FDA Clears AI-Enabled Chest CT Software for Aortic Analysis
The newly cleared Aview Aorta Metrics software reportedly offers automated segmentation of aortic regions. It also classifies aortic zones based on chest CT images.
Read the original →Article digest · Oct 28 AM
AI in Medicine — what each story says
Every story in this issue with a short summary, so you can decide what to read before leaving the site. Summaries are written by AI (Claude) only from each publisher's own description and checked by Jev for anything not in that description. They are not full translations. Follow "Read the original" for the full story.
38 stories21 with summaries9 topics
🩻 Diagnosis and imaging4
Eulji University Medical Center Blends AI with 70-Year Clinical Expertise
A symposium addresses AI in diagnosis, treatment, surgery and patient care. It uses 70 years of clinical data.
Read the original →Federated AI Framework Pairs Medical Imaging Accuracy with Auditable Explanations
A new federated learning framework for chest X-ray analysis reports modest accuracy gains over existing methods. It aligns explanations across simulated hospital clients under a certified approach.
Read the original →- Headlines only (the publisher did not provide a description)
- Powerful Medical Granted CPT Code for AI ECG Heart Attack Detection, Weeks After FDA De Novo Authorization of Queen of HeartsVentureBeat · 10/27 13:00 UTC
🩺 Clinical decision support5
Machine Learning May Help Predict Diabetes in NAFLD
A two-cohort study evaluates machine learning for predicting medium-term Type 2 diabetes risk in patients with biopsy-confirmed NAFLD.
Read the original →- Headlines only (the publisher did not provide a description)
- AI Is Already Inside the Doctor’s Visit. The Question Is Whether It’s Helping or Hunting for Billable Findings.The Clinical Trial Vanguard · 10/27 13:08 UTC
- Vera Health has secured $28.5M for clinical AI search platformMobiHealthNews · 10/27 17:36 UTC
- Why AI legal liability should be a red flag for health systemsModern Healthcare News · 10/27 10:00 UTC
- SGH AI cuts brain bleed reporting timeHealthcare IT News · 10/27 11:06 UTC
📝 Generative AI, documentation and records2
AI Drafts With Doctor Oversight Showed Quality Gains in Pilot Trial
A pilot randomized trial at Kyoto University Hospital tested a workflow in which clinicians review drafts generated by large language models. It found that the workflow improved most expert-rated quality domains for discharge summaries.
Read the original →- Headlines only (the publisher did not provide a description)
- AI Scribes Improved Note-Taking Time but Not Productivity, Study FindsMedPage Today · 10/26 21:00 UTC
🏥 Hospital operations and care delivery5
How To Measure Healthcare AI ROI
The payoff from healthcare artificial intelligence can take time. Three experts share how leaders can demonstrate value from AI and use it to support the health system's strategic priorities.
Read the original →When AI Enters the Hospital: What Changes for Doctors and Patients
This column follows an earlier piece that examined medical AI from a business perspective, focusing on opportunities for technology companies. It turns to a more fundamental question beyond innovation and new markets.
Read the original →Scaling Agentic & Physical AI in Hospitals: Roadmap
The article presents an operational roadmap for scaling agentic AI and physical hospital robotics beyond pilots. It covers ROI, governance and adoption strategy.
Read the original →- Headlines only (the publisher did not provide a description)
- UT Austin unveils UT Dell Medicine as it bets big on new AI-native hospitalFierce Healthcare · 10/27 14:30 UTC
- AI is a megaphone for bad dataHospitalogy · 10/27 17:17 UTC
💊 Drug discovery and clinical trials4
AI drug discovery is changing which experiments reach patients
Artificial intelligence is influencing the choices behind drug development, from molecular design to trial recruitment.
Read the original →- Headlines only (the publisher did not provide a description)
- Parallel Bio Appoints Jason Lew Chief AI Officer to Advance Human-First Drug Discoverycitybiz · 10/27 14:14 UTC
- AI Meets the LabJones Day · 10/26 23:02 UTC
- Anthropic lays out drug R&D ambitions, risks in mammoth IPO filingFierce Biotech · 10/26 15:44 UTC
⚖️ Device approvals and regulation5
Why Agentic AI Is Not Yet FDA Approved
The risks and behavior of autonomous, agentic AI in health care remain relatively uncharted, so regulators such as the FDA are taking a cautious approach to clearance. The FDA is soliciting industry feedback on evaluation frameworks. The article says rigorous, standardized oversight is essential for patient safety and clinical reliability before these tools enter real-world practice.
