Differentiating axillary lymph node metastasis in invasive breast cancer patients: A comparison of radiomic signatures from multiparametric breast MR sequences

A two-center study of a combined nomogram based on mammography and MRI to predict ALN metastasis in breast cancer

OBJECTIVES

To develop and validate a predictive method for axillary lymph node (ALN) metastasis of breast cancer by using radiomics based on mammography and MRI.

MATERIALS AND METHODS

A retrospective analysis of 492 women from center 1 (The affiliated Hospital of Qingdao University) and center 2 (Yantai Yuhuangding Hospital) with primary breast cancer from August 2013 to May 2021 was carried out. The radscore was calculated using the features screened based on preoperative mammography and MRI from the training cohort of Center 1 (n = 231), then tested in the validation cohort (n = 99), an internal test cohort (n = 90) from Center 1, and an external test cohort (n = 72) from Center 2. Univariate and multivariate analyses were used to screen for the clinical and radiological characteristics most associated with ALN metastasis. A combined nomogram was established in combination with radscore that predicted the clinicopathological and radiological characteristics. Calibration curves were used to test the effectiveness of the combined nomogram. The receiver operating characteristic (ROC) curve was used to evaluate the performance of the combined nomogram and then compare with the clinical and radiomic models. The decision curve analysis (DCA) value was used to evaluate the combined nomogram for clinical applications.

RESULTS

The constructed combined nomogram incorporating the radscore and MRI-reported ALN metastasis status exhibited good calibration and outperformed the radiomics signatures in predicting ALN metastasis (area under the curve [AUC]: 0.886 vs. 0.846 in the training cohort; 0.826 vs. 0.762 in the validation cohort; 0.925 vs. 0.899 in the internal test cohort; and 0.902 vs. 0.793 in the external test cohort). The combination nomogram achieved a higher AUC in the training cohort (0.886 vs. 0.786) and the internal test cohort (0.925 vs. 0.780) and similar AUCs in the validation (0.826 vs. 0.811) and external test (0.902 vs. 0.837) cohorts than the clinical model.

CONCLUSION

A combined nomogram based on mammography and MRI can be used for preoperative prediction of ALN metastasis in primary breast cancer.

Clinical study on the prediction of ALN metastasis based on intratumoral and peritumoral DCE-MRI radiomics and clinico-radiological characteristics in breast cancer

Objective

To investigate the value of predicting axillary lymph node (ALN) metastasis based on intratumoral and peritumoral dynamic contrast-enhanced MRI (DCE-MRI) radiomics and clinico-radiological characteristics in breast cancer.

Methods

A total of 473 breast cancer patients who underwent preoperative DCE-MRI from Jan 2017 to Dec 2020 were enrolled. These patients were randomly divided into training (n=378) and testing sets (n=95) at 8:2 ratio. Intratumoral regions (ITRs) of interest were manually delineated, and peritumoral regions of 3 mm (3 mmPTRs) were automatically obtained by morphologically dilating the ITR. Radiomics features were extracted, and ALN metastasis-related radiomics features were selected by the Mann-Whitney U test, Z score normalization, variance thresholding, K-best algorithm and least absolute shrinkage and selection operator (LASSO) algorithm. Clinico-radiological risk factors were selected by logistic regression and were also used to construct predictive models combined with radiomics features. Then, 5 models were constructed, including ITR, 3 mmPTR, ITR+3 mmPTR, clinico-radiological and combined (ITR+3 mmPTR+ clinico-radiological) models. The performance of models was assessed by sensitivity, specificity, accuracy, F1 score and area under the curve (AUC) of receiver operating characteristic (ROC), calibration curves and decision curve analysis (DCA).

Results

A total of 2264 radiomics features were extracted from each region of interest (ROI), 3 and 10 radiomics features were selected for the ITR and 3 mmPTR, respectively. 5 clinico-radiological risk factors were selected, including lesion size, human epidermal growth factor receptor 2 (HER2) expression, vascular cancer thrombus status, MR-reported ALN status, and time-signal intensity curve (TIC) type. In the testing set, the combined model showed the highest AUC (0.839), specificity (74.2%), accuracy (75.8%) and F1 Score (69.3%) among the 5 models. DCA showed that it had the greatest net clinical benefit compared to the other models.

Conclusion

The intra- and peritumoral radiomics models based on DCE-MRI could be used to predict ALN metastasis in breast cancer, especially for the combined model with clinico-radiological characteristics showing promising clinical application value.

Deep learning radiomics based prediction of axillary lymph node metastasis in breast cancer

This study aimed to develop and validate a deep learning radiomics nomogram (DLRN) for the preoperative evaluation of axillary lymph node (ALN) metastasis status in patients with a newly diagnosed unifocal breast cancer. A total of 883 eligible patients with breast cancer who underwent preoperative breast and axillary ultrasound were retrospectively enrolled between April 1, 2016, and June 30, 2022. The training cohort comprised 621 patients from Hospital I; the external validation cohorts comprised 112, 87, and 63 patients from Hospitals II, III, and IV, respectively. A DLR signature was created based on the deep learning and handcrafted features, and the DLRN was then developed based on the signature and four independent clinical parameters. The DLRN exhibited good performance, yielding areas under the receiver operating characteristic curve (AUC) of 0.914, 0.929, and 0.952 in the three external validation cohorts, respectively. Decision curve and calibration curve analyses demonstrated the favorable clinical value and calibration of the nomogram. In addition, the DLRN outperformed five experienced radiologists in all cohorts. This has the potential to guide appropriate management of the axilla in patients with breast cancer, including avoiding overtreatment.

The application of radiomics machine learning models based on multimodal MRI with different sequence combinations in predicting cervical lymph node metastasis in oral tongue squamous cell carcinoma patients

BACKGROUND

The purpose of this study was to explore preliminary the performance of radiomics machine learning models based on multimodal MRI to predict the risk of cervical lymph node metastasis (CLNM) for oral tongue squamous cell carcinoma (OTSCC) patients.

METHODS

A total of 400 patients were enrolled in this study and divided into six groups according to the different combinations of MRI sequences. Group I consisted of patients with T1-weighted images (T1WI) and FS-T2WI (fat-suppressed T2-weighted images), group II consisted of patients with T1WI, FS-T2WI, and contrast enhanced MRI (CE-MRI), group III consisted of patients with T1WI, FS-T2WI, and T2-weighted images (T2WI), group IV consisted of patients with T1WI, FS-T2WI, CE-MRI, and T2WI, group V consisted of patients with T1WI, FS-T2WI, T2WI, and apparent diffusion coefficient map (ADC), and group VI consisted of patients with T1WI, FS-T2WI, CE-MRI, T2WI, and ADC. Machine learning models were constructed. The performance of the models was compared in each group.

RESULTS

The machine learning model in group IV including T1WI, FS-T2WI, T2WI, and CE-MRI presented best prediction performance, with AUCs of 0.881 and 0.868 in the two sets. The models with CE-MRI performed better than the models without CE-MRI(I vs. II, III vs. IV, V vs. VI).

CONCLUSIONS

The radiomics machine learning models based on CE-MRI showed great accuracy and stability in predicting the risk of CLNM for OTSCC patients.

Pre- and Post-treatment Double-Sequential-Point Dynamic Radiomic Model in the Response Prediction of Gastric Cancer to Neoadjuvant Chemotherapy: 3-Year Survival Analysis

BACKGROUND

Prognosis prediction of patients with gastric cancer after neoadjuvant chemotherapy is suboptimal. This study aims to develop and validate a dynamic radiomic model for prognosis prediction of patients with gastric cancer on the basis of baseline and posttreatment features.

PATIENTS AND METHODS

This single-center cohort study included patients with gastric adenocarcinoma treated with neoadjuvant chemotherapy from June 2009 to July 2015 in the Gastrointestinal Cancer Center of Peking University Cancer Hospital. Their clinicopathological data, pre-treatment and post-treatment computed tomography (CT) images, and pathological reports were retrieved and analyzed. Four prediction models were developed and validated using tenfold cross-validation, with death within 3 years as the outcome. Model discrimination was compared by the area under the curve (AUC). The final radiomic model was evaluated for calibration and clinical utility using Hosmer-Lemeshow tests and decision curve analysis.

RESULTS

The study included 205 patients with gastric adenocarcinoma [166 (81%) male; mean age 59.9 (SD 10.3) years], with 71 (34.6%) deaths occurring within 3 years. The radiomic model alone demonstrated better discrimination than the pathological T stage (ypT) stage model alone (cross-validated AUC 0.598 versus 0.516, P = 0.009). The final radiomic model, which incorporated both radiomic and clinicopathological characteristics, had a significantly higher cross-validated AUC (0.769) than the ypT stage model (0.516), the radiomics alone model (0.598), and the ypT plus other clinicopathological characteristics model (0.738; all P < 0.05). Decision curve analysis confirmed the clinical utility of the final radiomic model.

CONCLUSIONS

The developed radiomic model had good accuracy and could be used as a decision aid tool in clinical practice to differentiate prognosis of patients with gastric cancer.

