Explore how AI is revolutionizing healthcare by enhancing diagnostics, predicting diseases, and supporting medical professionals in various capacities.
The integration of AI in colonoscopy improves adenoma detection but raises concerns about declining physician skills when AI support is withdrawn, highlighting a potential clinical 'de-skilling' effect.
Linea integrates generative AI and high-touch virtual nursing to proactively manage heart failure patients, achieving a 50% reduction in hospital readmissions within 90 days, substantially lowering costs and improving patient outcomes.
A comprehensive survey of 233 US healthcare systems reveals widespread AI deployment with key barriers including cost and resource limitations; highlights the critical role of CFOs and CIOs in AI decision-making and the growing influence of integrated EHR systems.
An observational study evaluates an AI chatbot's usability in promoting HIV testing among men who have sex with men in Malaysia, showcasing potential improvements for public health interventions.
This study explores the integration of line-field confocal optical coherence tomography (LC-OCT) with artificial intelligence (AI) to enhance wound healing evaluations, providing deeper insights into clinical interventions.
A 60-year-old man developed bromide toxicity after replacing dietary NaCl with NaBr based on ChatGPT guidance; the case highlights laboratory interference, diagnostic pitfalls, clinical management, and risks of decontextualized AI health advice.
A multisite RCT of a 12‑module smartphone CBT intervention for psychiatric inpatients did not change time to first suicide attempt overall, but sustained reductions in suicidal ideation and reduced recurrent attempts among patients with prior attempts in secondary analyses.
A single-blind RCT shows that integrating ChatGPT into nursing education significantly improves students’ problem-solving skills, AI attitudes, nursing process competency, and satisfaction compared to standard training.
A machine learning model using nationwide registry data accurately predicts long-term remission in refractory ulcerative colitis and identifies key prognostic factors such as pseudopolyps.
A proof-of-concept study demonstrates that AI-based image recognition, using the ResNet-18 algorithm, can accurately identify clinically important Aspergillus species, offering a rapid, cost-effective alternative to traditional methods.
An AI-based model for assessing ankylosing spondylitis severity using spinal X-rays and mSASSS demonstrates high accuracy, offering clinicians a rapid, reliable tool and reducing human resource demands.
AI analysis of smile videos demonstrates high accuracy in detecting Parkinson’s disease, offering a scalable and accessible screening solution with strong generalizability across populations.
A pioneering study demonstrates that visual cues of infection in virtual reality can trigger real immune activation, revealing a rapid neural-immune interface with potential implications for medicine, psychology, and vaccine development.
AI offers transformative potential for clinical trials, but realizing its value demands a robust, stakeholder-driven framework to guide adoption and assess impact across the trial lifecycle.
This study assesses ChatGPT-4's effectiveness as a virtual assistant in managing puerperal mastitis, focusing on accuracy, clarity, and safety in patient communication.
The VIDEOGO trial shows that online video consultations are non-inferior to face-to-face consultations for preoperative assessment in major abdominal surgery, achieving similar patient satisfaction and information recall.
Two randomized controlled trials evaluated immersive VR therapies targeting auditory verbal hallucinations and persecutory delusions in schizophrenia and psychosis, revealing modest benefits for hallucinations but no superior effect for cognitive VR therapy in delusions.
Limited demographic data transparency in FDA-authorized AI devices for Alzheimer’s may hinder equitable care and generalizability, highlighting a crucial need for improved dataset representativeness and regulatory oversight.