• Amalya Zeynalova, Burak Kocak, Emine Sebnem Durmaz, Nil Comunoglu, Kerem Ozcan, Gamze Ozcan, Okan Turk, Necmettin Tanriover, Naci Kocer, Osman Kizilkilic, and Civan Islak.
• Department of Radiology, Cerrahpasa Medical Faculty, Istanbul University-Cerrahpasa, Istanbul, Turkey.
• Neuroradiology. 2019 Jul 1; 61 (7): 767-774.

PurposeTo evaluate the potential value of machine learning (ML)-based histogram analysis (or first-order texture analysis) on T2-weighted magnetic resonance imaging (MRI) for predicting consistency of pituitary macroadenomas (PMA) and to compare it with that of signal intensity ratio (SIR) evaluation.MethodsFifty-five patients with 13 hard and 42 soft PMAs were included in this retrospective study. Histogram features were extracted from coronal T2-weighted original, filtered and transformed MRI images by manual segmentation. To achieve balanced classes (38 hard vs 42 soft), multiple samples were obtained from different slices of the PMAs with hard consistency. Dimension reduction was done with reproducibility analysis, collinearity analysis and feature selection. ML classifier was artificial neural network (ANN). Reference standard for the classifications was based on surgical and histopathological findings. Predictive performance of histogram analysis was compared with that of SIR evaluation. The main metric for comparisons was the area under the receiver operating characteristic curve (AUC).ResultsOnly 137 of 162 features had excellent reproducibility. Collinearity analysis yielded 20 features. Feature selection algorithm provided six texture features. For histogram analysis, the ANN correctly classified 72.5% of the PMAs regarding consistency with an AUC value of 0.710. For SIR evaluation, accuracy and AUC values were 74.5% and 0.551, respectively. Considering AUC values, ML-based histogram analysis performed better than SIR evaluation (z = 2.312, p = 0.021).ConclusionML-based T2-weighted MRI histogram analysis might be a useful technique in predicting the consistency of PMAs, with a better predictive performance than that of SIR evaluation.

20 keywords...

hide…