Ensemble Learning for Breast Cancer Lesion Classification: A Pilot Validation Using Correlated Spectroscopic Imaging and Diffusion-Weighted Imaging.

The main objective of this work was to evaluate the application of individual and ensemble machine learning models to classify malignant and benign breast masses using features from two-dimensional (2D) correlated spectroscopy spectra extracted from five-dimensional echo-planar correlated spectroscopic imaging (5D EP-COSI) and diffusion-weighted imaging (DWI). Twenty-four different metabolite and lipid ratios with respect to diagonal fat peaks (1.4 ppm, 5.4 ppm) from 2D spectra, and water and fat peaks (4.7 ppm, 1.4 ppm) from one-dimensional non-water-suppressed (NWS) spectra were used as the features. Additionally, water fraction, fat fraction and water-to-fat ratios from NWS spectra and apparent diffusion coefficients (ADC) from DWI were included. The nine most important features were identified using recursive feature elimination, sequential forward selection and correlation analysis. XGBoost (AUC: 93.0%, Accuracy: 85.7%, F1-score: 88.9%, Precision: 88.2%, Sensitivity: 90.4%, Specificity: 84.6%) and GradientBoost (AUC: 94.3%, Accuracy: 89.3%, F1-score: 90.7%, Precision: 87.9%, Sensitivity: 94.2%, Specificity: 83.4%) were the best-performing models. Conventional biomarkers like choline, myo-Inositol, and glycine were statistically significant predictors. Key features contributing to the classification were ADC, 2D diagonal peaks at 0.9 ppm, 2.1 ppm, 3.5 ppm, and 5.4 ppm, cross peaks between 1.4 and 0.9 ppm, 4.3 and 4.1 ppm, 2.3 and 1.6 ppm, and the triglyceryl-fat cross peak. The results highlight the contribution of the 2D spectral peaks to the model, and they demonstrate the potential of 5D EP-COSI for early breast cancer detection.

To scan or not to scan: effect of scanning the axilla of all patients undergoing diagnostic breast ultrasound.

BACKGROUND: Breast ultrasonography is a useful modality in patients undergoing diagnostic and screening breast imaging. However, breast ultrasound has a high false positive rate and can be time-consuming to perform. PURPOSE: The purpose of this study was to evaluate the clinical impact of incidental axillary findings found on diagnostic breast ultrasounds at a single multi-site institution that has a standard protocol of scanning the axilla for all breast ultrasound exams. METHODS: All diagnostic breast ultrasounds were retrospectively reviewed from January 2017 to September 2019. Follow-up imaging, relevant clinical history, and pathology results were also reviewed. All positive axillary findings were divided into incidental or non-incidental findings depending on whether there was a direct clinical indication to scan the axilla. Descriptive statistics were performed with a 5% level of significance. RESULTS: Of the 19,695 diagnostic ultrasounds performed during this timeframe, there were 91 (0.5%) incidental axillary findings given a BIRADS category 3 or 4, and none of these findings resulted in the diagnosis of an occult breast cancer. One biopsy-proven SLL/CLL lymphoma was diagnosed that was otherwise clinically occult. CONCLUSION: Routine axillary scanning in all patients undergoing a diagnostic breast ultrasound at a large multi-site institution yields a low rate of incidental findings and has minimal impact on detection of cancer.

Correlated MR spectroscopic imaging of breast cancer to investigate metabolites and lipids: acceleration and compressed sensing reconstruction.

