Healthcare AI Daily Briefing
New to OpenRounds? Get the free daily email or see how we work.
Operations & Workflow
EHR‑derived ML AUROC 0.709 matches clinician VTE risk scores.
Preprint
In 577,904 admissions across five NHS hospitals, machine‑learning models using EHR‑derived VTE risk factors available within 14 hours achieved an AUROC of 0.709, compared with 0.704 for clinician‑recorded factors. The analysis assessed completion, concordance with EHR data, and clinical validity of the risk assessment forms. [1]
Research & Evidence
AI‑guided wearable reduced instability but not HF rehospitalization in LINK‑HF2.
Journal article
In a prospective randomized trial of 171 heart‑failure patients at five Veterans Affairs medical centers, participants wore a continuous multisensor patch for up to 90 days and received AI‑guided wearable monitoring or usual care; the intervention lowered AI‑detected physiological instability but did not reduce HF rehospitalization. [2]
Operations & Workflow
Baptist Health early adopter of assistive coding, 20% faster billing.
Podcast from Lifers (Second Opinion)
The interview describes that Baptist Health leveraged automated assistive coding to reduce the meantime of submitting bills and obtaining prior auth by 20%, which the interviewee said amounts to almost a one‑million‑dollar savings. The head of coding said she was reluctant at first but later saw its value in filling staffing gaps. [3]
Industry & Products
Artera’s AI‑native platform self‑hosts LLMs and claims 40% effectiveness gain.
Artera’s AI‑native platform self‑hosts six LLMs and, in a Federally Qualified Health Center with a 200‑person call center, identifies and closes patient care gaps during the same call, boosting provider quality measures and yielding a claimed 40 % increase in effectiveness. [4]
Industry & Products
AI chatbot summaries aided physicians in three‑quarters of visits.
In a study at BIDMC’s ambulatory primary care clinic, 98 patients consulted the AMIE AI chatbot before urgent‑care visits. Clinicians reported AI summaries helped prepare visits in 75% of cases and influenced care in more than half, while AMIE’s diagnoses matched doctors’ final diagnoses 90% of the time. Findings suggest AI could enhance patient‑physician relationships pending larger trials. [5]
Sources
- Machine Learning-Supported Efficient VTE Risk Assessment using Routinely Collected Electronic Health Record Data · medRxiv Original source
- Artificial Intelligence-Enabled Detection and Management of Physiological Instability in Heart Failure: Results of the Randomized LINK-HF2 Trial. · JACC. Heart failure Original source
- Healthcare runs a 1980s model in 2026 | Aaron Miri, Baptist Health · Lifers (Second Opinion) Original source
- AI-Native Artera Fills Gaps in Care and Provider Performance · Healthcare IT Today Original source
- Study in The Lancet suggests AI could improve patient-physician relationships. · Google Health Blog Original source
Daily healthcare AI brief
Get the free healthcare AI briefing
One concise briefing built from reporting, research, policy, blogs, videos, and full podcast transcripts.
Free. Sign in to subscribe. Unsubscribe anytime. Already subscribed? Sign in to save and personalize.