Assessment of Background Parenchymal Enhancement at Dynamic Contrast-enhanced MRI in Predicting Breast Cancer Recurrence Risk

Background Background parenchymal enhancement (BPE) at dynamic contrast-enhanced (DCE) MRI of cancer-free breasts increases the risk of developing breast cancer; implications of quantitative BPE in ipsilateral breasts with breast cancer are largely unexplored. Purpose To determine whether quantitative BPE measurements in one or both breasts could be used to predict recurrence risk in women with breast cancer, using the Oncotype DX recurrence score as the reference standard. Materials and Methods This HIPAA-compliant retrospective single-institution study included women diagnosed with breast cancer between January 2007 and January 2012 (development set) and between January 2012 and January 2017 (internal test set). Quantitative BPE was automatically computed using an in-house-developed computer algorithm in both breasts. Univariable logistic regression was used to examine the association of BPE with Oncotype DX recurrence score binarized into high-risk (recurrence score >25) and low- or intermediate-risk (recurrence score ≤25) categories. Models including BPE measures were assessed for their ability to distinguish patients with high risk versus those with low or intermediate risk and the actual recurrence outcome. Results The development set included 127 women (mean age, 58 years ± 10.2 [SD]; 33 with high risk and 94 with low or intermediate risk) with an actual local or distant recurrence rate of 15.7% (20 of 127) at a minimum 10 years of follow-up. The test set included 60 women (mean age, 57.8 years ± 11.6; 16 with high risk and 44 with low or intermediate risk). BPE measurements quantified in both breasts were associated with increased odds of a high-risk Oncotype DX recurrence score (odds ratio range, 1.27-1.66 [95% CI: 1.02, 2.56]; P < .001 to P = .04). Measures of BPE combined with tumor radiomics helped distinguish patients with a high-risk Oncotype DX recurrence score from those with a low- or intermediate-risk score, with an area under the receiver operating characteristic curve of 0.94 in the development set and 0.79 in the test set. For the combined models, the negative predictive values were 0.97 and 0.93 in predicting actual distant recurrence and local recurrence, respectively. Conclusion Ipsilateral and contralateral DCE MRI measures of BPE quantified in patients with breast cancer can help distinguish patients with high recurrence risk from those with low or intermediate recurrence risk, similar to Oncotype DX recurrence score. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Zhou and Rahbar in this issue.

Update on DWI for Breast Cancer Diagnosis and Treatment Monitoring

DWI is a noncontrast MRI technique that measures the diffusion of water molecules within biologic tissue. DWI is increasingly incorporated into routine breast MRI examinations. Currently, the main applications of DWI are breast cancer detection and characterization, prognostication, and prediction of treatment response to neoadjuvant chemotherapy. In addition, DWI is promising as a noncontrast MRI alternative for breast cancer screening. Problems with suboptimal resolution and image quality have restricted the mainstream use of DWI for breast imaging, but these shortcomings are being addressed through several technologic advancements. In this review, we present an up-to-date assessment of the use of DWI for breast cancer imaging, including a summary of the clinical literature and recommendations for future use.

The use of radiomics in magnetic resonance imaging for the pre-treatment characterisation of breast cancers: A scoping review

Radiomics is an emerging field that aims to extract and analyse a comprehensive set of quantitative features from medical images. This scoping review is focused on MRI-based radiomic features for the molecular profiling of breast tumours and the implications of this work for predicting patient outcomes. A thorough systematic literature search and outcome extraction were performed to identify relevant studies published in MEDLINE/PubMed (National Centre for Biotechnology Information), EMBASE and Scopus from 2015 onwards. The following information was retrieved from each article: study purpose, study design, extracted radiomic features, machine learning technique(s), sample size/characteristics, statistical result(s) and implications on patient outcomes. Based on the study purpose, four key themes were identified in the included 63 studies: tumour subtype classification (n = 35), pathologically complete response (pCR) prediction (n = 15), lymph node metastasis (LNM) detection (n = 7) and recurrence rate prediction (n = 6). In all four themes, reported accuracies widely varied among the studies, for example, area under receiver characteristics curve (AUC) for detecting LNM ranged from 0.72 to 0.91 and the AUC for predicting pCR ranged from 0.71 to 0.99. In all four themes, combining radiomic features with clinical data improved the predictive models. Preliminary results of this study showed radiomics potential to characterise the whole tumour heterogeneity, with clear implications for individual-targeted treatment. However, radiomics is still in the pre-clinical phase, currently with an insufficient number of large multicentre studies and those existing studies are often limited by insufficient methodological transparency and standardised workflow. Consequently, the clinical translation of existing studies is currently limited.

Peritumoral and Intratumoral Texture Features Based on Multiparametric MRI and Multiple Machine Learning Methods to Preoperatively Evaluate the Pathological Outcomes of Pancreatic Cancer

BACKGROUND

Radiomics-based preoperative evaluation of lymph node metastasis (LNM) and histological grade (HG) might facilitate the decision-making for pancreatic cancer and further efforts are needed to develop effective models.

PURPOSE

To develop multiparametric MRI (MP-MRI)-based radiomics models to evaluate LNM and HG.

STUDY TYPE

Retrospective.

POPULATION

The pancreatic cancer patients from the main center (n = 126) were assigned to the training and validation sets at a 4:1 ratio. The patients from the other center (n = 40) served as external test sets.

FIELD STRENGTH/SEQUENCE

A 3.0 T and 1.5 T/T2-weighted imaging, diffusion-weighted imaging, and dynamic contrast enhancement T1-weighted imaging.

ASSESSMENT

A total of 10,686 peritumoral and intratumoral radiomics features were extracted which contained first-order, shape-based, and texture features. The following three-step method was applied to reduce the feature dimensionality: SelectKBest (a function from scikit-learn package), least absolute shrinkage and selection operator (LASSO), and recursive feature elimination based on random forest (RFE-RF). Six classifiers (random forest, logistic regression, support vector machine, K-nearest neighbor, decision tree, and XGBOOST) were trained and selected based on their performance to construct the clinical, radiomics, and combination models.

STATISTICAL TESTS

Delong's test was used to compare the models' performance. P value less than 0.05 was considered significant.

RESULTS

Twelve significant features for LNM and 11 features for HG were obtained. Random forest and logistic regression performed better than the other classifiers in evaluating LNM and HG, respectively, according to the surgical pathological results. The best performance was obtained with the models that combined peritumoral and intratumoral features with area under curve (AUC) values of 0.944 and 0.892 in the validation and external test sets for HG and 0.924 and 0.875 for LNM.

DATA CONCLUSION

Radiomics holds the potential to evaluate LNM and HG of pancreatic cancer. The combination of peritumoral and intratumoral features will make models more accurate.

EVIDENCE LEVEL

4.

TECHNICAL EFFICACY

Stage 2.

Deep Learning Radiomics of Preoperative Breast MRI for Prediction of Axillary Lymph Node Metastasis in Breast Cancer

The objective of this study is to develop a radiomic signature constructed from deep learning features and a nomogram for prediction of axillary lymph node metastasis (ALNM) in breast cancer patients. Preoperative magnetic resonance imaging data from 479 breast cancer patients with 488 lesions were studied. The included patients were divided into two cohorts by time (training/testing cohort, n = 366/122). Deep learning features were extracted from diffusion-weighted imaging-quantitatively measured apparent diffusion coefficient (DWI-ADC) imaging and dynamic contrast-enhanced MRI (DCE-MRI) by a pretrained neural network of DenseNet121. After the selection of both radiomic and clinicopathological features, deep learning signature and a nomogram were built for independent validation. Twenty-three deep learning features were automatically selected in the training cohort to establish the deep learning signature of ALNM. Three clinicopathological factors, including LN palpability (odds ratio (OR) = 6.04; 95% confidence interval (CI) = 3.06-12.54, P = 0.004), tumor size in MRI (OR = 1.45, 95% CI = 1.18-1.80, P = 0.104), and Ki-67 (OR = 1.01; 95% CI = 1.00-1.02, P = 0.099), were selected and combined with radiomic signature to build a combined nomogram. The nomogram showed excellent predictive ability for ALNM (AUC 0.80 and 0.71 in training and testing cohorts, respectively). The sensitivity, specificity, and accuracy were 65%, 80%, and 75%, respectively, in the testing cohort. MRI-based deep learning radiomics in patients with breast cancer could be used to predict ALNM, providing a noninvasive approach to structuring the treatment strategy.

Consistent Learning-Based Breast Tumor Segmentation and Its Application in Sentinel Lymph Node Metastasis Prediction

Accurate staging of lymph nodes provides crucial diagnostic information for breast cancer patients, where segmentation is of great importance by localizing and visualizing the breast tumor of interest. Nevertheless, current segmentation methods perform average when facing large span of tumor sizes, degraded image quality, blurred tumor boundaries, and resulting noise during manual annotation. Therefore, we develop a Multi-scale RepVGG-based Segmentation Network (MPSegNet) to segment breast tumor from MR images. In particular, we construct a consistent learning framework for the MPSegNet to alleviate the impact of noisy labels upon segmentation results. The rationale is that different views covering the same breast tumors are supposed to generate identical segmentation predictions. Then, we predict SLN metastasis given segmented breast tumors, where we evaluate the relationships between the predictive performance and tumor segmentations under different consistencies. The results show the superiority of our method over other state-of-the-art methods. A high consistency among multiple views can boost the segmentation performance during consistent learning. However, the optimal segmentation does not produce the best SLN metastatic prediction results, implying that the dependence of classification upon segmentation needs to be elaborately investigated further.Clinical Relevance- This study facilitates more accurate segmentation of breast tumors with consistent learning, and provides an initial analysis between tumor segmentation and subsequent prediction of SLN metastasis, which has potential significance for the precise medical care of breast cancer patients.

