The Potential Use of DCE-MRI Texture Analysis to Predict HER2 2+ Status

Discrimination between human epidermal growth factor receptor 2 (HER2)-low-expressing and HER2-overexpressing breast cancers: a comparative study of four MRI diffusion models

OBJECTIVES

To determine the value of conventional DWI, continuous-time random walk (CTRW), fractional order calculus (FROC), and stretched exponential model (SEM) in discriminating human epidermal growth factor receptor 2 (HER2) status of breast cancer (BC).

METHODS

This prospective study included 158 women who underwent DWI, CTRW, FROC, and SEM and were pathologically categorized into the HER2-zero-expressing group (n = 10), HER2-low-expressing group (n = 86), and HER2-overexpressing group (n = 62). Nine diffusion parameters, namely ADC, αCTRW, βCTRW, DCTRW, βFROC, DFROC, μFROC, αSEM, and DDCSEM of the primary tumor, were derived from four diffusion models. These diffusion metrics and clinicopathologic features were compared between groups. Logistic regression was used to determine the optimal diffusion metrics and clinicopathologic variables for classifying the HER2-expressing statuses. Receiver operating characteristic (ROC) curves were used to evaluate their discriminative ability.

RESULTS

The estrogen receptor (ER) status, progesterone receptor (PR) status, and tumor size differed between HER2-low-expressing and HER2-overexpressing groups (p < 0.001 to p = 0.009). The αCTRW, DCTRW, βFROC, DFROC, μFROC, αSEM, and DDCSEM were significantly lower in HER2-low-expressing BCs than those in HER2-overexpressing BCs (p < 0.001 to p = 0.01). Further multivariable logistic regression analysis showed that the αCTRW was the single best discriminative metric, with an area under the curve (AUC) being higher than that of ADC (0.802 vs. 0.610, p < 0.05); the addition of ER status, PR status, and tumor size to the αCTRW improved the AUC to 0.877.

CONCLUSIONS

The αCTRW could help discriminate the HER2-low-expressing and HER2-overexpressing BCs.

CLINICAL RELEVANCE STATEMENT

Human epidermal growth factor receptor 2 (HER2)-low-expressing breast cancer (BC) might also benefit from the HER2-targeted therapy. Prediction of HER2-low-expressing BC or HER2-overexpressing BC is crucial for appropriate management. Advanced continuous-time random walk diffusion MRI offers a solution to this clinical issue.

KEY POINTS

• Human epidermal receptor 2 (HER2)-low-expressing BC had lower αCTRW, DCTRW, βFROC, DFROC, μFROC, αSEM, and DDCSEM values compared with HER2-overexpressing breast cancer. • The αCTRW was the single best diffusion metric (AUC = 0.802) for discrimination between the HER2-low-expressing and HER2-overexpressing breast cancers. • The addition of αCTRW to the clinicopathologic features (estrogen receptor status, progesterone receptor status, and tumor size) further improved the discriminative ability.

Advances in artificial intelligence for the diagnosis and treatment of ovarian cancer (Review)

Artificial intelligence (AI) has emerged as a crucial technique for extracting high‑throughput information from various sources, including medical images, pathological images, and genomics, transcriptomics, proteomics and metabolomics data. AI has been widely used in the field of diagnosis, for the differentiation of benign and malignant ovarian cancer (OC), and for prognostic assessment, with favorable results. Notably, AI‑based radiomics has proven to be a non‑invasive, convenient and economical approach, making it an essential asset in a gynecological setting. The present study reviews the application of AI in the diagnosis, differentiation and prognostic assessment of OC. It is suggested that AI‑based multi‑omics studies have the potential to improve the diagnostic and prognostic predictive ability in patients with OC, thereby facilitating the realization of precision medicine.

Machine learning-based model constructed from ultrasound radiomics and clinical features for predicting HER2 status in breast cancer patients with indeterminate (2+) immunohistochemical results

BACKGROUND

We aimed to predict human epidermal growth factor receptor 2 (HER2) 2+ status in patients with breast cancer by constructing and validating machine learning models utilizing ultrasound (US) radiomics and clinical features.

