artificial intelligence
68 all articles · 20 / page
Latest news
Tag
The Sieve of Asclepius: A History of Navigating the Medical Literature, From Index to Algorithm
A historically grounded review of how physicians, librarians, publishers, and now algorithms have filtered the expanding medical literature, shaping what counts as visible, credible, and clinically actionable knowledge.
AI Model Shows Moderate Accuracy for Predicting Complete Response After Total Neoadjuvant Therapy in Locally Advanced Rectal Cancer
A fine-tuned ensemble of tabular foundation models moderately predicted pathological complete response after TNT and showed potential to support watch-and-wait selection after recalibration.
A Chest X-Ray AI Age Gap May Flag Early Spirometric Decline Before Overt Lung Disease
In a large Korean health-screening cohort, chest radiograph-derived age acceleration was associated with both prevalent and incident PRISm and obstructive lung disease, supporting chest X-ray AI as a potential opportunistic marker of early
AI-Driven Insights into Incidental Thyroid Findings: Prevalence, Clinical Impact, and the Overdiagnosis Dilemma
A retrospective cohort study using AI to analyze radiology reports reveals high prevalence of incidental thyroid findings, strong association with diagnostic cascades, and potential overdiagnosis of low-risk thyroid cancers.
AI vs. Human Clinicians: Study Reveals Gaps in AI-Generated Clinical Notes
A cross-sectional evaluation of AI-generated clinical notes found significant quality deficits compared to human-produced notes, particularly in thoroughness, organization, and usefulness.
AI-Enhanced ECG Screening Pinpoints Atrial Fibrillation Risk in Older Adults: Key Insights from VITAL-AF Trial
A VITAL-AF trial analysis reveals that combining ECG-based artificial intelligence with clinical risk scores significantly improves atrial fibrillation screening efficiency. The study demonstrates that high-risk individuals benefit most fro
Beyond Simple Quantification: How AI and Tumor-Stroma Ratio Are Redefining Prognosis Prediction in Hepatocellular Carcinoma
Researchers have developed an AI-powered framework that uncovers a non-linear relationship between tumor-stroma ratio and mortality in hepatocellular carcinoma, using a novel token-guided multimodal fusion approach to improve prognostic acc
Adaptive AI Transforms Cardiovascular Event Adjudication: New Algorithm Achieves Near-Human Accuracy Across Multiple Endpoints
Researchers developed and validated ADAPT-CEC, an artificial intelligence algorithm that can adjudicate cardiovascular events across different trial definitions. A hybrid approach combining AI with selective human review achieved 95.6% accu
Artificial Intelligence in Gastric Endoscopy: Decoding the Findings of a Landmark Multicenter Randomized Controlled Trial
A large-scale multicenter RCT investigating AI-assisted gastric neoplasm detection found no significant improvement in pathologically confirmed detection rates but demonstrated significant reductions in procedural blind spots and benefits f
Precision Prediction of Incident Heart Failure: The Clinical Integration of AI-Enabled Electrocardiography
This review evaluates the breakthrough ECG2HF model, a publicly available AI tool that outperforms standard clinical scores in predicting 10-year heart failure risk using 12-lead ECG waveforms.
Breaking the Hardware Barrier: How Domain-Shift AI Enables Vendor-Agnostic 3D OCT Macular Disease Detection
A multicenter study in JAMA Ophthalmology introduces a deep learning model using domain-shift technology to accurately detect macular diseases across different OCT hardware vendors, achieving high negative predictive value and establishing
AI-Supported Mammography Outperforms Standard Double Reading: Insights from the MASAI Trial
The MASAI trial demonstrates that AI-supported mammography screening achieves higher sensitivity and non-inferior interval cancer rates compared to standard double reading, while maintaining specificity and significantly reducing radiologis
AI-Enabled Stethoscopes in Primary Care: Why Implementation Science Matters More Than Algorithm Accuracy
The TRICORDER trial demonstrates that while AI stethoscopes are capable of detecting heart failure, atrial fibrillation, and valvular disease, their real-world implementation in UK primary care did not yield a statistically significant incr
Human-AI Collaboration Enhances Clinical Reasoning in Ophthalmology but Risks Overconfidence and Automation Bias
A crossover study involving 30 ophthalmologists and trainees reveals that while Claude-3.5-Sonnet collaboration improves diagnostic accuracy in complex cases, it also increases confidence in incorrect decisions and fails to reach the perfor
AI Assistant Achieves High Accuracy in Molecular-Based CNS Tumor Classification: A Multi-Institutional Validation
This multi-institutional study validates Neuropath-AI, a deep-learning system that predicts CNS tumor types by inferring molecular features from histopathology slides. Achieving 80% top-1 accuracy in high-confidence samples, the model offer
AI-Driven Detection of Extranodal Extension Outperforms Radiologists in Predicting Outcomes for HPV-Positive Oropharyngeal Cancer
A single-center study demonstrates that an AI pipeline for CT-based lymph node segmentation and extranodal extension classification significantly improves prognostic accuracy for HPV-positive oropharyngeal cancer compared to traditional rad
High Empathy, Low Accuracy: Can Generative AI Safely Navigate Alcohol Misuse Support?
A longitudinal study published in NEJM AI evaluates seven generative AI chatbots for alcohol misuse support, finding that while they excel in empathy and non-judgmental language, they frequently provide low-quality or inaccurate medical inf
AI-ECG Models for Heart Failure Screening Show High Accuracy in First-of-its-Kind Independent Comparison Study
This independent validation study confirms that AI-enhanced ECG models effectively detect left ventricular systolic dysfunction across diverse populations, achieving AUROCs up to 0.93. However, significant barriers remain regarding model tr
AI-Driven Decision Support for Antibiotic Switching: Why Clinicians Value Caution Over Speed
A randomized multimethod study reveals that while AI clinical decision support systems for antibiotic switching are well-received, their primary influence is reinforcing conservative prescribing. The study underscores that usability and cli
Multimodal AI Outperforms Clinical Nomograms in Predicting Metastasis for Post-Prostatectomy Biochemical Recurrence
A validated multimodal AI model using digital pathology and clinical data significantly improves risk stratification for prostate cancer patients with biochemical recurrence, identifying those who derive the greatest benefit from salvage ho