Home fever watch and a nurse desk for the risky days after each chemo cycle
TL;DR Nadirline watches the temperature of chemotherapy patients at home during the week or two when their white cells bottom out. A nurse desk acts on the hospital's own fever rule and gets the patient into the emergency room fast. Cancer hospitals and chemo day-care centres pay per cycle. It sells speed, not prediction.
India had an estimated 14.6 lakh new cancer cases in 2022, projected to reach about 15.7 lakh by 2025[1]. Many of these patients get chemotherapy as outpatients and go home for the days when infection risk peaks. A fever then is an emergency.
At AIIMS New Delhi, febrile neutropenia accounted for 34.3% of unplanned ICU admissions of adult cancer patients, and 52.94% of those ICU patients died[2]. In a Bengaluru children's emergency unit, only 22% of febrile neutropenia episodes got antibiotics within the 60-minute target before a quality drive, and delays beyond an hour were linked to intensive-care admission[3].
And much of the lost time happens before the patient even reaches the hospital.
What it does, in regulatory terms - remote temperature monitoring and alerting against a threshold the treating hospital sets. It does not diagnose infection, recommend drugs or predict flare-ups. It uses thermometers and patches that already hold approval, and its software sits under the Medical Devices Rules 2017, so confirm its class with CDSCO[4].
After each cycle, the patient goes home with a Bluetooth thermometer and a twice-daily reminder. Patients on high-risk regimens can also get a disposable skin patch that reads continuously. Readings flow to a nurse desk run by Nadirline or by the hospital. When a reading crosses the hospital's fever rule, a nurse calls within minutes, checks the symptom list and sends the patient to the emergency room. The ER is told they are coming, so antibiotics can start on arrival.
The seed promised biosensors that predict chronic-condition flare-ups, and I've narrowed that down a lot. No sensor has been shown to predict these, and published patch pilots catch fever only about an hour before nurse checks, with sensitivity of 50–82%[5][6]. That hour still matters here, so it is the whole pitch.
The patient benefits, but the hospital pays, because a missed fever becomes an ICU bed and a reputational problem. Private cancer hospitals and chemo day-care chains would buy it per enrolled cycle. There's no public price for a comparable Indian service, so my working assumption is ₹1,500–3,000 per cycle, which the hospital may pass on to the patient as a home-care add-on.
Bluetooth thermometers are cheap, and US-cleared single-use patches already exist[7]. Two wearable-patch studies in cancer wards were published in 2025[5][6], giving buyers numbers they can read. CDSCO has published a guidance document on medical device software under MDR 2017[4], so the regulatory route is clearer than it was a few years ago.
There is no published figure for chemo-fever home monitoring in India. The firmest base is patient numbers - about 15.7 lakh new cancer cases a year by 2025[1], a large share of whom will get several chemotherapy cycles. The ICU data shows what the hospital loses each time, since one in three unplanned cancer ICU admissions at AIIMS was febrile neutropenia[2]. Treat any rupee market size as unproven.
One US device maker sells an FDA-cleared 72-hour temperature patch to providers[7]. A Korean hospital tested a wireless temperature patch on cancer inpatients[6]. Both sell a device, not a response service. In India, I couldn't find a company offering a chemo-cycle fever watch with a nurse desk. Oncology chains run phone helplines, which rely on the patient deciding to call.
| Difficulty | Moderate (48/100) |
| Starting capital | ₹0.8–1.5 Cr |
| Time to launch | 6–9 months |
The device side is off the shelf. The hard part is clinical. Each hospital needs its own fever rule, ER handover and liability terms. Your nurses need oncology training, the desk has to answer at 3 a.m., and patients must still take readings on the days they feel worst.
Yes. You handle medical devices and patient alerts, so this is firmly regulated.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI diagnose infection or replace the hospital's fever rule. Published patch pilots miss many fevers, and any software that predicts infection takes on device obligations.
Want to work out the AI layer for Nadirline? Talk to Yash →
Here's what can go south -
I think this is worth trying for a founder with a nursing or oncology-operations background and one cancer hospital willing to pilot. If you're hoping to sell a gadget or an "AI predicts flares" story then skip it. Prove one number in a pilot first - median time from home fever to ER arrival, compared with the hospital's current patients.
Plan for roughly ₹0.8–1.5 Cr, mostly devices, a 24-hour nurse desk and integration work with the first cancer hospital.
Yes, in practice. Devices must be CDSCO-licensed, the software's class must be confirmed under the Medical Devices Rules 2017 and you must comply with the DPDP Act.
Yes, febrile neutropenia is a risk everywhere chemo is given. Check your local device rules and who carries liability for missed alerts.
The free Nadirline validation kit has the line-by-line capital breakdown, the effort scoring, the device and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for oncologists, nurses and patients, and 5 assumptions to test. Download the kit →
Thinking about building Nadirline or something close to it? The make-or-break work is agreeing a fever rule, ER handover and liability terms with the first cancer hospital. 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.