Abbreviated MRI for Comprehensive Regional Lymph Node Staging during Pre-Operative Breast MRI

Background: The detection of regional lymph node metastases (LNM), in particular significant LNM (≥N2), is important to guide treatment decisions in women with breast cancer. The purpose of this study was to determine whether a coronal pulse sequence as part of pre-operative breast MRI is useful to identify women without significant LNM. Material: Retrospective study between January 2017 and December 2019 on 414 consecutive women with breast cancer who underwent pre-operative breast MRI on a 1.5 T system. For lymph node (LN) staging, a coronal pre-contrast non-fat-suppressed T1-weighted TSE sequence was acquired with the system’s built-in body coil, covering the chest wall; acquisition time 3:12 min. Two radiologists rated the likelihood of LNM on a 3-point scale (absent/possible/present). Validation was obtained by histology from sentinel LN biopsy, axillary LN dissection, and/or PET/CT. Results: 368/414 women were staged to have no or non-significant LNM (pN0 in 282/414, pN1 in 86/414), and significant LNM (≥pN2) in 46/414. For identification of women with significant LNM, MRI was true-positive in 42/46, false-negative in 4/46, true-negative in 327/368, and false-positive in 41/83, the latter mostly caused by women with N1-disease (38/41), yielding an NPV and PPV for significant LNM of 98.8% [95%-CI: 97.0–100%] and 50.6% [43.1–58.1%], respectively. Conclusions: A 3 min coronal T1-weighted pulse sequence covering the chest wall as part of pre-operative breast MRI is useful to rule out significant LNM with high NPV. Where MRI staging is positive for significant LNM, additional work-up is indicated to improve the distinction of N1 and N2 disease.

Radiomics nomogram for predicting axillary lymph node metastasis in breast cancer based on DCE-MRI: A multicenter study

OBJECTIVES

This study aims to develop and validate a radiomics nomogram based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to noninvasively predict axillary lymph node (ALN) metastasis in breast cancer.

METHODS

This retrospective study included 263 patients with histologically proven invasive breast cancer and who underwent DCE-MRI examination before surgery in two hospitals. All patients had a defined ALN status based on pathological examination results. Regions of interest (ROIs) of the primary tumor and ipsilateral ALN were manually drawn. A total of 1,409 radiomics features were initially computed from each ROI. Next, the low variance threshold, SelectKBest, and least absolute shrinkage and selection operator (LASSO) algorithms were used to extract the radiomics features. The selected radiomics features were used to establish the radiomics signature of the primary tumor and ALN. A radiomics nomogram model, including the radiomics signature and the independent clinical risk factors, was then constructed. The predictive performance was evaluated by the receiver operating characteristic (ROC) curves, calibration curve, and decision curve analysis (DCA) by using the training and testing sets.

RESULTS

ALNM rates of the training, internal testing, and external testing sets were 43.6%, 44.3% and 32.3%, respectively. The nomogram, including clinical risk factors (tumor diameter) and radiomics signature of the primary tumor and ALN, showed good calibration and discrimination with areas under the ROC curves of 0.884, 0.822, and 0.813 in the training, internal and external testing sets, respectively. DCA also showed that radiomics nomogram displayed better clinical predictive usefulness than the clinical or radiomics signature alone.

CONCLUSIONS

The radiomics nomogram combined with clinical risk factors and DCE-MRI-based radiomics signature may be used to predict ALN metastasis in a noninvasive manner.

CNN-Based Approaches with Different Tumor Bounding Options for Lymph Node Status Prediction in Breast DCE-MRI

Simple Summary

Breast cancer represents the most frequent cancer in women in the world. The state of the axillary lymph node is considered an independent prognostic factor and is currently evaluated only with invasive methods. Deep learning approaches, especially the ones based on convolutional neural networks, offer a valid, non-invasive alternative, allowing extraction of large amounts of the quantitative data that are used to build predictive models. The aim of our work is to evaluate the influence of the peritumoral parenchyma through different bounding box techniques on the prediction of the axillary lymph node in breast cancer patients using a deep learning artificial intelligence approach.

Abstract

Background: The axillary lymph node status (ALNS) is one of the most important prognostic factors in breast cancer (BC) patients, and it is currently evaluated by invasive procedures. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), highlights the physiological and morphological characteristics of primary tumor tissue. Deep learning approaches (DL), such as convolutional neural networks (CNNs), are able to autonomously learn the set of features directly from images for a specific task. Materials and Methods: A total of 155 malignant BC lesions evaluated via DCE-MRI were included in the study. For each patient’s clinical data, the tumor histological and MRI characteristics and axillary lymph node status (ALNS) were assessed. LNS was considered to be the final label and dichotomized (LN+ (27 patients) vs. LN− (128 patients)). Based on the concept that peritumoral tissue contains valuable information about tumor aggressiveness, in this work, we analyze the contributions of six different tumor bounding options to predict the LNS using a CNN. These bounding boxes include a single fixed-size box (SFB), a single variable-size box (SVB), a single isotropic-size box (SIB), a single lesion variable-size box (SLVB), a single lesion isotropic-size box (SLIB), and a two-dimensional slice (2DS) option. According to the characteristics of the volumes considered as inputs, three different CNNs were investigated: the SFB-NET (for the SFB), the VB-NET (for the SVB, SIB, SLVB, and SLIB), and the 2DS-NET (for the 2DS). All the experiments were run in 10-fold cross-validation. The performance of each CNN was evaluated in terms of accuracy, sensitivity, specificity, the area under the ROC curve (AUC), and Cohen’s kappa coefficient (K). Results: The best accuracy and AUC are obtained by the 2DS-NET (78.63% and 77.86%, respectively). The 2DS-NET also showed the highest specificity, whilst the highest sensibility was attained by the VB-NET based on the SVB and SIB as bounding options. Conclusion: We have demonstrated that a selective inclusion of the DCE-MRI’s peritumoral tissue increases accuracy in the lymph node status prediction in BC patients using CNNs as a DL approach.

Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer Using CNN Based on Multiparametric MRI

BACKGROUND

Multiparametric magnetic resonance imaging (MRI) is widely used in breast cancer screening. Accurate prediction of the axillary lymph nodes metastasis (ALNM) is essential for breast cancer surgery and treatment. However, there is no mature and effective discerning method for ALNM based on multiparametric MRI.

PURPOSE

To evaluate the ALNM using T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and diffusion-weighted imaging (DWI) sequences, respectively, and construct a quantitative ALNM discerning model of integrated multiparametric MRI.

STUDY TYPE

Retrospective.

POPULATION

Three-hundred forty-eight breast cancer patients, 163 with ALNM (99.39% females), and 185 without ALNM (100% females). The dataset was randomly divided into the training set (315 cases) and the testing set (33 cases).

FIELD STRENGTH/SEQUENCE

1.5 T; T1WI (VIBRANT), T2WI (FSE), and DWI (echo planar imaging [EPI]).

ASSESSMENT

The lesion region of interest images were cropped and sent to a pretrained ResNet50 network. Then, the results of different sequences were sent to a classifier for ensemble learning to construct the ALNM model of multiparametric MRI.

STATISTICAL TESTS

Performance indicators such as accuracy, the receiver operating characteristic (ROC) curve, and the area under the ROC curve (AUC) were calculated. Student's t-test, chi-square test, Fisher's exact test, and Delong test were performed, and P < 0.05 was considered statistically significant.

RESULTS

T2WI performed the best among the three sequences, and achieved the accuracy and AUC of 0.933/0.989 in the testing set. Compared to T1WI with the accuracy and AUC of 0.691/0.806, the increase is significant. While compared to DWI with the accuracy and AUC of 0.800/0.910, the improvement is not significant (P = 0.126). After integrating three sequences, the accuracy and AUC improved to 0.970 and 0.996.

DATA CONCLUSION

T2WI performed better than DWI and T1WI in discerning ALNM in this breast cancer dataset. The proposed quantitative model of integrated multiparametric MRI could effectively help the ALNM diagnosis.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY STAGE: 2.

Beyond N Staging in Breast Cancer: Importance of MRI and Ultrasound-based Imaging

The correct N-staging in breast cancer is crucial to tailor treatment and stratify the prognosis. N-staging is based on the number and the localization of suspicious regional nodes on physical examination and/or imaging. Since clinical examination of the axillary cavity is associated with a high false negative rate, imaging modalities play a central role. In the presence of a T1 or T2 tumor and 0–2 suspicious nodes, on imaging at the axillary level I or II, a patient should undergo sentinel lymph node biopsy (SLNB), whereas in the presence of three or more suspicious nodes at the axillary level I or II confirmed by biopsy, they should undergo axillary lymph node dissection (ALND) or neoadjuvant chemotherapy according to a multidisciplinary approach, as well as in the case of internal mammary, supraclavicular, or level III axillary involved lymph nodes. In this scenario, radiological assessment of lymph nodes at the time of diagnosis must be accurate. False positives may preclude a sentinel lymph node in an otherwise eligible woman; in contrast, false negatives may lead to an unnecessary SLNB and the need for a second surgical procedure. In this review, we aim to describe the anatomy of the axilla and breast regional lymph node, and their diagnostic features to discriminate between normal and pathological nodes at Ultrasound (US) and Magnetic Resonance Imaging (MRI). Moreover, the technical aspects, the advantage and limitations of MRI versus US, and the possible future perspectives are also analyzed, through the analysis of the recent literature.

