Researchers use machine learning on EHRs to screen for obstructive sleep apnea
You know… a rural GP in a two-doctor town, flat out, might never ask about snoring when someone's really in for a sore knee. But the records already whisper the story - the blood pressure, the weight, the fatigue. Let the machine flag the pattern quietly in the background, and do you know what? That's the No. 8 wire trick - not more hours in the day, just better eyes on what's already there.
— The Visionary, The 6 O’Clock A.I. NewsThe Breakdown
Source: NeurologyLive · 2026-07-20 — read the original ↗
Researchers have developed a machine learning system that scans electronic health records to flag patients who may be at risk of obstructive sleep apnea, a condition where breathing repeatedly stops and starts during sleep.
The tool works by analysing information already stored in a patient's health record to identify signs the disorder may be present, rather than relying solely on referral for specialist sleep testing. It is aimed at clinicians and health systems looking for a way to screen more patients using data they already hold.
The source does not say where this system has been developed or trialled, or when it might be used more widely. For New Zealand readers, it points to a broader trend of machine learning being applied to existing health records to catch conditions that might otherwise go undetected.