Measures lung scarring on each follow-up CT so ILD clinics can see the trend
TL;DR Fibrogauge is software that measures how much of the lung is scarred on each high-resolution CT and plots the change between scans. Hospitals and imaging chains pay per study. Pulmonologists in India must decide whether to start or continue costly antifibrotic drugs, and eyeballing two scans side by side is unreliable.
Interstitial lung disease (ILD) is a group of conditions that scar the lungs. A 2025 review estimates 0.45–0.89 million Indians live with it[1]. In the national ILD-India registry, hypersensitivity pneumonitis made up 47.3% of cases and idiopathic pulmonary fibrosis 13.7%, and only 7.5% of patients had a biopsy[1]. HRCT and expert review are scarce outside tertiary centres, and IPF has often been mistaken for tuberculosis[1].
For families, treatment is a money decision. Generic pirfenidone costs about ₹2,880–15,120 a month[1]. So doctors need to know whether scarring is actually advancing, and visual scoring varies from one reader to the next.
What it does, in regulatory terms - quantitative imaging decision support for specialists. It measures and compares. It does not diagnose or recommend treatment. Under the Medical Devices Rules 2017 this is medical-device software that needs a CDSCO licence before sale[2].
Fibrogauge plugs into a hospital's image archive. When a patient already known to have ILD has a repeat HRCT, it segments fibrotic patterns, reports the percentage of lung affected by lobe and plots it against earlier scans. The radiologist checks the output and signs the report.
The seed proposed predicting ILD early from CT. Prediction in people without symptoms is unproven, and building it would mean years of trials, which I don't think a startup can carry. Measuring change in diagnosed patients has better evidence - a 2025 study found AI agreed with an expert reader 81% of the time and caught subtle changes visual reading missed[3]. So Fibrogauge starts there.
The patient benefits, but the buyer is a tertiary hospital's radiology department, an imaging chain or a teleradiology firm, paying per study or a yearly licence. ILD clinics and rheumatology departments, which follow connective-tissue ILD, drive the demand. I couldn't find a public Indian price for ILD quantification. Pharma firms selling antifibrotics may fund access programmes, and that raises conflict-of-interest questions you must handle in the open.
US regulators cleared two ILD CT tools in 2024 - one in January[4] and another in May[5]. CDSCO brought class C and D devices under licensing in October 2023[2], so the Indian route is defined. Generic antifibrotics have made treatment affordable enough that monitoring now changes decisions[1].
There's no public figure for India's ILD imaging software market. The narrowest number is the 0.45–0.89 million estimated patients[1], each needing repeat HRCTs during follow-up.
Do hospitals pay per scan for software-assisted reads? Teleradiology says yes. 5C Network says it reads for more than 2,000 hospitals and charges per scan[6]. That's adjacent context though, not your market.
India's best-known imaging AI firm lists chest X-ray and lung nodule products, which is useful, but no ILD tool on its site[7]. Teleradiology firms run AI-assisted reads[6]. Abroad, two ILD tools hold US clearances[4][5]. The gap is an ILD tool validated on Indian scans, where hypersensitivity pneumonitis dominates, and licensed by CDSCO.
| Difficulty | Hard (66/100) |
| Starting capital | ₹6–12 Cr |
| Time to launch | 18–30 months |
You need labelled Indian scans from many scanner models, a multi-centre validation study and a quality system good enough for a device licence. Then you get to sell into radiology departments that already juggle tonnes of AI vendors. Licensing a foreign engine is a faster but thinner alternative.
Yes, and it's the core of the business. This is medical-device software that needs a CDSCO licence before sale.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI is the product here, so the practical questions are about how you build and sell it.
Don't trust the output on scanner models or patient groups missing from your validation data, and never let it recommend treatment.
Want to work out the AI layer for Fibrogauge? Talk to Yash →
Here's what can go south -
This one is only for a team with a pulmonology or radiology co-founder, imaging-AI experience and patient capital. If you're a solo founder or want a quick SaaS play then skip it, honestly.
Prove one thing first - that three tertiary ILD centres will share de-identified serial scans and commit to a validation study.
Plan for roughly ₹6–12 Cr, because labelled scans, a multi-centre study and a device-grade quality system all come before revenue.
Yes. It's medical-device software under the Medical Devices Rules 2017 and needs a CDSCO licence before sale.
Yes, and similar tools have US clearance already. You'd need your own regulator's approval, since clearances don't carry across borders.
The free Fibrogauge validation kit has the line-by-line capital breakdown, the effort scoring, the CDSCO device-licence and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day plan, 10 interview questions for pulmonologists and radiologists, and 5 assumptions to test. Download the kit →
Thinking about building Fibrogauge or something close to it? The make-or-break work is securing multi-centre scan data and a realistic CDSCO pathway before money runs out, and I can help you map both. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.