Radiomics for Predicting CyberKnife response in acoustic neuroma: a pilot study

Abstract

Vestibular schwannomas, also known as acoustic neuromas, are a primary intracranial tumor of the myelin-forming cells of the 8 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> cranial nerve. Stereotactic radiosurgery, which can be performed with the CyberKnife robotic device, is one of the approaches for managing this disease, and has shown to be effective in controlling tumor growth. However, it may have side effects and up to two years may be needed to assess its efficacy. In this work we present a machine learning-based radiomics approach that first computes quantitative biomarkers from MR images routinely collected before the CyberKnife treatment and then predicts the treatment response. Furthermore, the degree of class imbalance observed in the available dataset suggested us to resample the data during the learning stage. The results achieved are promising, with an accuracy equal to 85.3%.