Multilevel analysis of spatiotemporal association features for differentiation of tumor enhancement patterns in breast DCE-MRI

A Pipelined Tracer-Aware Approach for Lesion Segmentation in Breast DCE-MRI

The recent spread of Deep Learning (DL) in medical imaging is pushing researchers to explore its suitability for lesion segmentation in Dynamic Contrast-Enhanced Magnetic-Resonance Imaging (DCE-MRI), a complementary imaging procedure increasingly used in breast-cancer analysis. Despite some promising proposed solutions, we argue that a "naive" use of DL may have limited effectiveness as the presence of a contrast agent results in the acquisition of multimodal 4D images requiring thorough processing before training a DL model. We thus propose a pipelined approach where each stage is intended to deal with or to leverage a peculiar characteristic of breast DCE-MRI data: the use of a breast-masking pre-processing to remove non-breast tissues; the use of Three-Time-Points (3TP) slices to effectively highlight contrast agent time course; the application of a motion-correction technique to deal with patient involuntary movements; the leverage of a modified U-Net architecture tailored on the problem; and the introduction of a new "Eras/Epochs" training strategy to handle the unbalanced dataset while performing a strong data augmentation. We compared our pipelined solution against some literature works. The results show that our approach outperforms the competitors by a large margin (+9.13% over our previous solution) while also showing a higher generalization ability.

Vascularity and Dynamic Contrast-Enhanced Breast Magnetic Resonance Imaging

Angiogenesis is a key step in the initiation and progression of an invasive breast cancer. High microvessel density by morphological characterization predicts metastasis and poor survival in women with invasive breast cancers. However, morphologic characterization is subject to variability and only can evaluate a limited portion of an invasive breast cancer. Consequently, breast Magnetic Resonance Imaging (MRI) is currently being evaluated to assess vascularity. Recently, through the new field of radiomics, dynamic contrast enhanced (DCE)-MRI is being used to evaluate vascular density, vascular morphology, and detection of aggressive breast cancer biology. While DCE-MRI is a highly sensitive tool, there are specific features that limit computational evaluation of blood vessels. These include (1) DCE-MRI evaluates gadolinium contrast and does not directly evaluate biology, (2) the resolution of DCE-MRI is insufficient for imaging small blood vessels, and (3) DCE-MRI images are very difficult to co-register. Here we review computational approaches for detection and analysis of blood vessels in DCE-MRI images and present some of the strategies we have developed for co-registry of DCE-MRI images and early detection of vascularization.

Multiblock Discriminant Analysis of Integrative 18F-FDG-PET/CT Radiomics for Predicting Circulating Tumor Cells in Early-Stage Non-small Cell Lung Cancer Treated With Stereotactic Body Radiation Therapy

PURPOSE

The main objective of the present study was to integrate 18F-FDG-PET/CT radiomics with multiblock discriminant analysis for predicting circulating tumor cells (CTCs) in early-stage non-small cell lung cancer (ES-NSCLC) treated with stereotactic body radiation therapy (SBRT).

METHODS

Fifty-six patients with stage I NSCLC treated with SBRT underwent 18F-FDG-PET/CT imaging pre-SBRT and post-SBRT (median, 5 months; range, 3-10 months). CTCs were assessed via a telomerase-based assay before and within 3 months after SBRT and dichotomized at 5 and 1.3 CTCs/mL. Pre-SBRT, post-SBRT, and delta PET/CT radiomics features (n = 1548 × 3/1562 × 3) were extracted from gross tumor volume. Seven feature blocks were constructed including clinical parameters (n = 12). Multiblock data integration was performed using block sparse partial least squares-discriminant analysis (sPLS-DA) referred to as Data Integration Analysis for Biomarker Discovery Using Latent Components (DIABLO) for identifying key signatures by maximizing common information between different feature blocks while discriminating CTC levels. Optimal input blocks were identified using a pairwise combination method. DIABLO performance for predicting pre-SBRT and post-SBRT CTCs was evaluated using combined AUC (area under the curve, averaged across different blocks) analysis with 20 × 5-fold cross-validation (CV) and compared with that of concatenation-based sPLS-DA that consisted of combining all features into 1 block. CV prediction scores between 1 class versus the other were compared using the Wilcoxon rank sum test.

RESULTS

For predicting pre-SBRT CTCs, DIABLO achieved the best performance with combined pre-SBRT PET radiomics and clinical feature blocks, showing CV AUC of 0.875 (P = .009). For predicting post-SBRT CTCs, DIABLO achieved the best performance with combined post-SBRT CT and delta CT radiomics feature blocks, showing CV AUCs of 0.883 (P = .001). In contrast, all single-block sPLS-DA models could not attain CV AUCs higher than 0.7.

