AI Learns When It Is Unsure: New Model Flags Borderline Hip Screenings Like Clinicians Do
Oxford researchers built an AI method that estimates uncertainty in landmark-based hip measurements and converts it into a screening confidence ...
Tag: Clinical decision support
Blinded experts flag unsafe actions in synthetic outputs from local clinical LLMs
A blinded panel of eight multidisciplinary experts judged the primary action acceptable in 338 of 400 synthetic clinical outputs from ...
Clinicians in the Cardiothoracic ICU Say AI Must Earn Its Place Before They Will Trust It
A field study of fourteen cardiothoracic ICU clinicians found that predictive AI tools must demonstrate added clinical value, earn trust ...
AI Moves Genetic Variant Interpretation Beyond Simple Pathogenicity Scores
A review in Cell & Bioscience argues that AI-assisted genetic variant interpretation should move beyond isolated pathogenicity scoring toward evidence-centered ...
AI Risk Scoring Cuts Unnecessary Kidney Surgery in Large Randomized Trial
A CT-based AI system reduced benign interventions among treated kidney lesions from 40.3 percent to 16.4 percent in a randomized ...
Perioperative AI needs to put humans first, anesthesiologists argue
Anesthesiology researchers argue that AI in perioperative care must be human-centered, warning that most models lack prospective validation and real-world ...
AI Model Designed to Predict Antiretroviral Resistance in HIV Patients in Cameroon
Researchers in Cameroon built and tested a machine learning model that predicts antiretroviral resistance from routine clinical data, with XGBoost ...
Deep learning model fuses clinical, imaging and genomic data to stratify coronary artery disease risk
A proof-of-concept deep learning framework called CardioHolisticAI combines clinical, radiomic and genomic data to stratify coronary artery disease risk, achieving ...
AI agent reads emergency abdominal ultrasounds with 99.6% diagnostic accuracy in study
A vision-language AI agent built by CU Anschutz researchers identified correct diagnoses in 99.6% of reviewed right upper quadrant ultrasound ...
Large Language Models Show Promise and Limits in Hospital Infection Control
A systematic review of fourteen studies finds large language models perform well on structured infection control tasks but fall short ...
AI System Splits TCM Knowledge From Expert Experience to Aid Prescribing
A modular AI system separates general Traditional Chinese Medicine knowledge from one senior practitioner's prescribing experience, scoring highest among evaluated ...
Artificial intelligence could reshape antimicrobial stewardship in Africa, but evidence remains early
A new commentary argues that artificial intelligence could strengthen antimicrobial stewardship across Africa, but finds most applications remain at pilot ...
Mixed-effects radiomics model hits AUC 0.91 for lung cancer prediction across seven UK hospitals
A retrospective UK study of 1,026 patients found that a mixed-effects radiomics model achieved an AUC of 0.91 for predicting ...
Generative AI in Cancer Care: Mapping Review Finds Rapid Growth, Weak Evidence
A mapping review of 218 studies finds generative AI tools in cancer care are expanding rapidly but are validated mostly ...
AI Chatbots Show an Accuracy-Practicality Trade-off in Anesthesia Crisis Scenarios
A benchmark study of OpenAI-o1 and DeepSeek-R1 across 30 anesthesia crisis scenarios found that the most accurate model gave the ...
Pediatricians rank top strategies for screening children for type 1 diabetes in primary care
A pre-implementation study with 29 primary care pediatricians identified highly prioritized strategies, including antibody order sets, protocols, and electronic health ...
AI matches expert severity ratings in half of hospital adverse event reports
A single-center study found a large language model exactly matched adjudicated expert severity classifications in 49 percent of 286 hospital ...
Most pediatricians in Türkiye expect AI to improve care, survey finds
A web-based survey of 368 pediatricians in Türkiye found broad optimism about artificial intelligence in pediatric care alongside strong concerns ...
Scoping review maps generative AI in healthcare and finds a carbon blind spot
A scoping review of 24 studies finds generative AI models achieving up to 99.9 percent accuracy in healthcare tasks, but ...
AI in MS, NMOSD, autoimmune encephalitis and MOGAD: promise still outruns proof
A new review finds AI applications in CNS autoimmune diseases are expanding rapidly, yet no model has been prospectively validated ...
POPULAR NEWS
-
Study Uncovers Botanical Compounds and Mechanisms Targeting Chronic Bronchitis Inflammation
-
Three Millimolar Caffeine Enhances Cryopreserved Nguni Bull Semen Post-Thaw Motility
-
Higher Brain Iron Levels Are Associated With MRI Alterations in Older Adults
-
Parent B Candida orthopsilosis caused fatal breakthrough candidemia during caspofungin therapy