Froncort.AI Grant: Google.org Backs TB Screening App With $2.3 Mn
The Froncort.AI grant from Google.org puts $2.3 Mn into TB screening in India, after a Pune pilot cut people screened per case by nearly two-thirds.
By Avinash Mishra, Business Correspondent
4 min read
Quick Take
- Google.org has given Froncort.AI $2.3 Mn, plus $200K in Google Cloud credits and free engineering help.
- In a Pune pilot, AI-guided teams screened 66,095 people to confirm 10 TB cases; standard teams screened 177,322 to find 9.
- The money will build mAI-NETRA, a mobile app headed to up to eight districts across three states, reaching two to three million people.
In this story4 sections
Pune-based healthcare AI startup Froncort.AI has won a $2.3 Mn grant from Google.org to scale tuberculosis (TB) screening across three Indian states.
The Froncort.AI grant is the startup’s second from Google.org. Its real pull is the efficiency, not the size of the cheque.
In its Pune pilot report, AI-guided teams screened about 6,610 people for each confirmed TB case. Standard campaigns screened about 19,702.
Google’s September 15 announcement named Froncort.AI one of 15 winners, chosen from more than 2,600 proposals for a $30 Mn programme. The award also brings $200K in Google Cloud credits and free engineering support. The first grant, in 2025, paid for the original AI-NETRA validation study.
Also read
What the Froncort.AI Grant Pays For
The Froncort.AI grant funds mAI-NETRA, the mobile version of its AI-NETRA platform. The app works offline, gives voice guidance in local languages, and runs on basic Android phones.
AI-NETRA mixes government health data with Google’s Earth AI satellite models. It adds local signals like settlement density, land use, rainfall and flood risk. From these it flags high-risk zones about 300 metres wide.
The system turns those zones into ranked work plans, household clusters and field routes. TB and malaria modules already work. All government health data stays and is processed inside India.
The team is building modules for six diseases in all: TB, leprosy, malaria, dengue, cholera and leptospirosis. Froncort.AI plans to put the app in up to eight high-burden districts across three states, reaching two to three million people.
Dr. Ujjwal Rao, a physician-scientist and engineer, founded Froncort.AI in 2024. The startup also runs RegDoc, a tool that drafts compliant paperwork for clinical research, drug and device firms.
Pune Pilot: Fewer People Screened, Similar Cases Found
In a two-week field test near Pune, AI targeting found about as many TB cases while screening far fewer people. The test ran across four TB units, in rural Bhor and Shirur and two urban Haveli zones.
AI-guided teams screened 66,095 people and confirmed 10 pulmonary TB cases. Standard campaigns screened 177,322 and found nine. That is about 63% fewer people screened.
Also read
The cost of finding one case fell too. It dropped from about $29K to about $10K. Among people suspected of TB, the share confirmed rose from 1.2% to 3.1%.
The diagnostics non-profit FIND ran the test with Maharashtra’s State TB Office. Froncort.AI says 92% of frontline workers reported a lighter workload.
| Measure | AI-guided (AI-NETRA) | Standard screening |
|---|---|---|
| People screened | 66,095 | 177,322 |
| Confirmed pulmonary TB cases | 10 | 9 |
| People screened per case | ~6,610 | ~19,702 |
| Cost per confirmed case | ~$10K | ~$29K |
The Numbers The Pilot Does Not Settle
The pilot also carries real caveats, which Froncort.AI lists itself. Counting all forms of confirmed TB, the AI arm found 15 cases against 32 in the standard arm.
The case numbers were small. The AI arm had no mobile X-ray van. The two arms also ran in different seasons.
A fresh trial under matched conditions is what Froncort.AI has asked for. A look back at a March 2025 campaign was striking. The targeting could have found about 89% of cases while screening under a third of those covered.
Dr. Ujjwal Rao, the founder and chief executive of Froncort.AI, put the aim plainly:
“The world does not lack experts. It lacks their reach. AI-NETRA puts that expertise in the health worker’s hands at the door: where to go, whom to screen, and when.”
Dr. Ujjwal Rao, founder and chief executive, Froncort.AI. From the company statement, November 10, 2026.
Why TB Targeting Matters In India
India carries the world’s largest TB burden, so better targeting saves money and lives. The WHO’s 2025 global TB report counted about 10.7 million TB cases worldwide in 2024, with India at about 25%.
Door-to-door screening is the main way to catch missed cases. In Pune’s standard campaigns, teams screened nearly 20,000 people to confirm one case. Targeting is meant to cut that cost sharply.
The open-source plan could matter well beyond India. Froncort.AI wants to publish mAI-NETRA as a blueprint any government can copy for free.
What this means for you
If you build public-health or healthtech tools for Indian states, Google.org just funded a rival approach that targets screening instead of widening it. Watch the eight-district rollout across three states; the open-source blueprint means you can study or adapt the model yourself.
StartupFeed Insight
The grant is small by funding-round standards, but the economics are the real story. Froncort.AI’s own pilot put the cost of finding one TB case at about $10K with AI targeting, against about $29K the standard way. That is the number state health budgets will notice, not the $2.3 Mn cheque. The caveat is real too. Counting all TB forms, the AI arm found fewer cases, and the two arms ran in different seasons. Expect the matched-conditions trial Froncort.AI itself asked for to settle this by the end of 2027.
By Avinash Mishra, Business Correspondent
Have a tip? Write to us at editorial@startupfeed.in.