A team of researchers at Harvard Medical School have developed a new artificial intelligence (AI) tool called CHARM (Computer-aided Human-Rated Malignant) that can help doctors more accurately detect brain tumors. The study, which was published in the journal Nature Medicine, found that CHARM was able to identify brain tumors with a 97% accuracy rate, compared to 87% for human radiologists.
CHARM works by analyzing MRI scans of the brain and identifying patterns that are associated with tumors. The AI tool then generates a score for each scan, which indicates the likelihood of a tumor being present. The scores are then reviewed by radiologists, who can use them to make a more informed diagnosis.
The study's authors say that CHARM could help to improve the accuracy of brain tumor detection and lead to earlier treatment for patients. They also believe that the AI tool could be used to develop new screening methods for brain tumors.
According to the study, CHARM uses machine learning to identify the genetic profile of glioma, a brain tumor that can lead to death in six months if untreated.
Surgeons usually rely on a glioma tumor’s genetic profile to decide the extent of tissue removal needed. Obtaining this information is time-consuming. Thankfully, CHARM, trained with a plethora of pictures from other patients’ diagnoses, can predict a tumor’s genetic profile almost instantly, allowing doctors to proceed with the proper treatment without scheduling and performing another surgery.
But while developed for glioma, a recent editorial published in the Lancet Oncology suggests CHARM could be retrained to accurately identify individuals with an elevated risk of pancreatic cancer, lung cancer, breast cancer, and more.
CHARM was particularly accurate at identifying small tumors and tumors in difficult-to-see areas of the brain and was able to generate scores for MRI scans in just a few minutes, which could help to speed up the diagnosis process.
The study's findings are promising, and CHARM could have a significant impact on the early detection and treatment of brain tumors. However, more research is needed to confirm the AI tool's accuracy and to assess its impact on patient outcomes.
On the date of publication, Andy Mukolo did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. For more information please view the Barchart Disclosure Policy here.
