With the help of artificial intelligence, computers are to simulate human thought processes. Machine learning is intended to support almost all medical specialties. But what is going on inside an AI algorithm, what are its decisions based on? Can you even entrust a medical diagnosis to a machine? Clarifying these questions remains a central aspect of AI research and development.
Henkjan Huisman has been appointed Professor of AI Guided Imaging at Radboud UMC/Radboud University. With a focus on AI applications in medical imaging, his goal is to provide better and more affordable care.
Mammography image interpretation AI models are unreliable – but so are human readers. A new hybrid strategy could reduce radiologist workload by 38% without compromising diagnostic efficacy.
To help people with hearing loss, researchers develop an AI-powered lip-reading camera that translates lip movements into sound. The device could also work in areas with a lot of background noise.
Gadolinium-based contrast agents enhance visibility but also pose a significant health risk. A new AI-powered virtual MRI imaging technique is designed to offer a safer diagnostic approach.
More clinically relevant tumors detected at an earlier stage and at lower costs: New research finds that AI can replace the second radiologist in the Dutch breast cancer screening program.
AI tools for adenoma polyp detection are designed to help increase the accuracy of colonoscopies. However, a new study finds that using AI may lead to loss of detection skills in health professionals.
New research finds that neurosurgery students receiving AI-augmented, personalized feedback from a human instructor have better surgical performance, risk management and skill transfer.
Microsatellite instability-high (MSI-H) tumors are associated with better clinical outcomes. A novel AI model for accurate MSI prediction could help battle gastric and colorectal cancers.
A new AI tool has been shown to accurately predict who would go on to develop leaks in the heart valves – conditions known as regurgitant valvular heart diseases, at a very early stage from an ECG.
By identifying the unique biological signatures that guide AI tissue classification, new research lays the foundation for smart surgical tools with the added feature of biomolecular-level precision.
RadNet's DeepHealth unit has completed its acquisition of iCAD, adding established AI breast cancer detection capabilities to its diagnostic platform.
A new AI-based tool measures cancer aggressiveness by analyzing the ‘stemness’ of tumors – their similarity to pluripotent stem cells. This could pave the way for new therapies.
By directing radiologists' attention to potentially suspicious areas, AI can help spot more lesions indicative of breast cancer in mammograms. This is suggested by a new study.
At Medical Taiwan 2025, manufacturers attracted attendees with a wide range of innovations. We took a closer look at some of the most exciting companies and their products on display at the medical, health and care expo.
A neural network AI has been trained to detect cardiac amyloidosis from a single echocardiogram video of the heart's apical four-chamber view and differentiate it from similar heart conditions.
Medical Taiwan has long been a showcase for cutting-edge healthcare solutions, but this year marked a pivotal moment. Visitors of the latest edition of the medical, health and care expo in Taipei witnessed a particularly noticeable step forward: the definitive transition of medical AI from promising research to actual clinical practice. Organized by the Taiwan External Trade Development Council…
A new study shows that retrieval-augmented generation (RAG) can eliminate hallucinations in clinical large language models (LLMs) while protecting patient privacy during contrast media consultations.
More consistent TIL assessments, more accurate patients' prognoses: New research shows how AI sharpens pathologists' interpretation of tissue samples for malignant melanoma.
A new technique called photoacoustic computed tomography (PACT) offers a breast imaging alternative without the discomfort, high costs, or risk associated with the conventional evaluation methods.
