Breast cancer patients who have pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) are more likely to have better clinical outcomes. The ability to predict which patient will respond to NAC early in the treatment course is important because it could help to minimize unnecessary toxic NAC and to modify regimens mid-treatment to achieve better efficacy. Machine learning (ML) is increasingly being used in radiology and medicine because it can identify relationships amongst complex data elements to inform outcomes without the need to specify such relationships a priori. One of the most popular deep learning methods that applies to medical images is the Convolutional Neural Networks (CNN). In contrast to supervised ML, deep learning CNN can operate on the whole images without requiring radiologists to manually contour the tumor on images. Although there have been many review papers on supervised ML prediction of pCR, review papers on deep learning prediction of pCR are sparse. Deep learning CNN could also incorporate multiple image types, clinical data such as demographics and molecular subtypes, as well as data from multiple treatment time points to predict pCR. The goal of this study is to perform a systematic review of deep learning methods that use whole-breast MRI images without annotation or tumor segmentation to predict pCR in breast cancer.

Purpose To predict pathological complete response (pCR) after neoadjuvant chemotherapy using extreme gradient boosting (XGBoost) with MRI and non-imaging data at multiple treatment timepoints. Material and methods This retrospective study included breast cancer patients (n = 117) who underwent neoadjuvant chemotherapy. Data types used included tumor ADC values, diffusion-weighted and dynamic-contrast-enhanced MRI at three treatment timepoints, and patient demographics and tumor data. GLCM textural analysis was performed on MRI data. An extreme gradient boosting machine learning algorithm was used to predict pCR. Prediction performance was evaluated using the area under the curve (AUC) of the receiver operating curve along with precision and recall. Results Prediction using texture features of DWI and DCE images at multiple treatment time points (AUC = 0.871; 95% CI: (0.768, 0.974; p<0.001) and (AUC = 0.903 95% CI: 0.854, 0.952; p<0.001) respectively), outperformed that using mean tumor ADC (AUC = 0.850 (95% CI: 0.764, 0.936; p<0.001)). The AUC using all MRI data was 0.933 (95% CI: 0.836, 1.03; p<0.001). The AUC using non-MRI data was 0.919 (95% CI: 0.848, 0.99; p<0.001). The highest AUC of 0.951 (95% CI: 0.909, 0.993; p<0.001) was achieved with all MRI and all non-MRI data at all time points as inputs. Conclusion Using XGBoost on extracted GLCM features and non-imaging data accurately predicts pCR. This early prediction of response can minimize exposure to toxic chemotherapy, allowing regimen modification mid-treatment and ultimately achieving better outcomes.

Deep learning analysis of radiological images has the potential to improve diagnostic accuracy of breast cancer, ultimately leading to better patient outcomes. This paper systematically reviewed the current literature on deep learning detection of breast cancer based on magnetic resonance imaging (MRI). The literature search was performed from 2015 to Dec 31, 2022, using Pubmed. Other database included Semantic Scholar, ACM Digital Library, Google search, Google Scholar, and pre-print depositories (such as Research Square). Articles that were not deep learning (such as texture analysis) were excluded. PRISMA guidelines for reporting were used. We analyzed different deep learning algorithms, methods of analysis, experimental design, MRI image types, types of ground truths, sample sizes, numbers of benign and malignant lesions, and performance in the literature. We discussed lessons learned, challenges to broad deployment in clinical practice and suggested future research directions.