Machine learning with multiparametric magnetic resonance imaging of the breast for early prediction of response to neoadjuvant chemotherapy. (February 2020)

• Record Type: Journal Article
• Authors: Lo Gullo, Roberto; Eskreis-Winkler, Sarah; Morris, Elizabeth A.; Pinker, Katja
• Is Part Of: Breast. Volume 49(2020)
• Page Start: 115
• Page End: 122
• Publication Date: 2020-02
• Subjects: Artificial intelligence -- Machine learning -- Multiparametric MRI -- Neoadjuvant chemotherapy; 1H-MRS proton magnetic resonance spectroscopy -- 23 N MRS sodium magnetic resonance spectroscopy -- ADC apparent diffusion coefficient -- ANN artificial neural network -- AUC area under the curve -- CNN convolutional neural network -- DCE-MRI dynamic contrast-enhanced magnetic resonance imaging -- DWI diffusion-weighted imaging -- EF enhancement fraction -- FGT fibroglandular tissue -- LR logistic regression -- MB Markov blanket -- NAC neoadjuvant chemotherapy -- pCR pathologic complete response -- SVM support vector machine -- TN triple negative; Breast -- Diseases -- Periodicals; Breast -- Tumors -- Periodicals; Breast -- Periodicals; Electronic journals; Periodicals; 616
• DOI: 10.1016/j.breast.2019.11.009
• Languages: English
• ISSNs: 0960-9776

Abstract

Abstract: In patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy (NAC), some patients achieve a complete pathologic response (pCR), some achieve a partial response, and some do not respond at all or even progress. Accurate prediction of treatment response has the potential to improve patient care by improving prognostication, enabling de-escalation of toxic treatment that has little benefit, facilitating upfront use of novel targeted therapies, and avoiding delays to surgery. Visual inspection of a patient's tumor on multiparametric MRI is insufficient to predict that patient's response to NAC. However, machine learning and deep learning approaches using a mix of qualitative and quantitative MRI features have recently been applied to predict treatment response early in the course of or even before the start of NAC. This is a novel field but the data published so far has shown promising results. We provide an overview of the machine learning and deep learning models developed to date, as well as discuss some of the challenges to clinical implementation. Highlights: Machine and deep learning have shown potential in predicting neoadjuvant treatment outcomes using multiparametric MRI data. Machine and deep learning-based early identification of chemotherapy non-responders could improve patient management. Deep learning techniques using CNN may prove to be more powerful and more robust than traditional machine learning classifiers.