AI enables more accurate tumor diagnosis in thyroid cancer

(CONNECT) A team of researchers from the Swiss Federal Laboratories for Materials Science and Technology (Empa) in Dübendorf and the University of Bern has developed an AI-supported analysis method for thyroid tumors. According to a press release, this is a non-invasive histopathological 3D imaging technique that should help to improve the diagnosis of various cancers in the future.

Together with pathologists from the University of Bern, the researchers were able to prove that the method "can detect clinically relevant tissue characteristics in thyroid tumors". X-ray phase-contrast micro-computed tomography (micro-CT) makes even the smallest differences in soft tissue visible and can also detect tumor characteristics - so-called biomarkers - in the deeper layers of tissue, which may be overlooked by conventional methods.

The method was developed by Robert Zboray, group leader at Empa's Centre for X-ray Analysis. According to him, the method allows a three-dimensional analysis of complete biopsy blocks of a tumor in a short time.  The removed tissue is spared from alteration and destruction. "This means that the sample can still be used for further molecular biological investigations," he is quoted as saying.

According to the press release, there are 300 million people affected by thyroid cancer worldwide, but the biomarkers are not the same for all patients. The researchers hope that the 3D analysis will lead to more accurate and therefore more precise diagnoses and prognoses. ce/heg