Jansen MJA, et al. - Given that in the automatic classification of focal liver lesions, only T2-weighted images are exploited till date, researchers examined the utility of additional MR sequences and risk factors in automatic classification for improving the results and aimed to make a step forward to a clinically useful aid for radiologists. Using clinical MRI data sets of 95 patients who had in total 125 benign lesions (40 adenomas, 29 cysts and 56 hemangiomas) and 88 malignant lesions (30 hepatocellular carcinomas (HCC) and 58 metastases), they extracted contrast curve, gray level histogram, and gray level co-occurrence matrix texture features. Furthermore, they used risk factors including the presence of steatosis, cirrhosis, and a known primary tumor as features. They selected 50 features with the highest ANOVA F-score and fed them to an extremely randomized trees classifier. Findings supported this classifier as capable of distinguishing five common types of lesions (adenoma, cyst, hemangioma, HCC, and metastasis). Lesions,