Advances in AI — October 2026
The item summarizes the top AI-related news and research in radiology over the past month.
Read the original →Article digest · Oct 27 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.
35 stories18 with summaries9 topics
🩻 Diagnosis and imaging5
AI detects “invisible” signs of pancreatic cancer years before diagnosis
A new AI system called REDMOD identified hidden warning signs an average of 475 days before diagnosis. It was nearly twice as sensitive as experienced radiologists. It is intended to help doctors detect pancreatic cancer at its earliest stages.
Read the original →AI Model Uses Routine Pathology Slides to Predict Breast Cancer Recurrence Risk
Agendia presented data on an AI pathology model. The model predicts breast cancer recurrence risk from routine biopsy slides.
Read the original →- Headlines only (the publisher did not provide a description)
- Agendia Unveils Clinical Validation of AI-Driven Digital Pathology ApproachBioSpace · 10/26 15:00 UTC
- Korean AI for Alzheimer's diagnosis, artery analysis bag 510(k) and more briefsMobiHealthNews · 10/26 07:47 UTC
🩺 Clinical decision support5
Without an Exit Plan, AI Partnerships in Africa Risk Becoming the New Tied Aid
Rachel George writes that AI partnerships risk becoming a form of tied aid. She argues they could bind recipients to the priorities, standards, and suppliers of tech donors.
Read the original →Machine learning model predicts 28-day mortality in sepsis patients with diabetic retinopathy
A machine learning model was trained on ICU data from two databases. It predicts 28-day mortality in sepsis patients with diabetic retinopathy, with AUCs of 0.805 to 0.870 across validation sets.
Read the original →- Headlines only (the publisher did not provide a description)
- NVIDIA Invests in Doctolib to Speed Healthcare AIKonsulteer · 10/26 11:31 UTC
- Over 1,000 FDA-Cleared AI Devices: Where They Actually Work in HospitalsIntelligent Living · 10/26 00:29 UTC
- India’s healthcare reaps AI boom dividendsMedical Buyer · 10/26 08:19 UTC
📝 Generative AI, documentation and records4
Mulk Technology launches AI-powered digital hospital platform for India
Mulk Technology announced an AI-enabled digital healthcare platform for India and international markets. It follows a strategic partnership with OnClarity USA signed in Dubai on 23 October 2026. The platform combines Mulk Technology’s Hospital Management System with OnClarity USA’s AI-based patient experience tools.
Read the original →- Headlines only (the publisher did not provide a description)
- eClinicalWorks offers new AI-enabled EHR toolsHealthcare IT News · 10/26 15:30 UTC
- Integrating AI Scribes to Improve Patient CommunicationThe Cardiology Advisor · 10/26 04:00 UTC
- Artificial Intelligence Streamlines Clinical Workflows and Patient Care for Family PhysiciansThe Cardiology Advisor · 10/26 04:00 UTC
🏥 Hospital operations and care delivery4
A hospital employee uploaded patient data to an unapproved AI tool. Where is it now?
A Baltimore hospital worker uploaded 3,182 patients’ records to an unapproved AI tool. Deleting the chat may not have erased them.
Read the original →Governing Healthcare AI Means Putting Patient Safety First
At the 2026 WEDI National Conference, HSCC task group co-leads Bill Reid and Ed Gaudet outlined a practical approach to AI governance. The approach addresses cybersecurity and clinical risk.
Read the original →- Headlines only (the publisher did not provide a description)
- Michigan Medicine’s new AI chief joins a fast-growing club of hospital executivesCrain's Detroit Business · 10/26 09:45 UTC
- AI Task Force Develops Unified Strategy to Integrate Tech InnovationsCleveland Clinic · 10/26 08:44 UTC
💊 Drug discovery and clinical trials5
Cleveland Clinic and Dyania’s Work on AI in Clinical Trial Matching Earns 2026 Gartner Eye on Innovation Award
Cleveland Clinic is recognized in the Healthcare and Life Sciences category. The recognition is for its work with Dyania Health's Synapsis Artificial Intelligence (AI) Platform.
