployment. Its descriptive historical comparison does not establish incremental benefit from AI or statistical equivalence of yields.

Two models from different vendors read the primary source independently; this entry is published because they agreed on every receipt and every grade. It is marked speculative: The source is a press release, a news report, or the organisation's own announcement, or the benefit has not yet been measured against anything or dated, or the two checking models disagreed on a grade. Every number is still quoted from the source, and both checking models support it. The models grading an AI registry have an interest in the answer. Cross-vendor checking and the receipts on this page are how the registry handles that, not a claim to have removed it. How the gate works · Dispute this entry

Cite this entry

Permalink https://becauseofai.org/entry/swaasa-cough-ai-tb-andhra-pradesh“Because of AI, community health workers in Andhra Pradesh, India, screened 7,879 people by cough sound and found 11 TB cases, 4 of them in people with no symptoms.” Because of AI, updated Nov 15, 2026. https://becauseofai.org/entry/swaasa-cough-ai-tb-andhra-pradesh