Neoplastic and Non-neoplastic Acute Intracerebral Hemorrhage in CT Brain Scans: Machine Learning-Based Prediction Using Radiomic Image Features

Background: Early differentiation of neoplastic and non-neoplastic intracerebral hemorrhage (ICH) can be difficult in initial radiological evaluation, especially for extensive ICHs. The aim of this study was to evaluate the potential of a machine learning-based prediction of etiology for acute ICHs based on quantitative radiomic image features extracted from initial non-contrast-enhanced computed tomography (NECT) brain scans. Methods: The analysis included NECT brain scans from 77 patients with acute ICH (n = 50 non-neoplastic, n = 27 neoplastic). Radiomic features including shape, histogram, and texture markers were extracted from non-, wavelet-, and log-sigma-filtered images using regions of interest of ICH and perihematomal edema (PHE). Six thousand and ninety quantitative predictors were evaluated utilizing random forest algorithms with five-fold model-external cross-validation. Model stability was assessed through comparative analysis of 10 randomly drawn cross-validation sets. Classifier performance was compared with predictions of two radiologists employing the Matthews correlation coefficient (MCC). Results: The receiver operating characteristic (ROC) area under the curve (AUC) of the test sets for predicting neoplastic vs. non-neoplastic ICHs was 0.89 [95% CI (0.70; 0.99); P < 0.001], and specificities and sensitivities reached >80%. Compared to the radiologists' predictions, the machine learning algorithm yielded equal or superior results for all evaluated metrics. The MCC of the proposed algorithm at its optimal operating point (0.69) was significantly higher than the MCC of the radiologist readers (0.54); P = 0.01. Conclusion: Evaluating quantitative features of acute NECT images in a machine learning algorithm provided high discriminatory power in predicting non-neoplastic vs. neoplastic ICHs. Utilized in the clinical routine, the proposed approach could improve patient care at low risk and costs.