Radiomic Signature Based on Dynamic Contrast-Enhanced MRI for Evaluation of Axillary Lymph Node Metastasis in Breast Cancer

Background. To construct and validate a radiomic-based model for estimating axillary lymph node (ALN) metastasis in patients with breast cancer by dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Methods. In this retrospective study, a radiomic-based model was established in a training cohort of 236 patients with breast cancer. Radiomic features were extracted from breast DCE-MRI scans. A method named the least absolute shrinkage and selection operator (LASSO) was applied to select radiomic features based on highly reproducible features. A radiomic signature was built by a support vector machine (SVM). Multivariate logistic regression analysis was adopted to establish a clinical characteristic-based model. The performance of models was analysed through discrimination ability and clinical benefits. Results. The radiomic signature comprised 6 features related to ALN metastasis and showed significant differences between the patients with ALN metastasis and without ALN metastasis ( $P<0.001$ ). The area under the curve (AUC) of the radiomic model was 0.990 and 0.858, respectively, in the training and validation sets. The clinical feature-based model, including MRI-reported status and palpability, performed slightly worse, with an AUC of 0.784 in the training cohort and 0.789 in the validation cohort. The radiomic signature was confirmed to provide more clinical benefits by decision curve analysis. Conclusions. The radiomic-based model developed in this study can successfully diagnose the status of lymph nodes in patients with breast cancer, which may reduce unnecessary invasive clinical operations.

Diffusion Breast MRI: Current Standard and Emerging Techniques

The role of diffusion weighted imaging (DWI) as a biomarker has been the subject of active investigation in the field of breast radiology. By quantifying the random motion of water within a voxel of tissue, DWI provides indirect metrics that reveal cellularity and architectural features. Studies show that data obtained from DWI may provide information related to the characterization, prognosis, and treatment response of breast cancer. The incorporation of DWI in breast imaging demonstrates its potential to serve as a non-invasive tool to help guide diagnosis and treatment. In this review, current technical literature of diffusion-weighted breast imaging will be discussed, in addition to clinical applications, advanced techniques, and emerging use in the field of radiomics.

Radiomics Based on DCE-MRI Improved Diagnostic Performance Compared to BI-RADS Analysis in Identifying Sclerosing Adenosis of the Breast

Purpose

Sclerosing adenosis (SA) is a benign lesion that could mimic breast carcinoma and be evaluated as malignancy by Breast Imaging-Reporting and Data System (BI-RADS) analysis. We aimed to construct and validate the performance of radiomic model based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) compared to BI-RADS analysis to identify SA.

Methods

Sixty-seven patients with invasive ductal carcinoma (IDC) and 58 patients with SA were included in this retrospective study from two institutions. The 125 patients were divided into a training cohort (n= 88) from institution I and a validation cohort from institution II (n=37). Dynamic contrast-enhanced sequences including one pre-contrast and five dynamic post-contrast series were obtained for all cases with different 3T scanners. Single-phase enhancement, multi-phase enhancement, and dynamic radiomic features were extracted from DCE-MRI. The least absolute shrinkage and selection operator (LASSO) logistic regression and cross-validation was performed to build the radscore of each single-phase enhancement and the final model combined multi-phase and dynamic radiomic features. The diagnostic performance of radiomics was evaluated by receiver operating characteristic (ROC) analysis and compared to the performance of BI-RADS analysis. The classification performance was tested using external validation.

Results

In the training cohort, the AUCs of BI-RADS analysis were 0.71 (95%CI [0.60, 0.80]), 0.78 (95%CI [0.67, 0.86]), and 0.80 (95%CI [0.70, 0.88]), respectively. In single-phase analysis, the second enhanced phase radiomic signature achieved the highest AUC of 0.88 (95%CI [0.79, 0.94]) in distinguishing SA from IDC. Nine multi-phase radiomic features and two dynamic radiomic features showed the best predictive ability for final model building. The final model improved the AUC to 0.92 (95%CI [0.84, 0.97]), and showed statistically significant differences with BI-RADS analysis (p<0.05 for all). In the validation cohort, the AUC of the final model was 0.90 (95%CI [0.75, 0.97]), which was higher than all BI-RADS analyses and showed statistically significant differences with one of the BI-RADS analysis observers (p = 0.03).

Conclusions

Radiomics based on DCE-MRI could show better diagnostic performance compared to BI-RADS analysis in differentiating SA from IDC, which may contribute to clinical diagnosis and treatment.

A Nomogram Based on Molecular Biomarkers and Radiomics to Predict Lymph Node Metastasis in Breast Cancer

Background

The aim of this study was to explore the feasibility and efficacy of a non-invasive quantitative imaging evaluation model to assess the lymphatic metastasis of breast cancer based on a radiomics signature constructed using conventional T1-weighted image (T1WI) enhanced MRI and molecular biomarkers.

Methods

Patients with breast cancer diagnosed via lymph biopsies between June 2015 and June 2019 were selected for the study. All patients underwent T1WI contrast-enhancement before treatment; lymph biopsy after surgery; and simultaneous Ki-67, COX-2, PR, Her2 and proliferating cell nuclear antigen detection. All images were imported into ITK-SNAP for whole tumor delineation, and AK software was used for radiomics feature extraction. Next, the radiomics signature Rad-score was constructed after reduction of specific radiomic features. A multiple regression logistic model was built by combining the Rad-score and molecular biomarkers based on the minimum AIC.

Results

In all, 100 patients were enrolled in this study, including 45 with non-lymph node (LN) metastasis and 55 with LN metastasis. A total of 1,051 texture feature parameters were extracted, and LASSO was used to reduce the dimensionality of the radiomics features. The log(λ) was set to 0.002786, and 19 parameters were retained for the construction of the radiomics tag Rad-score. ROC was used to evaluate the diagnostic efficiency of Rad-score: the area under the ROC curve (AUC) of the Rad-score for identifying non-lymphatic and lymphatic metastases was 0.891 in the training cohort and 0.744 in the validation cohort. With the incorporation of tumor molecular markers, the AUCs of the training cohort and validation cohort of the nomogram were 0.936 and 0.793, respectively, which were notably higher than the AUCs of the clinical parameters in the training and validation cohorts (0.719 and 0.588, respectively).

Conclusion

The combined model constructed using the Rad-score and molecular biomarkers can be used as an effective non-invasive method to assess LN metastasis of breast cancer. Furthermore, it can be used to quantitatively evaluate the risk of breast cancer LN metastasis before surgery.

Prediction of Axillary Lymph Node Metastasis in Breast Cancer using Intra-peritumoral Textural Transition Analysis based on Dynamic Contrast-enhanced Magnetic Resonance Imaging

RATIONALE AND OBJECTIVES

Intra-peritumoural textural transition (Ipris) is a new radiomics method, which includes a series of quantitative measurements of the image features that represent the differences between the inside and outside of the tumour. This study aimed to explore the feasibility of Ipris analysis for the preoperative prediction of axillary lymph node (ALN) status in patients with breast cancer based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).

MATERIALS AND METHODS

This study was approved by the Institutional Review Board (IRB) of our hospital. One hundred sixty-six patients with clinicopathologically confirmed invasive breast cancer and ALN status were enrolled. All patients underwent preoperative breast DCE-MRI examinations. The primary breast lesion was manually segmented using the ITK-SNAP software for each patient. Two sets of image features were extracted, including Ipris features and conventional intratumoural features. Feature selection was conducted using Spearman correlation analysis and support vector machine with recursive feature elimination (SVM-RFE). Next, three models were established in training dataset: Model 1 was established by Ipris features; Model 2 was established by intratumoural features; Model 3 was established by combining Ipris features and intratumoural features. The performances of the three models were evaluated for the prediction of ALN status in testing datasets.

RESULTS

Model 1 with four Ipris features achieved an AUC of 0.816 (95% CI, 0.733-0.883) and 0.829 (95% CI, 0.695-0.922) in the training and testing datasets, respectively. Model 2 with six intratumoural features achieved an AUC of 0.801 (95% CI, 0.716-0.870) and 0.824 (95% CI, 0.689-0.918) in the training and testing datasets, respectively. By incorporating the Ipris and intratumoural features, the AUC of Model 3 increased to 0.968 (95% CI, 0.916-0.992) and 0.855 (95% CI, 0.724-0.939) in the training and testing datasets, respectively.

CONCLUSION

Ipris features based on DCE-MRI can be used to predict ALN status in patients with breast cancer. The model combining intratumoural and Ipris features achieved higher prediction performance.

Development and Internal Validation of a Preoperative Prediction Model for Sentinel Lymph Node Status in Breast Cancer: Combining Radiomics Signature and Clinical Factors

Purpose

To develop and internally validate a nomogram combining radiomics signature of primary tumor and fibroglandular tissue (FGT) based on pharmacokinetic dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and clinical factors for preoperative prediction of sentinel lymph node (SLN) status in breast cancer patients.

Methods

This study retrospectively enrolled 186 breast cancer patients who underwent pretreatment pharmacokinetic DCE-MRI with positive (n = 93) and negative (n = 93) SLN. Logistic regression models and radiomics signatures of tumor and FGT were constructed after feature extraction and selection. The radiomics signatures were further combined with independent predictors of clinical factors for constructing a combined model. Prediction performance was assessed by receiver operating characteristic (ROC), calibration, and decision curve analysis. The areas under the ROC curve (AUCs) of models were corrected by 1,000-times bootstrapping method and compared by Delong's test. The added value of each independent model or their combinations was also assessed by net reclassification improvement (NRI) and integrated discrimination improvement (IDI) indices. This report referred to the "Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis" (TRIPOD) statement.

