> ## Content Index
> Fetch the complete content index at: https://www.businessbagel.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Netcare wants to use AI to catch medical emergencies 8–10 hours early
- URL: https://www.businessbagel.com/netcare-wants-ai-that-spots-a-patient-crashing-hours-before-doctors-can/
- Published: 2026-08-10T03:30:00.000Z
- Updated: 2026-08-10T19:12:02.000Z
- Description: South Africa's largest hospital group has asked the regulator to approve a tool that reads vital signs and flags danger up to ten hours early.
- Author: Christian Maidman
- Tags: Companies, Business Bagel News, #big

Imagine a monitor that notices a patient is heading for trouble long before the numbers look alarming to the human eye. That is what Netcare, South Africa's largest hospital group, has just asked the health-products regulator, SAHPRA, to license, after running a clinical study in its intensive- and high-care units.

## How it works

The tool is a machine-learning model that reads a patient's real-time vital signs — heart rate, blood pressure, oxygen levels and breathing — to predict a dangerous downturn before it becomes obvious. It is designed to catch events like heart failure, respiratory failure, sepsis or a dangerous heart rhythm, and Netcare's Professor Reitze Rodseth says it can give staff an eight- to ten-hour window to act. Built with German clinical-AI firm TCC in a partnership running since 2023, and refined over almost three years with the University of KwaZulu-Natal, it took years precisely because the team had to prove it works and is safe.

## Not the AI boogeyman

Handing patient monitoring to an algorithm invites obvious worry, and Rodseth is quick to head it off: “We don't need to be scared of the AI boogeyman... It's only a support. It's an additional safety layer.” The treatment decision, in other words, stays with the doctor. He is candid about the limits, too — noting that a model built on one population can lose accuracy when applied to another, which is part of why the work was slow.

The push fits a broader strategic shift toward data- and AI-driven care under Netcare CEO Dr Richard Friedland. And it is a big canvas to test on: Netcare runs 49 acute hospitals with almost 10,000 beds, plus 15 mental-health hospitals with about 1,100 beds through its Akeso division. In the six months to end-March, paid-patient growth was a thin 0.7%, while mental-health patient days rose 3.4%. Netcare says this is only the start: models for kidney failure, emergency-department admissions and ICU readmission are already in development, and Corsano wearable watches are monitoring thousands of at-risk patients. Whether SAHPRA signs off — and Netcare hopes for a yes within three to six months — is the next thing to watch.