CONCLUSIONS

Multiblock integration with discriminant analysis of 18F-FDG-PET/CT radiomics has the potential for predicting pre-SBRT and post-SBRT CTCs. Radiomics and CTC analysis may complement and together help guide the subsequent management of patients with ES-NSCLC.

Texture Analysis of DCE-MRI Intratumoral Subregions to Identify Benign and Malignant Breast Tumors

Purpose

To evaluate the potential of the texture features extracted from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) intratumoral subregions to distinguish benign from malignant breast tumors.

Materials and Methods

A total of 299 patients with pathologically verified breast tumors who underwent breast DCE-MRI examination were enrolled in this study, including 124 benign cases and 175 malignant cases. The whole tumor area was semi-automatically segmented on the basis of subtraction images of DCE-MRI in Matlab 2018b. According to the time to peak of the contrast agent, the whole tumor area was partitioned into three subregions: early, moderate, and late. A total of 467 texture features were extracted from the whole tumor area and the three subregions, respectively. Patients were divided into training (n = 209) and validation (n = 90) cohorts by different MRI scanners. The least absolute shrinkage and selection operator (LASSO) method was used to select the optimal feature subset in the training cohort. The Kolmogorov-Smirnov test was first performed on texture features selected by LASSO to test whether the samples followed a normal distribution. Two machine learning methods, decision tree (DT) and support vector machine (SVM), were used to establish classification models with a 10-fold cross-validation method. The performance of the classification models was evaluated with receiver operating characteristic (ROC) curves.

Results

In the training cohort, the areas under the ROC curve (AUCs) for the DT_Whole model and SVM_Whole model were 0.744 and 0.806, respectively. In contrast, the AUCs of the DT_Early model (P = 0.004), DT_Late model (P = 0.015), SVM_Early model (P = 0.002), and SVM_Late model (P = 0.002) were significantly higher: 0.863 (95% CI, 0.808-0.906), 0.860 (95% CI, 0.806-0.904), 0.934 (95% CI, 0.891-0.963), and 0.921 (95% CI, 0.876-0.954), respectively. The SVM_Early model and SVM_Late model achieved better performance than the DT_Early model and DT_Late model (P = 0.003, 0.034, 0.008, and 0.026, respectively). In the validation cohort, the AUCs for the DT_Whole model and SVM_Whole model were 0.670 and 0.708, respectively. In comparison, the AUCs of the DT_Early model (P = 0.006), DT_Late model (P = 0.043), SVM_Early model (P = 0.001), and SVM_Late model (P = 0.007) were significantly higher: 0.839 (95% CI, 0.747-0.908), 0.784 (95% CI, 0.601-0.798), 0.890 (95% CI, 0.806-0.946), and 0.865 (95% CI, 0.777-0.928), respectively.

Conclusion

The texture features from intratumoral subregions of breast DCE-MRI showed potential in identifying benign and malignant breast tumors.

Pattern Classification Approaches for Breast Cancer Identification via MRI: State-Of-The-Art and Vision for the Future

Mining algorithms for Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) of breast tissue are discussed. The algorithms are based on recent advances in multi-dimensional signal processing and aim to advance current state-of-the-art computer-aided detection and analysis of breast tumours when these are observed at various states of development. The topics discussed include image feature extraction, information fusion using radiomics, multi-parametric computer-aided classification and diagnosis using information fusion of tensorial datasets as well as Clifford algebra based classification approaches and convolutional neural network deep learning methodologies. The discussion also extends to semi-supervised deep learning and self-supervised strategies as well as generative adversarial networks and algorithms using generated confrontational learning approaches. In order to address the problem of weakly labelled tumour images, generative adversarial deep learning strategies are considered for the classification of different tumour types. The proposed data fusion approaches provide a novel Artificial Intelligence (AI) based framework for more robust image registration that can potentially advance the early identification of heterogeneous tumour types, even when the associated imaged organs are registered as separate entities embedded in more complex geometric spaces. Finally, the general structure of a high-dimensional medical imaging analysis platform that is based on multi-task detection and learning is proposed as a way forward. The proposed algorithm makes use of novel loss functions that form the building blocks for a generated confrontation learning methodology that can be used for tensorial DCE-MRI. Since some of the approaches discussed are also based on time-lapse imaging, conclusions on the rate of proliferation of the disease can be made possible. The proposed framework can potentially reduce the costs associated with the interpretation of medical images by providing automated, faster and more consistent diagnosis.