METHODS

We analyzed 203 breast cancer cases immunohistochemically determined as HER2 2+ and used fluorescence in situ hybridization (FISH) as the confirmation method. From each case, the study analyzed 840 extracted radiomics features and 11 clinicopathologic features. Cases were randomly split into training (n = 141) and validation sets (n = 62) at a 7:3 ratio. Univariate logistic regression analysis was first performed on the 11 clinicopathologic characteristics. The least absolute shrinkage and selection operator (LASSO) and decision tree (DT) techniques were employed for post-feature selection. Finally, 19 radiomics features were utilized in logistic regression (LR) and Naive Bayesian (NB) classifiers. Model performance was gauged using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity.

RESULTS

Our models exhibited notable diagnostic efficacy in differentiating HER2-positive from negative breast cancer cases. In the validation sets, the LR model outperformed the NB model with an AUC of 0.860 and accuracy of 83.8% compared to NB's AUC of 0.684 and accuracy of 79.0%. The LR model demonstrated higher sensitivity (92.3% vs. 46.2%) while the NB model had a better specificity (91.8% vs. 63.3%) in the validation set.

CONCLUSIONS

Machine learning models grounded on radiomics efficiently predicted IHC HER2 2+ status in breast cancer patients, suggesting potential enhancements in clinical decision-making for treatment and management.

HER2GAN: Overcome the Scarcity of HER2 Breast Cancer Dataset Based on Transfer Learning and GAN Model

INTRODUCTION

Immunohistochemistry (IHC) is crucial for breast cancer diagnosis, classification, and individualized treatment. IHC is used to measure the levels of expression of hormone receptors (estrogen and progesterone receptors), human epidermal growth factor receptor 2 (HER2), and other biomarkers, which are used to make treatment decisions and predict how well a patient will do. The evaluation of the breast cancer score on IHC slides, taking into account structural and morphological features as well as a scarcity of relevant data, is one of the most important issues in the IHC debate. Several recent studies have utilized machine learning and deep learning techniques to resolve these issues.

MATERIALS AND METHODS

This paper introduces a new approach for addressing the issue based on supervised deep learning. A GAN-based model is proposed for generating high-quality HER2 images and identifying and classifying HER2 levels. Using transfer learning methodologies, the original and generated images were evaluated.

RESULTS AND CONCLUSION

All of the models have been trained and evaluated using publicly accessible and private data sets, respectively. The InceptionV3 and InceptionResNetV2 models achieved a high accuracy of 93% with the combined generated and original images used for training and testing, demonstrating the exceptional quality of the details in the synthesized images.

Potential of the Non-Contrast-Enhanced Chest CT Radiomics to Distinguish Molecular Subtypes of Breast Cancer: A Retrospective Study

Objectives

The molecular subtype plays an important role in breast cancer, which is the main reference to guide treatment and is closely related to prognosis. The objective of this study was to explore the potential of the non-contrast-enhanced chest CT-based radiomics to predict breast cancer molecular subtypes non-invasively.

Methods

A total of 300 breast cancer patients (153 luminal types and 147 non-luminal types) who underwent routine chest CT examination were included in the study, of which 220 cases belonged to the training set and 80 cases to the time-independent test set. Identification of the molecular subtypes is based on immunohistochemical staining of postoperative tissue samples. The region of interest (ROI) of breast masses was delineated on the continuous slices of CT images. Forty-two models to predict the luminal type of breast cancer were established by the combination of six feature screening methods and seven machine learning classifiers; 5-fold cross-validation (cv) was used for internal validation. Finally, the optimal model was selected for external validation on the independent test set. In addition, we also took advantage of SHapley Additive exPlanations (SHAP) values to make explanations of the machine learning model.

Results

During internal validation, the area under the curve (AUC) values for different models ranged from 0.599 to 0.842, and the accuracy ranged from 0.540 to 0.775. Eventually, the LASSO_SVM combination was selected as the final model, which included 9 radiomics features. The AUC, accuracy, sensitivity, and specificity of the model to distinguish luminal from the non-luminal type were 0.842 [95% CI: 0.728−0.957], 0.773, 0.818, and 0.773 in the training set and 0.757 [95% CI: 0.640–0.866], 0.713, 0.767, and 0.676 in the test set.

Conclusion

The radiomics based on chest CT may provide a new idea for the identification of breast cancer molecular subtypes.

