Healthcare AI Daily Briefing

Diagnostics & Pathology

Machine learning identifies seizure onset zone with 95% within-subject accuracy.

Journal article

We tested whether pHFO network properties could characterize seizure propagation in 20 patients with >9,000 channel‑hours of intracranial EEG from 46 seizures. Within‑subject clustering of pHFO network features identified each patient’s seizure onset zone with 95% accuracy. [1]

Medical Imaging

AI shows promise for MASLD risk stratification but lacks outcome‑improving trials.

Journal article

A narrative review finds that machine‑learning and deep‑learning models matched or exceeded conventional non‑invasive tests for fibrosis assessment, histologic interpretation and hepatocellular carcinoma risk stratification in selected datasets. However, the evidence is mostly retrospective, single‑center and lacks external validation, and no randomized trial has shown AI‑guided management improves clinical outcomes in MASLD. [2]

Industry & Products

Utah expands AI sandbox to let an acne‑treating bot prescribe drugs after clinician sign‑off.

In Utah’s AI sandbox, regulators authorized a pilot where Nolla Health’s AI evaluates mild‑to‑moderate acne via structured intake and skin photos, then recommends a low‑risk drug; a licensed clinician reviews and signs each prescription initially, with possible autonomous prescribing later. [4]

Industry & Products

Hippocratic AI cites open‑source build for cost advantage in patient interactions.

Podcast from Lifers (Second Opinion)

The interview describes that Hippocratic AI built its models with open source and says this lets it ensure good costs, while other AI firms spend so much on model development they lose money on every minute of patient interaction. [6]

Industry & Products

Interview: Guide’s AI platform aims to acquire broker businesses to boost returns.

Podcast from Claims Denied: A Hospitalogy Podcast

As reported in the interview, Guide’s AI platform targets brokers serving Medicare Advantage and small‑to‑medium employer markets (individuals under 65); the company acquires the broker firms it serves, surrounds each with a dedicated operations team, and expects five‑times economic and operational returns from the model. [3]

Industry & Products

Martin McKee turns his lens on AI in health policy.

Martin McKee, professor of European public health at the London School of Hygiene and Tropical Medicine and research director of the European Observatory on Health Systems and Policies, has spent four decades studying collapse and recovery—including the post‑Soviet transition, the 2008 financial crisis, and the COVID‑19 pandemic—and has now turned his lens on AI. [5]