Development of an online, publicly accessible naive Bayesian decision support tool for mammographic mass lesions based on the American College of Radiology (ACR) BI-RADS lexicon

• Published in: European radiology Vol. 25; no. 6; pp. 1768 - 1775
• Main Authors: Benndorf, Matthias, Kotter, Elmar, Langer, Mathias, Herda, Christoph, Wu, Yirong, Burnside, Elizabeth S.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2015; Springer Nature B.V
• Subjects: Adult; Aged; Bayes Theorem; Breast Neoplasms - diagnostic imaging; Cancer; Computer Applications; Data collection; Decision Support Systems, Clinical - standards; Diagnostic Radiology; Female; Humans; Imaging; Internal Medicine; Internet; Interventional Radiology; Mammography - methods; Mammography - standards; Medicine; Medicine & Public Health; Middle Aged; Neuroradiology; Radiology; Radiology - education; ROC Curve; Terminology as Topic; Ultrasound; United States; Variables; United States; United States > US; Decision support techniques; Mammography; BI-RADS, CAD; Bayesian analysis
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-014-3570-6

Purpose To develop and validate a decision support tool for mammographic mass lesions based on a standardized descriptor terminology (BI-RADS lexicon) to reduce variability of practice. Materials and methods We used separate training data (1,276 lesions, 138 malignant) and validation data (1,177 lesions, 175 malignant). We created naïve Bayes (NB) classifiers from the training data with tenfold cross-validation. Our “inclusive model” comprised BI-RADS categories, BI-RADS descriptors, and age as predictive variables; our “descriptor model” comprised BI-RADS descriptors and age. The resulting NB classifiers were applied to the validation data. We evaluated and compared classifier performance with ROC-analysis. Results In the training data, the inclusive model yields an AUC of 0.959; the descriptor model yields an AUC of 0.910 ( P  < 0.001). The inclusive model is superior to the clinical performance (BI-RADS categories alone, P  < 0.001); the descriptor model performs similarly. When applied to the validation data, the inclusive model yields an AUC of 0.935; the descriptor model yields an AUC of 0.876 ( P  < 0.001). Again, the inclusive model is superior to the clinical performance ( P  < 0.001); the descriptor model performs similarly. Conclusion We consider our classifier a step towards a more uniform interpretation of combinations of BI-RADS descriptors. We provide our classifier at www.ebm-radiology.com/nbmm/index.html . Key Points • We provide a decision support tool for mammographic masses at www.ebm-radiology.com/nbmm/index.html . • Our tool may reduce variability of practice in BI-RADS category assignment. • A formal analysis of BI-RADS descriptors may enhance radiologists’ diagnostic performance.