Radiomics of dynamic contrast-enhanced magnetic resonance imaging parametric maps and apparent diffusion coefficient maps to predict Ki-67 status in breast cancer

Purpose

This study was aimed at evaluating whether a radiomics model based on the entire tumor region from breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) parametric maps and apparent diffusion coefficient (ADC) maps could indicate the Ki-67 status of patients with breast cancer.

Materials and methods

This retrospective study enrolled 205 women with breast cancer who underwent clinicopathological examination. Among them, 93 (45%) had a low Ki-67 amplification index (Ki-67 positivity< 14%), and 112 (55%) had a high Ki-67 amplification index (Ki-67 positivity ≥ 14%). Radiomics features were extracted from three DCE-MRI parametric maps and ADC maps calculated from two different b values of diffusion-weighted imaging sequences. The patients were randomly divided into a training set (70% of patients) and a validation set (30% of patients). After feature selection, we trained six support vector machine classifiers by combining different parameter maps and used 10-fold cross-validation to predict the expression level of Ki-67. The performance of six classifiers was evaluated with receiver operating characteristic (ROC) analysis, sensitivity, and specificity in both cohorts.

Results

Among the six classifiers constructed, a radiomics feature set combining three DCE-MRI parametric maps and ADC maps yielded an area under the ROC curve (AUC) of 0.839 (95% confidence interval [CI], 0.768-0.895) within the training set and 0.795 (95% CI, 0.674-0.887) within the independent validation set. Additionally, the AUC value, compared with that for a single parameter map, was moderately increased by combining features from the three parametric maps.

Conclusions

Radiomics features derived from the DCE-MRI parametric maps and ADC maps have the potential to serve as imaging biomarkers to determine Ki-67 status in patients with breast cancer.

MRI-Based Machine Learning in Differentiation Between Benign and Malignant Breast Lesions

Background

Differential diagnosis between benign and malignant breast lesions is of crucial importance relating to follow-up treatment. Recent development in texture analysis and machine learning may lead to a new solution to this problem.

Method

This current study enrolled a total number of 265 patients (benign breast lesions:malignant breast lesions = 71:194) diagnosed in our hospital and received magnetic resonance imaging between January 2014 and August 2017. Patients were randomly divided into the training group and validation group (4:1), and two radiologists extracted their texture features from the contrast-enhanced T1-weighted images. We performed five different feature selection methods including Distance correlation, Gradient Boosting Decision Tree (GBDT), least absolute shrinkage and selection operator (LASSO), random forest (RF), eXtreme gradient boosting (Xgboost) and five independent classification models were built based on Linear discriminant analysis (LDA) algorithm.

Results

All five models showed promising results to discriminate malignant breast lesions from benign breast lesions, and the areas under the curve (AUCs) of receiver operating characteristic (ROC) were all above 0.830 in both training and validation groups. The model with a better discriminating ability was the combination of LDA + gradient boosting decision tree (GBDT). The sensitivity, specificity, AUC, and accuracy in the training group were 0.814, 0.883, 0.922, and 0.868, respectively; LDA + random forest (RF) also suggests promising results with the AUC of 0.906 in the training group.

Conclusion

The evidence of this study, while preliminary, suggested that a combination of MRI texture analysis and LDA algorithm could discriminate benign breast lesions from malignant breast lesions. Further multicenter researches in this field would be of great help in the validation of the result.

Mask-Guided Convolutional Neural Network for Breast Tumor Prognostic Outcome Prediction on 3D DCE-MR Images

In this proof-of-concept work, we have developed a 3D-CNN architecture that is guided by the tumor mask for classifying several patient-outcomes in breast cancer from the respective 3D dynamic contrast-enhanced MRI (DCE-MRI) images. The tumor masks on DCE-MRI images were generated using pre- and post-contrast images and validated by experienced radiologists. We show that our proposed mask-guided classification has a higher accuracy than that from either the full image without tumor masks (including background) or the masked voxels only. We have used two patient outcomes for this study: (1) recurrence of cancer after 5 years of imaging and (2) HER2 status, for comparing accuracies of different models. By looking at the activation maps, we conclude that an image-based prediction model using 3D-CNN could be improved by even a conservatively generated mask, rather than overly trusting an unguided, blind 3D-CNN. A blind CNN may classify accurately enough, while its attention may really be focused on a remote region within 3D images. On the other hand, only using a conservatively segmented region may not be as good for classification as using full images but forcing the model's attention toward the known regions of interest.