A new machine learning tool extracts imaging data "invisible" to humans and distinguishes pre-cancerous from non-cancerous polyps on CT colonography scans, researchers reported Tuesday in Radiology.
German investigators trained their algorithm on a set of more than 100 colorectal polyps from 63 patients, sharing their process in Radiology. The proof-of-concept study was a success, they noted, with the approach notching an “excellent” area-under-the-curve score of 0.91 for distinguishing between polyp types.
Radiologists may be able to use the algorithm to help diagnose one of the top three causes of cancer-related death among men and women: colorectal cancer.
"Adding machine learning-assisted image analysis to conventional, radiological image reading could further improve the clinical significance of CT colonography-based colorectal cancer screening by allowing for a more precise selection of patients eligible for subsequent polypectomy," lead author Sergio Grosu, MD, a radiologist with University Hospital in Munich, Germany, explained in a statement. “This method could be used routinely as a second reader in all CT colonography examinations in the distant future."
Grosu and colleagues trained their radiomics-based machine learning platform to extract quantitative imaging features and predict individual polyp types. They tested their creation on a set of 77 polyps from 59 patients.
In that set, the tool notched a sensitivity of 82% and specificity of 85% for noninvasively differentiating benign from premalignant polyps.
Additional testing with more patients will help validate their findings, Grosu noted.
"Further refinement of the machine learning-based image analysis is necessary to achieve higher precision in polyp differentiation as well as workflow optimization for better applicability in clinical routine," he added.
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.