Implementation of machine learning into clinical breast MRI: Potential for objective and accurate decision-making in suspicious breast masses

Abstract

We investigated whether the integration of machine learning (ML) into MRI interpretation can provide accurate decision rules for the management of suspicious breast masses. A total of 173 consecutive patients with suspicious breast masses upon complementary assessment (BI-RADS IV/V: n = 100/76) received standardized breast MRI prior to histological verification. MRI findings were independently assessed by two observers (R1/R2: 5 years of experience/no experience in breast MRI) using six (semi-)quantitative imaging parameters. Interobserver variability was studied by ICC (intraclass correlation coefficient). A polynomial kernel function support vector machine was trained to differentiate between benign and malignant lesions based on the six imaging parameters and patient age. Tenfold cross-validation was applied to prevent overfitting. Overall diagnostic accuracy and decision rules (rule-out criteria) to accurately exclude malignancy were evaluated. Results were integrated into a web application and published online. Malignant lesions were present in 107 patients (60.8%). Imaging features showed excellent interobserver variability (ICC: 0.81-0.98) with variable diagnostic accuracy (AUC: 0.65-0.82). Overall performance of the ML algorithm was high (AUC = 90.1%; BI-RADS IV: AUC = 91.6%). The ML algorithm provided decision rules to accurately rule-out malignancy with a false negative rate <1% in 31.3% of the BI-RADS IV cases. Thus, integration of ML into MRI interpretation can provide objective and accurate decision rules for the management of suspicious breast masses, and could help to reduce the number of potentially unnecessary biopsies. OPEN ACCESS Citation: Ellmann S, Wenkel E, Dietzel M, Bielowski C, Vesal S, Maier A, et al. (2020) Implementation of machine learning into clinical breast MRI: Potential for objective and accurate decision-making in suspicious breast masses. PLoS ONE 15(1): e0228446. https://doi.org/10.Machine learning in breast MRI PLOS ONE | https://doi.org/10.1371/journal.pone.0228446 January 30, 2020 2 / 15 manuscript, and in the decision to publish the results. Competing interests: The authors of this manuscript declare relationships with the following companies: Michael Uder is on the speakers' bureau for Bracco, Medtronic, Siemens and Bayer Schering. Rolf Janka is on the speakers' bureau for Bracco. Tobias Bäuerle is on the speakers' bureau for Bracco and Boehringer Ingelheim. For the remaining authors no potential conflicts of interest were declared. This does not alter our adherence to PLOS ONE policies on sharing data and materials.An ML algorithm (polynomial kernel function support vector machine) was used for lesion classification [22]. These algorithms aim to define a decision boundary between two classes Fig 3. Clinical cases: Breast MRI of three different patients with suspicious lesions. Case 1: A 55-year-old woman presenting with a mass in her left breast, measuring 21 × 18 mm. ADC was 1015 × 10 −6 mm 2 /s. There was a type-2 curve. T2w SI was 4.6. The SVM diagnosed malignancy (er...

Diagnostic performance of the Kaiser score for characterizing lesions on breast MRI with comparison to a multiparametric classification system

“…To improve the specificity and diagnostic accuracy of MRI findings, researchers have explored the value of multiparametric scoring and evaluation schemes, with varying results [8][9][10][11]. Furthermore, a recent meta-analysis of breast MRI studies concluded that diffusion-weighted imaging with a multiparametric protocol improved the specificity for the diagnosis of malignant lesions but it did not affect the sensitivity [12].…”

Images Are Data: A Breast Imaging Perspective on a Contemporary Paradigm

“…A large variety of methods belong to the machine learning family, such as artificial neural networks (ANNs), support vector machines (SVMs), but also classic statistical procedures, such as decision trees and regression analysis [46]. Each method provides inherent advantages and has been successfully applied to breast imaging research [14,33,53,54].…”

Bilder sind Daten: Eine Perspektive der Brustbildgebung auf ein zeitgenössisches Paradigma

“…Zur Familie des maschinellen Lernens gehört eine Vielzahl von Methoden wie künstliche neuronale Netze (ANNs), Support Vector Machines (SVMs), aber auch klassische statistische Verfahren wie Entscheidungsbäume und Regressionsanalysen [46]. Jede Methode bietet inhärente Vorteile und wurde bereits erfolgreich in der Brustbildforschung eingesetzt [14,33,53,54].…”