Objectives: The main objective of this work was to detect novel biomarkers in breast cancer by spreading the MR spectra over two dimensions in multiple spatial locations using an accelerated 5D EP-COSI technology. Methods: The 5D EP-COSI data were non-uniformly undersampled with an acceleration factor of 8 and reconstructed using group sparsity-based compressed sensing reconstruction. Different metabolite and lipid ratios were then quantified and statistically analyzed for significance. Linear discriminant models based on the quantified metabolite and lipid ratios were generated. Spectroscopic images of the quantified metabolite and lipid ratios were also reconstructed. Results: The 2D COSY spectra generated using the 5D EP-COSI technique showed differences among healthy, benign, and malignant tissues in terms of their mean values of metabolite and lipid ratios, especially the ratios of potential novel biomarkers based on unsaturated fatty acids, myo-inositol, and glycine. It is further shown the potential of choline and unsaturated lipid ratio maps, generated from the quantified COSY signals across multiple locations in the breast, to serve as complementary markers of malignancy that can be added to the multiparametric MR protocol. Discriminant models using metabolite and lipid ratios were found to be statistically significant for classifying benign and malignant tumor from healthy tissues. Conclusions: Accelerated 5D EP-COSI technique demonstrates the potential to detect novel biomarkers such as glycine, myo-inositol, and unsaturated fatty acids in addition to commonly reported choline in breast cancer, and facilitates metabolite and lipid ratio maps which have the potential to play a significant role in breast cancer detection. Advances in knowledge: This study presents the first evaluation of a multidimensional MR spectroscopic imaging technique for the detection of potentially novel biomarkers based on glycine, myo-inositol, and unsaturated fatty acids, in addition to commonly reported choline. Spatial mapping of choline and unsaturated fatty acid ratios with respect to water in malignant and benign breast masses are also shown. These metabolic characteristics may serve as additional biomarkers for improving the diagnostic and therapeutic evaluation of breast cancer.

Quantitative Predictors of Response to Neoadjuvant Chemotherapy on Dynamic Contrast-enhanced 3T Breast MRI.

OBJECTIVE: To assess whether changes in quantitative parameters on breast MRI better predict pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer than change in volume. METHODS: This IRB-approved retrospective study included women with newly diagnosed breast cancer who underwent 3T MRI before and during NAC from January 2013 to December 2019 and underwent surgery at our institution. Clinical data such as age, histologic diagnosis and grade, biomarker status, clinical stage, maximum index cancer dimension and volume, and surgical pathology (presence or absence of in-breast pCR) were collected. Quantitative parameters were calculated using software. Correlations between clinical features and MRI quantitative measures in pCR and non-pCR groups were assessed using univariate and multivariate logistic regression. RESULTS: A total of 182 women with a mean age of 52 years (range, 26-79 years) and 187 cancers were included. Approximately 45% (85/182) of women had pCR at surgery. Stepwise multivariate regression analysis showed statistical significance for changes in quantitative parameters (increase in time to peak and decreases in peak enhancement, wash out, and Kep [efflux rate constant]) for predicting pCR. These variables in combination predicted pCR with 81.2% accuracy and an area under the curve (AUC) of 0.878. The AUCs of change in index cancer volume and maximum dimension were 0.767 and 0.613, respectively. CONCLUSION: Absolute changes in quantitative MRI parameters between pre-NAC MRI and intra-NAC MRI could help predict pCR with excellent accuracy, which was greater than changes in index cancer volume and maximum dimension.

Fear of the Unknown: The Benefits of a Patient Educational Handout on Breast Biopsy Markers.

OBJECTIVE: To determine whether providing a biopsy marker informational handout to patients improves patient knowledge and comfort with receiving a marker. METHODS: In this IRB-exempt prospective study, a patient educational handout on breast biopsy markers was developed. A questionnaire was created with four questions asking patients to self-evaluate their knowledge of biopsy markers and their comfort level with marker placement before and after reading the handout. Technologists distributed the educational handouts to patients presenting for a percutaneous breast biopsy under any modality from December 11, 2020, to April 23, 2021. Data from the completed questionnaires were entered into a database. Statistical analyses included paired t-test and Wilcoxon analyses. RESULTS: In total, 141 completed surveys were included in the analysis. The mean scores prior to reading the handout for knowledge and comfort were 2.59 and 3.40, respectively. After reading the handout, there was a significant increase in mean scores for knowledge and comfort (4.26 and 4.20, respectively) (P < 0.001). There was a 64% increase vs 23% increase for knowledge and comfort, respectively. CONCLUSION: Patient-assessed knowledge of biopsy markers increased significantly after reading our educational handout. Patient-assessed comfort with biopsy marker placement also increased significantly after reading the educational handout, though to a lesser degree than knowledge. Although not included in our study, use of an educational handout may impact patient acceptance of marker placement. Future directions may include quantitatively assessing the effect of the handout on time to consent for a biopsy or influence on acceptance of marker placement.

