In a recent editorial published in the November issue of The New England Journal of Medicine (NEJM), Werner Hacke, MD, PhD, DSc, discussed a new era of imaging selections for patients who are looking to benefit from a thrombectomy performed long after the onset of a stroke.

Athletes suffering suspected tears of the anterior cruciate ligament (ACL) are often adequately assessed with clinical diagnostic tests performed in the clinician’s office. When these are inconclusive, diagnostic arthroscopy is the gold standard—and MRI is a generally low-value option due to its time and cost burdens.

According to statistics, roughly two billion people worldwide are social media users, with that number expected to double by 2018. So why might there be little attention on social media in health communications?  

Patients presenting in the emergency department with pronounced chest pain but no other signs of acute coronary syndromes don’t have better outcomes when they’re sent for stress testing or coronary CT angiography (CCTA) on top of gurney-side clinical assessment with blood testing, echocardiogram and so on.

Free-text radiology reports can be automatically classified by convolutional neural networks (CNNs) powered by deep-learning algorithms with accuracy that’s equal to or better than that achieved by traditional—and more labor-intensive—natural language processing (NLP) methods.