Over the past week, many of our Instagram and Twitter feeds have been filled with avatars of our friends looking like heroes in a cyberpunk thriller. With their Day-Glo pink hair, shimmery skin, and Mad Max–meets–Ren Faire corsets, many of these profile pics looked so good that it was tempting to download the Lensa app and splurge on 100 A.I.-generated portraits of one's own. But by the time TMZ built its slideshow of the best celebrity Lensa looks, the tide had turned. Yes, the app, which continues to top Apple's free apps chart (users are prompted within Lensa to pay for the avatars), was succeeding in making hot people hotter and helping the selfie-averse find their superhero inside them. But there was also the matter of theft.

CHICAGO--An artificial intelligence machine-learning program has demonstrated the ability to accurately forecast which head and neck cancer patients are likely to experience severe weight loss, necessitating the use of a feeding tube, researchers at MD Anderson Cancer Center in Houston told attendees at the 2019 ASTRO Annual Meeting (Abstract 141). It marks the first time that such a sophisticated "self-teaching" computer algorithm has accurately identified patients likely to develop problems eating, said Jay Reddy, MD, PhD, Assistant Professor of Radiation Oncology and lead author of the study. "With head and neck radiation, a lot of toxicity occurs; however it's not always clear which patients will experience serious side effects," he told a press conference. Reddy and his colleagues used machine learning models to analyze large datasets from three sources--electronic health records, an internal web-based patient charting tool, and the hospital's records and verification system--in an effort to discern and eventually predict patients with weight loss exceeding 10 percent of total body weight, the need for a feeding tube, and/or any unplanned hospitalization within 3 months of radiation. Machine learning is a relatively powerful application of artificial intelligence (AI)–think facial recognition software--by which a computer program can automatically learn and improve itself by analyzing large quantities of data.

MD Anderson researchers have developed the first machine learning algorithm to predict acute toxicities in patients receiving radiation therapy for head and neck cancers. The results of the study were presented today at the 61st Annual Meeting of the American Society for Radiation Oncology (ASTRO). "With head and neck radiation, a lot of toxicity occurs, however it's not always clear which patients will experience serious side effects," says study lead Jay Reddy, M.D., Ph.D., assistant professor of Radiation Oncology. Reddy's team set out to develop algorithms that could predict significant weight loss ( 10% during radiation therapy), feeding tube placement and unplanned hospitalizations with three months of beginning radiation treatment. "It's virtually unheard of for these patients to not lose any weight at all, but many patients are able to complete treatment without a feeding tube. Thus, we don't want to unnecessarily place one on a hunch. Prolonged time with a feeding tube can hamper efforts to rehabilitate swallowing muscles. "The challenge is to balance this concern with the knowledge that some patients can't get through treatment without assistance, and their need for nutrition becomes dire.