Results

The AUCs of the tumor radiomic model (eight features) and the FGT radiomic model (three features) were 0.783 (95% confidence interval [CI], 0.717-0.849) and 0.680 (95% CI, 0.604-0.757), respectively. A higher AUC of 0.799 (95% CI, 0.737-0.862) was obtained by combining tumor and FGT radiomics signatures. By further combining tumor and FGT radiomics signatures with progesterone receptor (PR) status, a nomogram was developed and showed better discriminative ability for SLN status [AUC 0.839 (95% CI, 0.783-0.895)]. The IDI and NRI indices also showed significant improvement when combining tumor, FGT, and PR compared with each independent model or a combination of any two of them (all p < 0.05).

Conclusion

FGT and clinical factors improved the prediction performance of SLN status in breast cancer. A nomogram integrating the DCE-MRI radiomics signature of tumor and FGT and PR expression achieved good performance for the prediction of SLN status, which provides a potential biomarker for clinical treatment decision-making.

Radiomics in breast MRI: current progress toward clinical application in the era of artificial intelligence

Breast magnetic resonance imaging (MRI) is the most sensitive imaging modality for breast cancer diagnosis and is widely used clinically. Dynamic contrast-enhanced MRI is the basis for breast MRI, but ultrafast images, T2-weighted images, and diffusion-weighted images are also taken to improve the characteristics of the lesion. Such multiparametric MRI with numerous morphological and functional data poses new challenges to radiologists, and thus, new tools for reliable, reproducible, and high-volume quantitative assessments are warranted. In this context, radiomics, which is an emerging field of research involving the conversion of digital medical images into mineable data for clinical decision-making and outcome prediction, has been gaining ground in oncology. Recent development in artificial intelligence has promoted radiomics studies in various fields including breast cancer treatment and numerous studies have been conducted. However, radiomics has shown a translational gap in clinical practice, and many issues remain to be solved. In this review, we will outline the steps of radiomics workflow and investigate clinical application of radiomics focusing on breast MRI based on published literature, as well as current discussion about limitations and challenges in radiomics.

Radiomics-based machine learning analysis and characterization of breast lesions with multiparametric diffusion-weighted MR

Background

This study aimed to evaluate the utility of radiomics-based machine learning analysis with multiparametric DWI and to compare the diagnostic performance of radiomics features and mean diffusion metrics in the characterization of breast lesions.

Methods

This retrospective study included 542 lesions from February 2018 to November 2018. One hundred radiomics features were computed from mono-exponential (ME), biexponential (BE), stretched exponential (SE), and diffusion-kurtosis imaging (DKI). Radiomics-based analysis was performed by comparing four classifiers, including random forest (RF), principal component analysis (PCA), L1 regularization (L1R), and support vector machine (SVM). These four classifiers were trained on a training set with 271 patients via ten-fold cross-validation and tested on an independent testing set with 271 patients. The diagnostic performance of the mean diffusion metrics of ME (mADC_{all b}, mADC_0–1000), BE (mD, mD^*, mf), SE (mDDC, mα), and DKI (mK, mD) were also calculated for comparison. The area under the receiver operating characteristic curve (AUC) was used to compare the diagnostic performance.

Results

RF attained higher AUCs than L1R, PCA and SVM. The AUCs of radiomics features for the differential diagnosis of breast lesions ranged from 0.80 (BE_D*) to 0.85 (BE_D). The AUCs of the mean diffusion metrics ranged from 0.54 (BE_mf) to 0.79 (ME_mADC_0–1000). There were significant differences in the AUCs between the mean values of all diffusion metrics and radiomics features of AUCs (all P < 0.001) for the differentiation of benign and malignant breast lesions. Of the radiomics features computed, the most important sequence was BE_D (AUC: 0.85), and the most important feature was FO-10 percentile (Feature Importance: 0.04).

Conclusions

The radiomics-based analysis of multiparametric DWI by RF enables better differentiation of benign and malignant breast lesions than the mean diffusion metrics.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-021-03117-5.

Radiomics as a New Frontier of Imaging for Cancer Prognosis: A Narrative Review

The evaluation of the efficacy of different therapies is of paramount importance for the patients and the clinicians in oncology, and it is usually possible by performing imaging investigations that are interpreted, taking in consideration different response evaluation criteria. In the last decade, texture analysis (TA) has been developed in order to help the radiologist to quantify and identify parameters related to tumor heterogeneity, which cannot be appreciated by the naked eye, that can be correlated with different endpoints, including cancer prognosis. The aim of this work is to analyze the impact of texture in the prediction of response and in prognosis stratification in oncology, taking into consideration different pathologies (lung cancer, breast cancer, gastric cancer, hepatic cancer, rectal cancer). Key references were derived from a PubMed query. Hand searching and clinicaltrials.gov were also used. This paper contains a narrative report and a critical discussion of radiomics approaches related to cancer prognosis in different fields of diseases.

Radiomic Evaluations of the Diagnostic Performance of DM, DBT, DCE MRI, DWI, and Their Combination for the Diagnosisof Breast Cancer

Objectives

This study aims to evaluate digital mammography (DM), digital breast tomosynthesis (DBT), dynamic contrast-enhanced (DCE), and diffusion-weighted (DW) MRI, individually and combined, for the values in the diagnosis of breast cancer, and propose a visualized clinical-radiomics nomogram for potential clinical uses.

Methods

A total of 120 patients were enrolled between September 2017 and July 2018, all underwent preoperative DM, DBT, DCE, and DWI scans. Radiomics features were extracted and selected using the least absolute shrinkage and selection operator (LASSO) regression. A radiomics nomogram was constructed integrating the radiomics signature and important clinical predictors, and assessed with the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

Results

The radiomics signature derived from DBT plus DM generated a lower area under the ROC curve (AUC) and sensitivity, but a higher specificity compared with that from DCE plus DWI. The nomogram integrating the combined radiomics signature, age, and menstruation status achieved the best diagnostic performance in the training (AUCs, nomogram vs. combined radiomics signature vs. clinical model, 0.975 vs. 0.964 vs. 0.782) and validation (AUCs, nomogram vs. combined radiomics signature vs. clinical model, 0.983 vs. 0.978 vs. 0.680) cohorts. DCA confirmed the potential clinical usefulness of the nomogram.

Conclusions

The DBT plus DM provided a lower AUC and sensitivity, but a higher specificity than DCE plus DWI for detecting breast cancer. The proposed clinical-radiomics nomogram has diagnostic advantages over each modality, and can be considered as an efficient tool for breast cancer screening.

Radiomic Features of T2-weighted Imaging and Diffusion Kurtosis Imaging in Differentiating Clinicopathological Characteristics of Cervical Carcinoma

RATIONALE AND OBJECTIVES

Clinicopathological characteristics including histological subtypes, tumour grades and International Federation of Gynecology and Obstetrics (FIGO) stages are crucial factors in the clinical decision for cervical carcinoma (CC). The purpose of this study was to evaluate the ability of T2-weighted imaging (T2WI) and diffusion kurtosis imaging (DKI) radiomics in differentiating clinicopathological characteristics of CC.

MATERIALS AND METHODS

One hundred and seventeen histologically confirmed CC patients (mean age 56.5 ± 14.0 years) with pre-treatment magnetic resonance imaging were retrospectively reviewed. DKI was acquired with 4 b-values (0-1500 s/mm²). Volumes of interest were contoured around the tumours on T2WI and DKI. Radiomic features including shape, first-order and grey-level co-occurrence matrix with wavelet transforms were extracted. Intraclass correlation coeffient between 2 radiologists was used for features reduction. Feature selection was achieved by elastic net and minimum redundancy maximum relevance. Selected features were used to build random forest (RF) models. The performances for differentiating histological subtypes, tumour grades and FIGO stages were assessed by receiver operating characteristic analysis.

RESULTS

Area under the curves (AUCs) for T2WI-only RF models for discriminating histological subtypes, tumour grades and FIGO stages were 0.762, 0.686, and 0.719. AUCs for DWI-only models were 0.663, 0.645, and 0.868, respectively. AUCs of the combined T2WI and DKI models were 0.823, 0.790, and 0.850, respectively.

CONCLUSION

T2WI and DKI radiomic features could differentiate the clinicopathological characteristics of CC. A combined model showed excellent diagnostic discrimination for histological subtypes, while a DKI-only model presented the best performance in differentiating FIGO stages.

Convolutional Neural Network of Multiparametric MRI Accurately Detects Axillary Lymph Node Metastasis in Breast Cancer Patients With Pre Neoadjuvant Chemotherapy

BACKGROUND

Accurate assessment of the axillary lymph nodes (aLNs) in breast cancer patients is essential for prognosis and treatment planning. Current radiological staging of nodal metastasis has poor accuracy. This study aimed to investigate the machine learning convolutional neural networks (CNNs) on multiparametric MRI to detect nodal metastasis with 18FDG-PET as ground truths.

MATERIALS AND METHODS

Data were obtained via a retrospective search. Inclusion criteria were patients with bilateral breast MRI and 18FDG-PETand/or CT scans obtained before neoadjuvant chemotherapy. In total, 238 aLNs were obtained from 56 breast cancer patients with 18FDG-PET and/or CT and breast MRI data. Radiologists scored each node based on all MRI as diseased and non-diseased nodes. Five models were built using T1-W MRI, T2-W MRI, DCE MRI, T1-W + T2-W MRI, and DCE + T2-W MRI model. Performance was evaluated using receiver operating curve (ROC) analysis, including area under the curve (AUC).

RESULTS

All CNN models yielded similar performance with an accuracy ranging from 86.08% to 88.50% and AUC ranging from 0.804 to 0.882. The CNN model using T1-W MRI performed better than that using T2-W MRI in detecting nodal metastasis. CNN model using combined T1- and T2-W MRI performed the best compared to all other models (accuracy = 88.50%, AUC = 0.882), but similar in AUC to the DCE + T2-W MRI model (accuracy = 88.02%, AUC = 0.880). All CNN models performed better than radiologists in detecting nodal metastasis (accuracy = 65.8%).

