Separating the art of medicine from artificial intelligence
February 21, 2018 -- Is it feasible to develop an artificial intelligence algorithm that can actually "read" chest x-rays? The task is definitely not an easy one and major challenges must be addressed, writes radiologist Dr. Hugh Harvey in his new column.  Discuss
Should radiologists become informatics professionals?
February 9, 2018 -- Nearly every medical discipline depends on imaging informatics, but training and accreditation in this emerging field are still very much in their infancy. This interview article looks at the issues under discussion and the range of courses, qualifications, and diplomas being developed.  Discuss
Machine learning can help assess atherosclerosis
February 7, 2018 -- Machine-learning techniques analyze imaging measurements to automatically stratify patients by the level of atherosclerotic burden, offering the potential of personalized prediction of disease progression and more effective treatment for individual patients, according to researchers from Italy.  Discuss
AI powers liver volumetry on contrast-enhanced MRI
January 18, 2018 -- Artificial intelligence (AI) can accurately segment the liver on contrast-enhanced MRI examinations, potentially yielding significant time savings for radiologists without affecting the accuracy of volumetry results, according to researchers from the University Hospital Regensburg in Germany.  Discuss
Can artificial intelligence enable a 10-minute MRI scan?
January 17, 2018 -- Artificial intelligence (AI) algorithms promise to improve the imaging workflow, leading to significantly faster MR image acquisition times as well as vastly improved image quality in low-dose CT images. Consultant radiologist Dr. Hugh Harvey from London investigates in a new column.  Discuss
PACS with AI can replace PACS without AI
January 8, 2018 -- Artificial intelligence (AI) can help solve the huge reporting backlog, but this requires industry's cooperation, writes Dr. Neelam Dugar. PACS with AI will replace PACS without AI in the next five to 10 years, so don't get left behind, she urges.  Discuss
What's next for deep learning? Dutch show the way
January 3, 2018 -- A multidisciplinary team from the Netherlands collected a top award at RSNA 2017 for identifying four key areas of future development for deep learning: multitask learning, image-to-image synthesis, unsupervised learning, and uncertainty estimation and novelty detection.  Discuss
AI algorithm matches radiologists in breast screening exams
December 14, 2017 -- How does an artificial intelligence (AI) algorithm compare to radiologists when it comes to interpreting screening mammography exams? Their performances were fairly comparable, according to research presented at the recent RSNA 2017 meeting in Chicago.  Discuss
AI assesses cardiovascular risk on routine chest CT
December 12, 2017 -- An artificial intelligence (AI) algorithm can automatically quantify coronary and other arterial calcification from routine chest CT studies, enabling cardiovascular risk assessment to be performed on these patients, Dutch researchers reported at the RSNA 2017 meeting.  Discuss
German innovation supports structured data mining
December 8, 2017 -- Structured reporting offers the prospect of speedier reporting throughput, but it also has major implications for epidemiology due to its capacity to store easily accessible data, notably numerical measurements, according to a German expert.  Discuss