Read the original →AI is reshaping healthcare economics – from drug discovery to patient care
Andy Craggs, head of energy finance at Paragon SME Lending, writes that healthcare is emerging as one of the biggest beneficiaries of AI.
Read the original →A Drug Has to Answer 5 Questions, AI Made Only One Cheap. Who Pays for the Rest?
This is a monthly column on biotech for founders and builders in Central and Eastern Europe. The first installment asks why cheaper design has not led to cheaper drugs.
Read the original →- Headlines only (the publisher did not provide a description)
- MitoCareX Bio completes first AI drug-discovery screening stageStreetInsider · 10/26 12:18 UTC
- Florence automating the connection between study sites and sponsors, with AI tools already liveChinook Observer · 10/26 13:50 UTC
⚖️ Device approvals and regulation3
How AI-Enabled Devices Are Changing Pharmacy
Pharmacists can pair advanced medical devices such as digital blood pressure monitors with emerging AI tools to enhance routine counseling with personalized patient education. Combining intelligent technology with direct pharmacist guidance is described as giving patients deeper health insights and supporting better self-management between clinical visits.
Read the original →Heartflow Takes to The Courts Over AI CCTA Tech
Heartflow, the incumbent in AI plaque analysis and FFR-CT, has again turned to legal action to defend its cardiac CT software market share. It is suing Elucid for patent infringement and trade secret theft.
Read the original →- Headlines only (the publisher did not provide a description)
- The FDA Cleared Over 1,500 Medical AI Devices and Next-Generation Obesity Pills, but the Difference Between Clearance and Real Clinical Proof Is StaggeringIstituto Impariamo · 10/26 01:41 UTC
⚠️ Safety, errors and bias3
AI in healthcare: No matter how good it is at detecting lesions, it still needs verification by doctors.
AI helps detect difficult lesions, aids in diagnosis and drug development. Algorithmic errors can be dangerous if doctors become overly reliant on the technology.
Read the original →- Headlines only (the publisher did not provide a description)
- Clinical AI Should Earn Its Autonomy the Way Doctors DoPetrie-Flom Center · 10/26 13:08 UTC
- The Algorithmic Prescription Pad: What AI Prescribing Means for Public Healthorfonline.org · 10/26 12:58 UTC
🔬 Research results (major journals)4
LLMs Recommend Far More Care Than Doctors
The piece describes a study as serendipitous evidence for the case on AI alignment.
Read the original →How Autonomous AI Is Reshaping Biopharma Operations
Agentic AI is moving biopharma from tools that inform decisions to systems that execute them. It is reshaping discovery, clinical trials, commercial launch and regulatory work.
Read the original →- Headlines only (the publisher did not provide a description)
- AI chatbots may worsen symptoms in people with psychosisMedical Xpress · 10/26 18:00 UTC
- adsilico raises £10M to test medical devices on AI-generated virtual patientsDealroom · 10/26 08:58 UTC
🇯🇵 Medicine and AI in Japan2
Precision cuts disease names suspected of being hallucinated by about 80% in medical LLM trained on case reports
Precision, a medical AI startup in Bunkyo, Tokyo, announced on October 26 that additional training cut the share of suspected hallucinated disease names by about 80% in its medical-specific large language model SIP-jmed-LLM-Precision. The company will also sell a Japanese input software product using this research, "Kyo no Iryo AI Nihongo Nyuryoku (Precision IME-med)", from October.
Read the original →AI Business Navi launched to diagnose first steps in generative AI use for small businesses
"DXCruise", jointly operated by ASTRO Siblings and Shijuman-do, began offering the generative AI diagnostic service "AI Business Navi" for small and medium-sized companies on October 26. Fees are quoted individually, and the diagnosis content and cost are explained before a request is made.
Read the original →The analysis and visual summary are in the Oct 27 AM issue. Japanese version here.