BRCA and Beyond: Comprehensive Image-rich Review of Hereditary Breast and Gynecologic Cancer Syndromes.

In addition to the well-characterized BRCA1 and BRCA2 hereditary breast and ovarian cancer syndromes, many other syndromes that are associated with genetic mutations predispose individuals to an increased risk of breast and gynecologic malignancies. Many mutated genes encode for tumor-suppressor products and are inherited in an autosomal dominant manner. Mutations markedly increase an individual's lifetime risk of cancers in different organ systems, depending on the associated syndrome. These syndromes include Lynch syndrome, the most common hereditary cause of endometrial cancer, and Peutz-Jeghers syndrome, which increases the risks of breast cancer, ovarian cancer, and cervical adenoma malignum. Li-Fraumeni syndrome and Cowden syndrome increase the risk of breast cancer, and Gorlin syndrome increases the risk of ovarian fibromas. With advances in genetic testing, clinicians' knowledge and awareness of the numerous additional genes associated with breast and ovarian cancers, such as ATM, CHEK2, and PALB2, are rapidly expanding. Radiologists have essential roles in patient management, which include developing optimal screening protocols for these patients and closely monitoring them for the development or recurrence of disease-specific malignancies. Radiologists' roles continue to increase and evolve as more mutations are identified and high-risk imaging screening recommendations expand to identify these patients. Understanding the epidemiologic, genetic, and pathophysiologic features and the cancers associated with these syndromes enables radiologists to appropriately contribute to patient management, ensure accurate and timely diagnosis, and make syndrome-specific imaging recommendations. ©RSNA, 2020.

Abbreviated Breast MRI for Estimating Extent of Disease in Newly Diagnosed Breast Cancer.

OBJECTIVE: To evaluate extent of disease estimation of abbreviated protocol (ap) magnetic resonance imaging (MRI) compared with full protocol (fp) MRI in newly diagnosed breast cancer. METHODS: In this institutional review board-approved, Health Insurance Portability and Accountability Act-compliant, retrospective study of women with breast cancer who underwent pretreatment fpMRI on a 3 Tesla MRI in 2013, axial fat-saturated pre- and first postcontrast T1, maximum-intensity projection, and subtraction sequences were interpreted independently by three breast radiologists in two sessions, without and with prior imaging, respectively. Agreement was calculated using Cohen's kappa. Interpretations were compared with histology or clinical stability. Diagnostic performances were compared using Bennett's statistic. P < 0.05 was significant. RESULTS: Eighty-one women (mean age 56 years, range 32-92 years), 116 lesions, and 95 cancers (mean size 27 mm, range 4-110 mm) were included. Agreement among radiologists for lesion assessment was excellent (0.83). apMRI cancer detection improved with prior imaging (mean sensitivity from 95% to 99%, specificity from 91% to 97%, positive predictive value [PPV] from 92% to 98%, and negative predictive value [NPV] from 95% to 99%) versus fpMRI (sensitivity 98% [93/95], specificity 94% [76/81], PPV 95% [93/98], and NPV 97% [76/78]). apMRI detected all multifocal, multicentric, and contralateral disease seen in 19% (15/81) of women to the same extent as fpMRI. apMRI axillary metastases detection improved with prior imaging (mean sensitivity from 78% to 86%, specificity from 90% to 92%, PPV from 76% to 82%, and NPV from 89% to 94%) versus fpMRI (sensitivity 71% [17/24], specificity 88% [51/58]), PPV 71% [17/24], and NPV 88% [51/58]). CONCLUSION: apMRI may be acceptable for women with newly diagnosed cancer.

Musculoskeletal interventional ultrasound.

Ultrasound is a nonionizing, low-cost, portable imaging technique for the evaluation of tendons, muscles, joints, soft tissue masses, and cysts, especially in patients unable to tolerate computed tomography or magnetic resonance imaging. These advantages make ultrasound an ideal modality for guiding musculoskeletal interventions. Its real-time capabilities allow continuous observation of needle placement into the targeted area and direct visualization of interventions such as injection of medication while avoiding other soft tissue structures or nearby neurovascular bundles. After a brief overview of the technical factors involved in performing ultrasound-guided musculoskeletal interventions, this article reviews commonly performed percutaneous procedures in the musculoskeletal system.