CONCLUSION

xxxxxx.

Radiomics: A Primer for Breast Radiologists

Radiomics has a long-standing history in breast imaging with computer-aided detection (CAD) for screening mammography developed in the late 20th century. Although conventional CAD had widespread adoption, the clinical benefits for experienced breast radiologists were debatable due to high false-positive marks and subsequent increased recall rates. The dramatic growth in recent years of artificial intelligence-based analysis, including machine learning and deep learning, has provided numerous opportunities for improved modern radiomics work in breast imaging. There has been extensive radiomics work in mammography, digital breast tomosynthesis, MRI, ultrasound, PET-CT, and combined multimodality imaging. Specific radiomics outcomes of interest have been diverse, including CAD, prediction of response to neoadjuvant therapy, lesion classification, and survival, among other outcomes. Additionally, the radiogenomics subfield that correlates radiomics features with genetics has been very proliferative, in parallel with the clinical validation of breast cancer molecular subtypes and gene expression assays. Despite the promise of radiomics, there are important challenges related to image normalization, limited large unbiased data sets, and lack of external validation. Much of the radiomics work to date has been exploratory using single-institution retrospective series for analysis, but several promising lines of investigation have made the leap to clinical practice with commercially available products. As a result, breast radiologists will increasingly be incorporating radiomics-based tools into their daily practice in the near future. Therefore, breast radiologists must have a broad understanding of the scope, applications, and limitations of radiomics work.

"Real-world" radiomics from multi-vendor MRI: an original retrospective study on the prediction of nodal status and disease survival in breast cancer, as an exemplar to promote discussion of the wider issues

BACKGROUND

Most MRI radiomics studies to date, even multi-centre ones, have used "pure" datasets deliberately accrued from single-vendor, single-field-strength scanners. This does not reflect aspirations for the ultimate generalisability of AI models. We therefore investigated the development of a radiomics signature from heterogeneous data originating on six different imaging platforms, for a breast cancer exemplar, in order to provide input into future discussions of the viability of radiomics in "real-world" scenarios where image data are not controlled by specific trial protocols but reflective of routine clinical practice.

METHODS

One hundred fifty-six patients with pathologically proven breast cancer underwent multi-contrast MRI prior to neoadjuvant chemotherapy and/or surgery. From these, 92 patients were identified for whom T2-weighted, diffusion-weighted and contrast-enhanced T1-weighted sequences were available, as well as key clinicopathological variables. Regions-of-interest were drawn on the above image types and, from these, semantic and calculated radiomics features were derived. Classification models using a variety of methods, both with and without recursive feature elimination, were developed to predict pathological nodal status. Separately, we applied the same methods to analyse the information carried by the radiomic features regarding the originating scanner type and field strength. Repeated, ten-fold cross-validation was employed to verify the results. In parallel work, survival modelling was performed using random survival forests.

RESULTS

Prediction of nodal status yielded mean cross-validated AUC values of 0.735 ± 0.15 (SD) for clinical variables alone, 0.673 ± 0.16 (SD) for radiomic features only, and 0.764 ± 0.16 (SD) for radiomics and clinical features together. Prediction of scanner platform from the radiomics features yielded extremely high values of AUC between 0.91 and 1 for the different classes examined indicating the presence of confounding features for the nodal status classification task. Survival analysis, gave out-of-bag prediction errors of 19.3% (clinical features only), 36.9-51.8% (radiomic features from different combinations of image contrasts), and 26.7-35.6% (clinical plus radiomics features).

CONCLUSIONS

Radiomic classification models whose predictive ability was consistent with previous single-vendor, single-field strength studies have been obtained from multi-vendor, multi-field-strength data, despite clear confounding information being present. However, our sample size was too small to obtain useful survival modelling results.

3T MRI-Radiomic Approach to Predict for Lymph Node Status in Breast Cancer Patients

Simple Summary

Breast cancer is the most common cancer in women worldwide. The axillary lymph node status is one of the main prognostic factors. Currently, the methods to define the lymph node status are invasive and not without sequelae (from biopsy to lymphadenectomy). Radiomics is a new tool, and highly varied, but with high potential that has already shown excellent results in numerous fields of application. In our study, we have developed a classifier validated on a relatively large number of patients, which is able to predict lymph node status using a combination of patients clinical features, primary breast cancer histological features and radiomics features based on 3 Tesla post contrast—MR images. This approach can accurately select breast cancer patients who may avoid unnecessary biopsy and lymphadenectomy in a non-invasive way.

Abstract

Background: axillary lymph node (LN) status is one of the main breast cancer prognostic factors and it is currently defined by invasive procedures. The aim of this study is to predict LN metastasis combining MRI radiomics features with primary breast tumor histological features and patients’ clinical data. Methods: 99 lesions on pre-treatment contrasted 3T-MRI (DCE). All patients had a histologically proven invasive breast cancer and defined LN status. Patients’ clinical data and tumor histological analysis were previously collected. For each tumor lesion, a semi-automatic segmentation was performed, using the second phase of DCE-MRI. Each segmentation was optimized using a convex-hull algorithm. In addition to the 14 semantics features and a feature ROI volume/convex-hull volume, 242 other quantitative features were extracted. A wrapper selection method selected the 15 most prognostic features (14 quantitative, 1 semantic), used to train the final learning model. The classifier used was the Random Forest. Results: the AUC-classifier was 0.856 (label = positive or negative). The contribution of each feature group was lower performance than the full signature. Conclusions: the combination of patient clinical, histological and radiomics features of primary breast cancer can accurately predict LN status in a non-invasive way.

Predicting cell invasion in breast tumor microenvironment from radiological imaging phenotypes

Background

The abundance of immune and stromal cells in the tumor microenvironment (TME) is informative of levels of inflammation, angiogenesis, and desmoplasia. Radiomics, an approach of extracting quantitative features from radiological imaging to characterize diseases, have been shown to predict molecular classification, cancer recurrence risk, and many other disease outcomes. However, the ability of radiomics methods to predict the abundance of various cell types in the TME remains unclear. In this study, we employed a radio-genomics approach and machine learning models to predict the infiltration of 10 cell types in breast cancer lesions utilizing radiomic features extracted from breast Dynamic Contrast Enhanced Magnetic Resonance Imaging.

Methods

We performed a retrospective study utilizing 73 patients from two independent institutions with imaging and gene expression data provided by The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA), respectively. A set of 199 radiomic features including shape-based, morphological, texture, and kinetic characteristics were extracted from the lesion volumes. To capture one-to-one relationships between radiomic features and cell type abundance, we performed linear regression on each radiomic feature/cell type abundance combination. Each regression model was tested for statistical significance. In addition, multivariate models were built for the cell type infiltration status (i.e. “high” vs “low”) prediction. A feature selection process via Recursive Feature Elimination was applied to the radiomic features on the training set. The classification models took the form of a binary logistic extreme gradient boosting framework. Two evaluation methods including leave-one-out cross validation and external independent test, were used for radiomic model learning and testing. The models’ performance was measured via area under the receiver operating characteristic curve (AUC).

Results

Univariate relationships were identified between a set of radiomic features and the abundance of fibroblasts. Multivariate models yielded leave-one-out cross validation AUCs ranging from 0.5 to 0.83, and independent test AUCs ranging from 0.5 to 0.68 for the multiple cell type invasion predictions.

Conclusions

On two independent breast cancer cohorts, breast MRI-derived radiomics are associated with the tumor’s microenvironment in terms of the abundance of several cell types. Further evaluation with larger cohorts is needed.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-08122-x.

Preoperative T2-weighted MR imaging texture analysis of gastric cancer: prediction of TNM stages

PURPOSE

To explore the capability of algorithms to build multivariate models integrating morphological and texture features derived from preoperative T₂-weighted magnetic resonance (MR) images of gastric cancer (GC) to evaluate tumor- (T), node- (N), and metastasis- (M) stages.

METHODS

A total of 80 patients at our hospital who underwent abdominal MR imaging and were diagnosed with GC from December 2011 to November 2016 were retrospectively included. Texture features were calculated using T₂-weighted images with a manual region of interest. Morphological characteristics were also evaluated. Classifiers and regression analyses were used to build multivariate models. Receiver operating characteristic (ROC) curve analysis was performed to assess diagnostic efficacy.

RESULTS

There were 8, 10, and 3 texture parameters that showed significant differences in GCs at different overall (I-II vs. III-IV), T (1-2 vs. 3-4), and N (- vs. +) stages (all p < 0.05), respectively. Mild thickening was more common in stages I-II, T1-2, and N- GCs (all p < 0.05). An irregular outer contour was more commonly observed in stages III-IV (p = 0.001) and T3-4 (p = 0.001) GCs. T3-4 and N+ GCs tended to be thickening type lesions (p = 0.005 and 0.032, respectively). The multivariate models using the naive bayes algorithm showed the highest diagnostic efficacy in predicting T and N stages (area under the ROC curves [AUC] = 0.900 and 0.863, respectively), and the model based on regression analysis had the best predictive performance in overall staging (AUC = 0.839).

CONCLUSION

Multivariate models combining morphological characteristics with texture parameters based on machine learning algorithms were able to improve diagnostic efficacy in predicting the overall, T, and N stages of GCs.