Read the original →AI medical device regulation after approval: What NHS trusts should set up now
The article describes a hospital trust eighteen months into using an AI tool that flags suspected abnormalities on chest X-rays, with no formal check on whether it still performs as it did on day one. It says UK regulation of AI medical devices is now turning toward that question.
Read the original →- Headlines only (the publisher did not provide a description)
- AI Adoption in Clinical Trials Slowed by Regulatory Gaps, Not Technology LimitsThe Clinical Trial Vanguard · 10/27 07:53 UTC
- Stanford Scholars Respond to FDA on Regulating Gen-AI Medical DevicesFreeman Spogli Institute for International Studies · 10/26 12:00 UTC
- Heartflow sues AI rival Elucid over alleged IP violationsCardiovascular Business · 10/26 19:38 UTC
⚠️ Safety, errors and bias5
The Liability Battle Over Medical AI Is Coming
As medical AI advances, lawmakers and health leaders face the question of who is liable when the technology fails.
Read the original →Medical AI Models That Read Images and Write Reports Still Face a Trust Gap
A new narrative review finds that medical vision-language AI models show strong benchmark results. It also reports inconsistent evaluation reporting, hallucination risks and narrow training data.
Read the original →- Headlines only (the publisher did not provide a description)
- Telekom invests in Tandem Health: AI-powered medical assistant eases the burden on healthcare professionalsDeutsche Telekom · 10/27 10:17 UTC
- UMich and European professors discuss AI in healthcareThe Michigan Daily · 10/27 06:05 UTC
- AI Medical Alerts: 90% Ignored at Vanderbilt [2026]shattered.io · 10/27 02:05 UTC
🔬 Research results (major journals)4
Years earlier, a powerful artificial intelligence (AI) system was developed that could detect the in..
Years earlier, a powerful AI system was developed that could detect pancreatic cancer, described as the "silent killer," with the naked eye. Academia is paying keen attention.
Read the original →AI Predicts Death List from Single Face Analysis: 50% of Subjects Passed Away Within 90 Days
The technology is described as helping medical professionals identify patients who need prioritized urgent care. It is also presented as giving employers a data-driven way to screen for high-risk individuals with potential health conditions. The description says it uses health assessment algorithms to support care triage and workplace wellness initiatives.
Read the original →Medical AI Systems Outperform Doctors in Clinical Accuracy as Microsoft and Google Validate MAI-DxO and AMIE
Global big tech companies are releasing a series of research findings evaluating the clinical performance of medical AI.
Read the original →Robotic surgery and AI in urology: how they are transforming care, but with the conundrum of costs
New techniques will be the focus of the Italian Society's forthcoming Congress. The description says a number of unknowns remain between the promises and their large-scale implementation.
Read the original →🇯🇵 Medicine and AI in Japan4
What Changes When AI Is Introduced in Hospitals: Effects on Doctors and Patients
This column follows an earlier one that covered medical AI from a business perspective and the opportunities for technology companies. It asks what these technologies change for the doctors who use them and the patients who rely on those doctors. The author says the value of medical innovation lies in improving the quality and accessibility of care, and that remarkable changes in how care is delivered are already under way.
Read the original →AI Used to Predict Future Risks for Diabetes Patients, Fukushima Medical University and Japan Diabetes Society Among Others
Researchers at Fukushima Medical University, Chiba University and the National Institute of Health and Crisis Management (JIHS), working with Fcom Corporation, developed a web platform called "Fukushima Medical University Diabetes Future Prediction Navi". It uses AI to classify diabetes into five subtypes and estimate future risks of complications and comorbidities, including diabetic kidney disease, dialysis initiation and cardiovascular disorders.
Read the original →South Korea Develops AI Robots as Surgical Assistants at Samsung-Affiliated Hospital
Samsung Seoul Hospital is working to put humanoid robots for surgery into practical use. Generative AI predicts the situation and the robot acts accordingly. When a surgeon gives a spoken command, the robot moves its hands within two seconds, handing over instruments and returning them to their original position after use.
Read the original →Tairex Announces MHAC, With 54 AI Doctors Reviewing Cases Alongside Physicians
Tairex, which originated at Tsinghua University, announced MHAC, in which 54 AI doctors join clinical care alongside physicians. The figures of over 96% on MedQA and roughly 30% efficiency improvement are self-reported by the company.
Read the original →The analysis and visual summary are in the Oct 28 AM issue. Japanese version here.