Clinical value of radiomics and machine learning in breast ultrasound: a multicenter study for differential diagnosis of benign and malignant lesions.

OBJECTIVES: We aimed to assess the performance of radiomics and machine learning (ML) for classification of non-cystic benign and malignant breast lesions on ultrasound images, compare ML's accuracy with that of a breast radiologist, and verify if the radiologist's performance is improved by using ML. METHODS: Our retrospective study included patients from two institutions. A total of 135 lesions from Institution 1 were used to train and test the ML model with cross-validation. Radiomic features were extracted from manually annotated images and underwent a multistep feature selection process. Not reproducible, low variance, and highly intercorrelated features were removed from the dataset. Then, 66 lesions from Institution 2 were used as an external test set for ML and to assess the performance of a radiologist without and with the aid of ML, using McNemar's test. RESULTS: After feature selection, 10 of the 520 features extracted were employed to train a random forest algorithm. Its accuracy in the training set was 82% (standard deviation, SD, ± 6%), with an AUC of 0.90 (SD ± 0.06), while the performance on the test set was 82% (95% confidence intervals (CI) = 70-90%) with an AUC of 0.82 (95% CI = 0.70-0.93). It resulted in being significantly better than the baseline reference (p = 0.0098), but not different from the radiologist (79.4%, p = 0.815). The radiologist's performance improved when using ML (80.2%), but not significantly (p = 0.508). CONCLUSIONS: A radiomic analysis combined with ML showed promising results to differentiate benign from malignant breast lesions on ultrasound images. KEY POINTS: • Machine learning showed good accuracy in discriminating benign from malignant breast lesions • The machine learning classifier's performance was comparable to that of a breast radiologist • The radiologist's accuracy improved with machine learning, but not significantly.

Prostate MRI technical parameters standardization: A systematic review on adherence to PI-RADSv2 acquisition protocol.

PURPOSE: The Prostate Imaging-Reporting and Data System has been developed to standardize prostate MRI in terms of acquisition, interpretation and reporting. It received a major revision in late 2014 (PI-RADSv2). Recently, doubts have been raised on imaging facilities adherence to its acquisition protocol. With this systematic review, we assessed adherence to PI-RADSv2 minimum technical specifications in literature, to achieve a better understanding of issues limiting their diffusion. METHOD: Multiple medical literature databases were extensively searched to retrieve original studies published after January 2016 performing prostate MRI. Information pertaining acquisition protocols and patient enrolment were recorded for analysis. Technical parameters were dichotomized in relation to adherence to the corresponding minimal technical requirements. RESULTS: A total of 150 studies were included for analysis. Only 5% reported every technical parameter specified in the PI-RADSv2 document requirements, none of which completely met guideline specifications. Overall, 19% were in line with PI-RADSv2 for all reported MRI acquisition parameters. The adherence was lowest for T2-weighted frequency in-plane resolution (12%), diffusion-weighted imaging field of view (40%), apparent diffusion coefficient map low b-value (27%) and dynamic contrast-enhanced imaging temporal resolution (43%). Considering its role in image interpretation, it must be highlighted that only 59% of studies reporting diffusion-weighted imaging high b-value follow recommendations. CONCLUSIONS: Adherence to PI-RADSv2 minimum technical standards is heterogeneous in the scientific community. Our findings endorse the need for greater diffusion of PI-RADSv2 guidelines to achieve protocol standardization and support the notion that some requirements might benefit from streamlining to improve clinical applicability.