In an analysis of more than 32,000 CT pulmonary angiograms, the AI triage tool matched radiologist interpretations in 97.8% of cases, while discordant findings highlighted the continued importance of radiologist oversight.
Stanford researchers have developed MUSK, an AI model that integrates visual and language data to improve cancer prognosis and treatment guidance, outperforming conventional methods in accuracy and versatility.
RSNA's 2024 AI challenge used the largest open-source lumbar spine MRI dataset to develop AI models for detecting and classifying degenerative spine conditions, with participation from 1,874 teams worldwide.
ChatGPT-4 Vision shows strong performance on text-based radiology questions but struggles with image interpretation, highlighting limitations in its current application to radiology.
Read More
A study from Japan’s Osaka Metropolitan University found that while ChatGPT shows potential as a diagnostic tool in radiology, its accuracy falls short compared to board-certified radiologists, especially in musculoskeletal cases.
Read More
Radiologist Lisa C. Adams, MD, contributed to a study highlighting Meta Llama 3’s strong performance in radiology exams, showcasing the potential of open-source AI in healthcare. She shares her insights here.
Read More
A study published in Radiology found that an off-label AI tool was effective in autonomously excluding pathology in chest X-rays with equal or lower rates of critical misses than radiologists, but AI mistakes were generally more clinically severe, suggesting the need for careful implementation.
Read More
A new deep learning model detects clinically significant prostate cancer on MRI as accurately as experienced abdominal radiologists, showing potential as an adjunct tool to improve cancer detection rates.
Read More