Deciding which patients should go into the intensive care unit (ICU) after surgery is a difficult call and typically made entirely at the surgeon's discretion. The result is that surgeons typically err on the side of caution by putting more post-operative patients in the ICU than necessary. To aid in better ICU decision-making, physicians at New York University Langone Hospital System (NYU Langone) developed a machine learning algorithm that combs through a patient's electronic medical record to identify relevant factors to determine if they needed the ICU after surgery.

Genomic testing firm Veracyte Inc. is eyeing 2021 for the launch of its noninvasive nasal swab classifier for early lung cancer detection and diagnosis, following preliminary clinical data demonstrating high sensitivity in low-risk patients and high specificity in high-risk patients with known lung nodules. The South San Francisco-based company is developing the nasal swab test in collaboration with Johnson & Johnson Inc.'s Lung Cancer Initiative, part of a long-term strategic collaboration that also aims to speed commercialization of Veracyte's Percepta genomic sequencing classifier.

Radiologists review thousands of images a day. The hope is that artificial intelligence (AI) applications will become useful soon to verify diagnoses, prioritize queued images and even to offer a level of detection and measurement that aren't feasible for humans. One of the latest efforts on this front is by researchers at the University of California at San Francisco (UCSF) and the University of California at Berkeley.

A recent medical journal article says the terminology used by physicians to denote a fatality in the FDA adverse event reporting system has led to underreporting of fatalities associated with two prominent cardiology devices, a predicament the authors say skews the public understanding of these devices' safety profiles.

Diagnosis and treatment of infections typically occurs after people exhibit obvious signs of illness, such as fever or a cough. By then, they may already have exposed others and are well on the way to developing more serious symptoms themselves. In the military, such delays can hamper medical countermeasures to contain potential outbreaks and reduce downtime among active duty personnel. Now, Amsterdam-based Royal Philips NV and the U.S. Department of Defense's Defense Threat Reduction Agency and Defense Innovations Unit have built an early warning algorithm – using artificial intelligence – to detect infection before a person shows any signs or symptoms of infection.

The FDA's September 2019 final guidance for the humanitarian device exemption program brought some clarity to several issues, but device makers must still untangle the question of which tasks an institutional review board (IRB) has delegated to an appropriate local committee for a specific clinical site.

The U.S. FDA has given 510(k) clearance to the Advanced Intelligent Clear-IQ Engine (AiCE) for Canon Medical Systems USA Inc.'s Aquilion Precision CT scanner. The regulatory green light brings artificial intelligence (AI)-based image reconstruction capabilities to the world's first ultra-high resolution CT imaging system.

Researchers at the Abramson Cancer Center at the University of Pennsylvania have developed an algorithm to better personalize immunotherapy treatment. The algorithm works by examining neoantigen quality, not just their quantity. Neoantigens are proteins that are the result of genetic mutations in a tumor.