Radiomics MRI for lymph node status prediction in breast cancer patients: the state of art

OBJECTIVES

To create a review of the existing literature on the radiomic approach in predicting the lymph node status of the axilla in breast cancer (BC).

MATERIALS AND METHODS

Two reviewers conducted the literature search on MEDLINE databases independently. Ten articles on the prediction of sentinel lymph node metastasis in breast cancer with a radiomic approach were selected. The study characteristics and results were reported. The quality of the methodology was evaluated according to the Radiomics Quality Score (RQS).

RESULTS

All studies were retrospective in design and published between 2017 and 2020. The majority of studies used DCE-MRI sequences and two investigated only pre-contrast images. The sample size was lower than 200 patients for 7 studies. The pre-processing used software, feature extraction and selection methods and classifier development are heterogeneous and a standardization of results is not yet possible. The average RQS score was 11.1 (maximum possible value = 36). The criteria with the lowest scores were the type of study, validation, comparison with a gold standard, potential clinical utility, cost-effective analysis and open science data.

CONCLUSION

The field of radiomics is a diagnostic approach of relative recent development. The results in predicting axillary lymph node status are encouraging, but there are still weaknesses in the quality of studies that may limit the reproducibility of the results.

Radiomic features of axillary lymph nodes based on pharmacokinetic modeling DCE-MRI allow preoperative diagnosis of their metastatic status in breast cancer

OBJECTIVE

To study the feasibility of use of radiomic features extracted from axillary lymph nodes for diagnosis of their metastatic status in patients with breast cancer.

MATERIALS AND METHODS

A total of 176 axillary lymph nodes of patients with breast cancer, consisting of 87 metastatic axillary lymph nodes (ALNM) and 89 negative axillary lymph nodes proven by surgery, were retrospectively reviewed from the database of our cancer center. For each selected axillary lymph node, 106 radiomic features based on preoperative pharmacokinetic modeling dynamic contrast enhanced magnetic resonance imaging (PK-DCE-MRI) and 5 conventional image features were obtained. The least absolute shrinkage and selection operator (LASSO) regression was used to select useful radiomic features. Logistic regression was used to develop diagnostic models for ALNM. Delong test was used to compare the diagnostic performance of different models.

RESULTS

The 106 radiomic features were reduced to 4 ALNM diagnosis-related features by LASSO. Four diagnostic models including conventional model, pharmacokinetic model, radiomic model, and a combined model (integrating the Rad-score in the radiomic model with the conventional image features) were developed and validated. Delong test showed that the combined model had the best diagnostic performance: area under the curve (AUC), 0.972 (95% CI [0.947-0.997]) in the training cohort and 0.979 (95% CI [0.952-1]) in the validation cohort. The diagnostic performance of the combined model and the radiomic model were better than that of pharmacokinetic model and conventional model (P<0.05).

CONCLUSION

Radiomic features extracted from PK-DCE-MRI images of axillary lymph nodes showed promising application for diagnosis of ALNM in patients with breast cancer.

Dedicated Axillary MRI-Based Radiomics Analysis for the Prediction of Axillary Lymph Node Metastasis in Breast Cancer

Radiomics features may contribute to increased diagnostic performance of MRI in the prediction of axillary lymph node metastasis. The objective of the study was to predict preoperative axillary lymph node metastasis in breast cancer using clinical models and radiomics models based on T2-weighted (T2W) dedicated axillary MRI features with node-by-node analysis. From August 2012 until October 2014, all women who had undergone dedicated axillary 3.0T T2W MRI, followed by axillary surgery, were retrospectively identified, and available clinical data were collected. All axillary lymph nodes were manually delineated on the T2W MR images, and quantitative radiomics features were extracted from the delineated regions. Data were partitioned patient-wise to train 100 models using different splits for the training and validation cohorts to account for multiple lymph nodes per patient and class imbalance. Features were selected in the training cohorts using recursive feature elimination with repeated 5-fold cross-validation, followed by the development of random forest models. The performance of the models was assessed using the area under the curve (AUC). A total of 75 women (median age, 61 years; interquartile range, 51-68 years) with 511 axillary lymph nodes were included. On final pathology, 36 (7%) of the lymph nodes had metastasis. A total of 105 original radiomics features were extracted from the T2W MR images. Each cohort split resulted in a different number of lymph nodes in the training cohorts and a different set of selected features. Performance of the 100 clinical and radiomics models showed a wide range of AUC values between 0.41-0.74 and 0.48-0.89 in the training cohorts, respectively, and between 0.30-0.98 and 0.37-0.99 in the validation cohorts, respectively. With these results, it was not possible to obtain a final prediction model. Clinical characteristics and dedicated axillary MRI-based radiomics with node-by-node analysis did not contribute to the prediction of axillary lymph node metastasis in breast cancer based on data where variations in acquisition and reconstruction parameters were not addressed.

Machine learning prediction of axillary lymph node metastasis in breast cancer: 2D versus 3D radiomic features

PURPOSE

The purpose of this study was to distinguish axillary lymph node (ALN) status using preoperative breast DCE-MRI radiomics and compare the effects of two-dimensional (2D) and three-dimensional (3D) analysis.

METHODS

A retrospective study including 154 breast cancer patients all confirmed by pathology; 80 with ALN metastasis and 74 without. All MRI scans were achieved at a 3.0 Tesla scanner with 7 post-contrast MR phases sequentially acquired with a temporal resolution of 60 s. MRI radiomic features were extracted separately from a 2D single slice (i.e., the representative slice) and the 3D tumor volume. Several machine learning classifiers were built and compared using 2D or 3D analysis to distinguish positive vs negative ALN status. We performed independent test and 10-fold cross validation with multiple repetitions, and used bootstrap test, least absolute shrinkage selection operator, and receiver operating characteristic (ROC) curve analysis as statistical tests.

RESULTS

The highest area under the ROC curve (AUC) was 0.81 (95% confidence intervals [CI]: 0.80-0.83) and 0.82 (95% CI: 0.81-0.82) for 2D and 3D analysis, respectively; the corresponding accuracy was 79% and 80%. The linear discriminant analysis (LDA) classifier achieved the highest classification performance. None of the AUC differences between 2D and 3D analysis was statistically significant for the several tested machine learning classifiers (all P> 0.05).

CONCLUSIONS

Radiomic features from segmented tumor region in breast MRI were associated with ALN status. The separate radiomic analysis on 3D tumor volume showed a similar effect to the 2D analysis on the single representative slice in the tested machine learning classifiers.

Radiomic signatures derived from multiparametric MRI for the pretreatment prediction of response to neoadjuvant chemotherapy in breast cancer

Objectives:

To investigate the ability of radiomic signatures based on MRI to evaluate the response and efficiency of neoadjuvant chemotherapy (NAC) for treating breast cancers.

Methods:

152 patients were included in this study at our institution between March 2017 and September 2019. All patients with breast cancer underwent a preoperative breast MRI and the Miller–Payne grading system was applied to evaluate response to NAC. Quantitative parameters were compared between patients with sensitive and insensitive responses to NAC and between those with pathological complete responses (pCR) and non-pCR. Four radiomic signatures were built based on T2W imaging, diffusion-weighted imaging, dynamic contrast-enhanced imaging and their combination, and radiomics scores (Rad-score) were calculated. The combination of the clinical factors and Rad-scores created a nomogram model. Multivariate logistic regression was performed to assess the association between MRI features and independent clinical risk factors.

Results:

20 features and 18 features were selected to build the radiomic signature for evaluating sensitivity and the possibility of pCR, respectively. The combined radiomic signature and nomogram model showed a similar discrimination in the training (AUC 0.91, 0.92, 95% confidence interval [CI], 0.85–0.96, 0.86–0.98) and validation (AUC 0.93, 0.91, 95% CI, 0.86–1.00, 0.82–1.00) sets. The clinical factor model exhibited reduced performance (AUC 0.74, 0.64, 95% CI, 0.64–0.84, 0.46–0.82) in terms of NAC sensitivity and pCR.

Conclusions:

The combined radiomic signature and nomogram model exhibited potential predictive power for predicting effective NAC treatment which can aid in the prognosis and guidance of treatment regimens.

Advances in knowledge:

Identifying a means of assessing the efficacy of NAC before surgery can guide follow-up treatment and avoid chemotherapy-induced toxicity.

Mammography-based radiomics nomogram: a potential biomarker to predict axillary lymph node metastasis in breast cancer

OBJECTIVE

To establish a radiomics nomogram by integrating clinical risk factors and radiomics features extracted from digital mammography (MG) images for pre-operative prediction of axillary lymph node (ALN) metastasis in breast cancer.

METHODS

216 patients with breast cancer lesions confirmed by surgical excision pathology were divided into the primary cohort (n = 144) and validation cohort (n = 72). Radiomics features were extracted from craniocaudal (CC) view of mammograms, and radiomics features selection were performed using the methods of ANOVA F-value and least absolute shrinkage and selection operator; then a radiomics signature was constructed with the method of support vector machine. Multivariate logistic regression analysis was used to establish a radiomics nomogram based on the combination of radiomics signature and clinical factors. The C-index and calibration curves were derived based on the regression analysis both in the primary and validation cohorts.

RESULTS

95 of 216 patients were confirmed with ALN metastasis by pathology, and 52 cases were diagnosed as ALN metastasis based on MG-reported criteria. The sensitivity, specificity, accuracy and AUC (area under the receiver operating characteristic curve of MG-reported criteria were 42.7%, 90.8%, 24.1% and 0.666 (95% confidence interval: 0.591-0.741]. The radiomics nomogram, comprising progesterone receptor status, molecular subtype and radiomics signature, showed good calibration and better favorite performance for the metastatic ALN detection (AUC 0.883 and 0.863 in the primary and validation cohorts) than each independent clinical features (AUC 0.707 and 0.657 in the primary and validation cohorts) and radiomics signature (AUC 0.876 and 0.862 in the primary and validation cohorts).

