FDA clears Epic deterioration model; Bayer sells Blackford radiology AI; chest X-ray AI flags osteoporosis; Houston Methodist ties OR AI to more cases
Listen to this post
read by AI voices ยท all episodesTHE SHORT READ
The Food and Drug Administration has cleared Epic Systems' Deterioration Index Version 2, a model in Epic's record that predicts which hospitalized adults will decline, in a 510(k) decided Sept. 18, agency records show.
The model is Epic's first AI clearance from the agency's device center, according to a review of September clearances by the consultancy Innolitics. Its validation showed lower sensitivity for Black and Hispanic patients than for white patients at the medium-risk threshold, Becker's Hospital Review reported.
In radiology, Sopris Capital closed its purchase of Blackford Analysis from Bayer, and Korean researchers reported a chest X-ray model that identified osteoporosis in three outside cohorts. In surgery, Houston Methodist's innovation chief described an operating-room camera system linked to 7% more cardiothoracic cases a month, and a Northwell Health team published a skills score built from robotic console data.
THE DETAILS
- FYIRegulation Oct 8, 2026The FDA cleared the model Sept. 18 as substantially equivalent to AgileMD's eCARTv5, about 8.5 months after Epic filed, according to the agency's 510(k) record and Innolitics. It scores adult inpatients and observation patients from 0 to 100 for the risk of an ICU transfer, rapid response call or code within 24 hours or death within 72 hours, using 195 inputs from the Epic record, and each hospital sets its own alert thresholds, Becker's Hospital Review reported. In Epic's retrospective testing on about 10.7 million scores from three U.S. sites, the area under the curve was 0.83 for the combined outcome, and at the medium-risk threshold sensitivity was 26% for Black patients, 32% for Hispanic patients and 51% for white patients, according to Becker's. The gap between Black and white patients narrowed in a site-by-site analysis, Becker's reported. The model is locked and the clearance includes no predetermined change control plan, so a retrained version would likely need a new submission, Innolitics' review said. A 2024 Yale New Haven Health study ranked the earlier version near the bottom of six early warning scores, Becker's reported. Correction (Oct. 14): This item has been corrected. An earlier version said the FDA cleared the model about eight months after Epic filed, did not say that the retrospective testing was Epic's, and omitted that the gap in sensitivity between Black and white patients narrowed in a site-by-site analysis; the FDA received Epic's application Jan. 2 and cleared the model Sept. 18, 259 days or about 8.5 months later, and Becker's reported that Epic tested the model retrospectively and that the gap between Black and white patients narrowed in a site-by-site analysis.
- FYIMoney Oct 9, 2026Sopris Capital closed its purchase of Blackford Analysis, a Scottish company whose platform gives imaging departments access to more than 130 AI applications from more than 45 vendors, AuntMinnie reported Oct. 9. The price was not disclosed. Bayer bought Blackford in 2023 and moved away from its AI platform business the following year, according to AuntMinnie. Sopris plans to fold Blackford into Azra AI, a company it already owns, in the coming weeks, while Blackford remains a separate legal entity with operations in Scotland and the United States.Source: AuntMinnie
- FYIEvidence Oct 9, 2026Researchers led by Seoul National University Bundang Hospital trained the model on 46,478 chest radiographs paired with DXA scans and tested it on a later hospital cohort of 6,391 exams and two health-checkup cohorts of 24,132 and 122,535, according to the npj Digital Medicine paper published Oct. 9. The area under the curve for osteoporosis was 0.88 in the hospital cohort and 0.93 and 0.92 in the checkup cohorts. In a simulation, using the model to decide who gets DXA cut scans by about half while detecting 98% to 99% of osteoporosis cases, the authors reported. The study was observational and retrospective, and several authors reported ties to imaging AI companies, which they said were unrelated to the work.Source: npj Digital Medicine
- FYIDeployment Oct 9, 2026Roberta Schwartz, Houston Methodist's chief innovation officer, described the results of Apella's image-based platform in its 15-room cardiothoracic suite to Healthcare IT News on Oct. 9. In the hospital's study, published June 27 in the Journal of Imaging, volume rose about 25 cases a month over a baseline of about 339 against a synthetic control built from 11 sites without the system, while case mix and operative minutes held steady. A placebo permutation test of that gain did not reach conventional significance, and a 2.7-minute drop in median late-start minutes was exploratory, the authors wrote. The study disclosed that Houston Methodist holds a minority equity interest in Apella and that an Apella employee ran the original analyses, which the hospital's team replicated.Sources: Journal of Imaging, Healthcare IT News
- FYIEvidence Oct 9, 2026A Northwell Health team built task-level skill scores from 19 kinematic measures recorded by the surgical robot across 392 inguinal hernia repairs by six attendings and 41 trainees, according to the npj Digital Medicine paper published Oct. 9. Cross-validated AUCs for telling attendings from trainees ranged from 0.62 to 0.74 by task, scores rose steadily from second-year residents to attendings, and the overall AUC in an external set of 21 cases was 0.84. The authors said the framework allows objective skills assessment without subjective rating scales. One co-author is an employee of Intuitive Surgical, which makes the robot used in the study.Source: npj Digital Medicine