Comparative analysis of machine learning algorithms for computer-assisted reporting based on fully automated cross-lingual RadLex mappings

• Published in: Scientific reports Vol. 11; no. 1; p. 5529
• Main Authors: Maros, Máté E., Cho, Chang Gyu, Junge, Andreas G., Kämpgen, Benedikt, Saase, Victor, Siegel, Fabian, Trinkmann, Frederik, Ganslandt, Thomas, Groden, Christoph, Wenz, Holger
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 09-03-2021; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114; 631/114/1305; 631/114/2164; 631/114/2397; 631/114/2415; 692/308/53/2421; 692/308/53/2422; 692/308/53/2423; Algorithms; Automation; Calibration; Comparative analysis; Computed tomography; Humanities and Social Sciences; Learning algorithms; Machine learning; multidisciplinary; Radiology; Science; Science (multidisciplinary); Support vector machines
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-021-85016-9

Computer-assisted reporting (CAR) tools were suggested to improve radiology report quality by context-sensitively recommending key imaging biomarkers. However, studies evaluating machine learning (ML) algorithms on cross-lingual ontological (RadLex) mappings for developing embedded CAR algorithms are lacking. Therefore, we compared ML algorithms developed on human expert-annotated features against those developed on fully automated cross-lingual (German to English) RadLex mappings using 206 CT reports of suspected stroke. Target label was whether the Alberta Stroke Programme Early CT Score (ASPECTS) should have been provided (yes/no:154/52). We focused on probabilistic outputs of ML-algorithms including tree-based methods, elastic net, support vector machines (SVMs) and fastText (linear classifier), which were evaluated in the same 5 × fivefold nested cross-validation framework. This allowed for model stacking and classifier rankings. Performance was evaluated using calibration metrics (AUC, brier score, log loss) and -plots. Contextual ML-based assistance recommending ASPECTS was feasible. SVMs showed the highest accuracies both on human-extracted- (87%) and RadLex features (findings:82.5%; impressions:85.4%). FastText achieved the highest accuracy (89.3%) and AUC (92%) on impressions. Boosted trees fitted on findings had the best calibration profile. Our approach provides guidance for choosing ML classifiers for CAR tools in fully automated and language-agnostic fashion using bag-of-RadLex terms on limited expert-labelled training data.