CONCLUSION

The MG-based radiomics nomogram could be used as a non-invasive and reliable tool in predicting ALN metastasis and may facilitate to assist clinicians for pre-operative decision-making.

ADVANCES IN KNOWLEDGE

ALN status remains among the most important breast cancer prognostic factors and is essential for making treatment decisions. However, the value of detecting metastatic ALN by MG is very limited. The studies on pre-operative ALN metastasis prediction using the method of MG-based radiomics in breast cancer are very few. Therefore, we studied whether MG-based radiomics nomogram could be used as a predictive biomarker for the detection of metastatic ALN.

Predictive performance of ultrasonography-based radiomics for axillary lymph node metastasis in the preoperative evaluation of breast cancer

PURPOSE

The purpose of this study was to evaluate the predictive performance of ultrasonography (US)-based radiomics for axillary lymph node metastasis and to compare it with that of a clinicopathologic model.

METHODS

A total of 496 patients (mean age, 52.5±10.9 years) who underwent breast cancer surgery between January 2014 and December 2014 were included in this study. Among them, 306 patients who underwent surgery between January 2014 and August 2014 were enrolled as a training cohort, and 190 patients who underwent surgery between September 2014 and December 2014 were enrolled as a validation cohort. To predict axillary lymph node metastasis in breast cancer, we developed a preoperative clinicopathologic model using multivariable logistic regression and constructed a radiomics model using 23 radiomic features selected via least absolute shrinkage and selection operator regression.

RESULTS

In the training cohort, the areas under the curve (AUC) were 0.760, 0.812, and 0.858 for the clinicopathologic, radiomics, and combined models, respectively. In the validation cohort, the AUCs were 0.708, 0.831, and 0.810, respectively. The combined model showed significantly better diagnostic performance than the clinicopathologic model.

CONCLUSION

A radiomics model based on the US features of primary breast cancers showed additional value when combined with a clinicopathologic model to predict axillary lymph node metastasis.

MRI Radiomic Features: Association with Disease-Free Survival in Patients with Triple-Negative Breast Cancer

Radiomic features hold potential to improve prediction of disease-free survival (DFS) in triple-negative breast cancer (TNBC) and may show better performance if developed from TNBC patients. We aimed to develop a radiomics score based on MRI features to estimate DFS in patients with TNBC. A total of 228 TNBC patients who underwent preoperative MRI and surgery between April 2012 and December 2016 were included. Patients were temporally divided into the training (n = 169) and validation (n = 59) set. Radiomic features of the tumor were extracted from T2-weighted and contrast-enhanced T1- weighted MRI. Then a radiomics score was constructed with the least absolute shrinkage and selection operator regression in the training set. Univariate and multivariate Cox proportional hazards models were used to determine what associations the radiomics score and clinicopathologic variables had with DFS. A combined clinicopathologic-radiomic (CCR) model was constructed based on multivariate Cox analysis. The incremental values of the radiomics score were evaluated by using the integrated area under the receiver operating characteristic curve (iAUC) and bootstrapping (n = 1000). The radiomics score, which consisted of 5 selected MRI features, was significantly associated with worse DFS in both the training and validation sets (p = 0.002, p = 0.033, respectively). In both the training and validation set, the radiomics score showed comparable performance with the clinicopathologic model. The CCR model demonstrated better performance than the clinicopathologic model in the training set (iAUC, 0.844; difference in iAUC, p < 0.001) and validation set (iAUC, 0.765, difference in iAUC, p < 0.001). In conclusion, MRI-based radiomic features can improve the prediction of DFS when integrated with clinicopathologic data in patients with TNBC.

Radiomics Based on Multimodal MRI for the Differential Diagnosis of Benign and Malignant Breast Lesions

BACKGROUND

MRI-based radiomics has been used to diagnose breast lesions; however, little research combining quantitative pharmacokinetic parameters of dynamic contrast-enhanced MRI (DCE-MRI) and diffusion kurtosis imaging (DKI) exists.

PURPOSE

To develop and validate a multimodal MRI-based radiomics model for the differential diagnosis of benign and malignant breast lesions and analyze the discriminative abilities of different MR sequences.

STUDY TYPE

Retrospective.

POPULATION

In all, 207 female patients with 207 histopathology-confirmed breast lesions (95 benign and 112 malignant) were included in the study. Then 159 patients were assigned to the training group, and 48 patients comprised the validation group.

FIELD STRENGTH/SEQUENCE

T₂ -weighted (T₂ W), T₁ -weighted (T₁ W), diffusion-weighted MR imaging (b-values = 0, 500, 800, and 2000 seconds/mm² ) and quantitative DCE-MRI were performed on a 3.0T MR scanner.

ASSESSMENT

Radiomics features were extracted from T₂ WI, T₁ WI, DKI, apparent diffusion coefficient (ADC) maps, and DCE pharmacokinetic parameter maps in the training set. Models based on each sequence or combinations of sequences were built using a support vector machine (SVM) classifier and used to differentiate benign and malignant breast lesions in the validation set.

STATISTICAL TESTS

Optimal feature selection was performed by Spearman's rank correlation coefficients and the least absolute shrinkage and selection operator algorithm (LASSO). Receiver operating characteristic (ROC) curves were used to assess the diagnostic performance of the radiomics models in the validation set.

RESULTS

The area under the ROC curve (AUC) of the optimal radiomics model, including T₂ WI, DKI, and quantitative DCE-MRI parameter maps was 0.921, with an accuracy of 0.833. The AUCs of the models based on T₁ WI, T₂ WI, ADC map, DKI, and DCE pharmacokinetic parameter maps were 0.730, 0.791, 0.770, 0.788, and 0.836, respectively.

DATA CONCLUSION

The model based on radiomics features from T₂ WI, DKI, and quantitative DCE pharmacokinetic parameter maps has a high discriminatory ability for benign and malignant breast lesions.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:596-607.

The Application of Radiomics in Breast MRI: A Review

Breast cancer has been a worldwide burden of women's health. Although concerns have been raised for early diagnosis and timely treatment, the efforts are still needed for precision medicine and individualized treatment. Radiomics is a new technology with immense potential to obtain mineable data to provide rich information about the diagnosis and prognosis of breast cancer. In our study, we introduced the workflow and application of radiomics as well as its outlook and challenges based on published studies. Radiomics has the potential ability to differentiate between malignant and benign breast lesions, predict axillary lymph node status, molecular subtypes of breast cancer, tumor response to chemotherapy, and survival outcomes. Our study aimed to help clinicians and radiologists to know the basic information of radiomics and encourage cooperation with scientists to mine data for better application in clinical practice.

Automated deep learning method for whole-breast segmentation in diffusion-weighted breast MRI

BACKGROUND

Diffusion-weighted imaging (DWI) in MRI plays an increasingly important role in diagnostic applications and developing imaging biomarkers. Automated whole-breast segmentation is an important yet challenging step for quantitative breast imaging analysis. While methods have been developed on dynamic contrast-enhanced (DCE) MRI, automatic whole-breast segmentation in breast DWI MRI is still underdeveloped.

PURPOSE

To develop a deep/transfer learning-based segmentation approach for DWI MRI scans and conduct an extensive study assessment on four imaging datasets from both internal and external sources.

STUDY TYPE

Retrospective.

SUBJECTS

In all, 98 patients (144 MRI scans; 11,035 slices) of four different breast MRI datasets from two different institutions.

FIELD STRENGTH/SEQUENCES

1.5T scanners with DCE sequence (Dataset 1 and Dataset 2) and DWI sequence. A 3.0T scanner with one external DWI sequence.

ASSESSMENT

Deep learning models (UNet and SegNet) and transfer learning were used as segmentation approaches. The main DCE Dataset (4,251 2D slices from 39 patients) was used for pre-training and internal validation, and an unseen DCE Dataset (431 2D slices from 20 patients) was used as an independent test dataset for evaluating the pre-trained DCE models. The main DWI Dataset (6,343 2D slices from 75 MRI scans of 29 patients) was used for transfer learning and internal validation, and an unseen DWI Dataset (10 2D slices from 10 patients) was used for independent evaluation to the fine-tuned models for DWI segmentation. Manual segmentations by three radiologists (>10-year experience) were used to establish the ground truth for assessment. The segmentation performance was measured using the Dice Coefficient (DC) for the agreement between manual expert radiologist's segmentation and algorithm-generated segmentation.

STATISTICAL TESTS

The mean value and standard deviation of the DCs were calculated to compare segmentation results from different deep learning models.

RESULTS

For the segmentation on the DCE MRI, the average DC of the UNet was 0.92 (cross-validation on the main DCE dataset) and 0.87 (external evaluation on the unseen DCE dataset), both higher than the performance of the SegNet. When segmenting the DWI images by the fine-tuned models, the average DC of the UNet was 0.85 (cross-validation on the main DWI dataset) and 0.72 (external evaluation on the unseen DWI dataset), both outperforming the SegNet on the same datasets.

DATA CONCLUSION

The internal and independent tests show that the deep/transfer learning models can achieve promising segmentation effects validated on DWI data from different institutions and scanner types. Our proposed approach may provide an automated toolkit to help computer-aided quantitative analyses of breast DWI images.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:635-643.