AI predicts malignancy, survival in patients with brain tumors
AI models can be trained to predict outcomes in meningioma patients, according to new research published in npj Digital Medicine. The study’s authors even developed a free smartphone app so others can explore their work.
“Meningiomas are the most common primary central nervous system (CNS) tumor, with an incidence of 8.14 per 100,000 population,” wrote lead author Jeremy T. Moreau, McGill University in Montreal, Canada, and colleagues. “They typically present with gradual onset of symptoms in the later decades of life and have generally favorable outcomes relative to other CNS tumors. However, there is a great deal of variability in both aggressiveness and outcomes.”
That variability, Moreau et al. explained, can make characterizing meningiomas especially challenging. And that’s where AI—or machine learning, to be more specific—enters the equation. The research team used data from more than 62,000 meningioma patients to develop its AI models, noting that they were able to predict patient outcomes with considerable accuracy.
“We report the development and validation of predictive models of meningioma malignancy and associated survival,” the authors wrote. “On the basis of a very limited set of clinical variables such as age, sex, and tumor size, our models are shown to be capable of predicting individual-patient outcomes.”
However, the team added, much more research is still needed in this area. One way they hoped to help that research along was through the development of an open-source smartphone app that helps interested parties interact with the completed algorithms.
“We have gotten great feedback on how the app can be used to explore how different clinical factors might influence malignancy and survival,” Moreau said in a prepared statement. “We believe it provides a unique entry point for furthering the translatability and transparency of machine learning models, which too often remain impossible for the average clinician to evaluate because of the time and programming knowledge this would require.”
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.