Multi-view radiomics and dosiomics analysis with machine learning for predicting acute-phase weight loss in lung cancer patients treated with radiotherapy

We propose a multi-view data analysis approach using radiomics and dosiomics (R&D) texture features for predicting acute-phase weight loss (WL) in lung cancer radiotherapy. Baseline weight of 388 patients who underwent intensity modulated radiation therapy (IMRT) was measured between one month prior to and one week after the start of IMRT. Weight change between one week and two months after the commencement of IMRT was analyzed, and dichotomized at 5% WL. Each patient had a planning CT and contours of gross tumor volume (GTV) and esophagus (ESO). A total of 355 features including clinical parameter (CP), GTV and ESO (GTV&ESO) dose-volume histogram (DVH), GTV radiomics, and GTV&ESO dosiomics features were extracted. R&D features were categorized as first- (L1), second- (L2), higher-order (L3) statistics, and three combined groups, L1 + L2, L2 + L3 and L1 + L2 + L3. Multi-view texture analysis was performed to identify optimal R&D input features. In the training set (194 earlier patients), feature selection was performed using Boruta algorithm followed by collinearity removal based on variance inflation factor. Machine-learning models were developed using Laplacian kernel support vector machine (lpSVM), deep neural network (DNN) and their averaged ensemble classifiers. Prediction performance was tested on an independent test set (194 more recent patients), and compared among seven different input conditions: CP-only, DVH-only, R&D-only, DVH + CP, R&D + CP, R&D + DVH and R&D + DVH + CP. Combined GTV L1 + L2 + L3 radiomics and GTV&ESO L3 dosiomics were identified as optimal input features, which achieved the best performance with an ensemble classifier (AUC = 0.710), having statistically significantly higher predictability compared with DVH and/or CP features (p < 0.05). When this performance was compared to that with full R&D-only features which reflect traditional single-view data, there was a statistically significant difference (p < 0.05). Using optimized multi-view R&D input features is beneficial for predicting early WL in lung cancer radiotherapy, leading to improved performance compared to using conventional DVH and/or CP features.

Current Status and Future Perspectives of Artificial Intelligence in Magnetic Resonance Breast Imaging

Recent advances in artificial intelligence (AI) and deep learning (DL) have impacted many scientific fields including biomedical maging. Magnetic resonance imaging (MRI) is a well-established method in breast imaging with several indications including screening, staging, and therapy monitoring. The rapid development and subsequent implementation of AI into clinical breast MRI has the potential to affect clinical decision-making, guide treatment selection, and improve patient outcomes. The goal of this review is to provide a comprehensive picture of the current status and future perspectives of AI in breast MRI. We will review DL applications and compare them to standard data-driven techniques. We will emphasize the important aspect of developing quantitative imaging biomarkers for precision medicine and the potential of breast MRI and DL in this context. Finally, we will discuss future challenges of DL applications for breast MRI and an AI-augmented clinical decision strategy.

Decision support system for breast lesions via dynamic contrast enhanced magnetic resonance imaging

The presented study aims to design a computer-aided detection and diagnosis system for breast dynamic contrast enhanced magnetic resonance imaging. In the proposed system, the segmentation task is performed in two stages. The first stage is called breast region segmentation in which adaptive noise filtering, local adaptive thresholding, connected component analysis, integral of horizontal projection, and breast region of interest detection algorithms are applied to the breast images consecutively. The second stage of segmentation is breast lesion detection that consists of 32-class Otsu thresholding and Markov random field techniques. Histogram, gray level co-occurrence matrix and neighboring gray tone difference matrix based feature extraction, Fisher score based feature selection and, tenfold and leave-one-out cross-validation steps are carried out after segmentation to increase the reliability of the designed system while decreasing the computational time. Finally, support vector machines, k- nearest neighbor, and artificial neural network classifiers are performed to separate the breast lesions as benign and malignant. The average accuracy, sensitivity, specificity, and positive predictive values of each classifier are calculated and the best results are compared with the existing similar studies. According to the achieved results, the proposed decision support system for breast lesion segmentation distinguishes the breast lesions with 86%, 100%, 67%, and 85% accuracy, sensitivity, specificity, and positive predictive values, respectively. These results show that the proposed system can be used to support the radiologists during a breast cancer diagnosis.

Empirical assessment of bias in machine learning diagnostic test accuracy studies

OBJECTIVE

Machine learning (ML) diagnostic tools have significant potential to improve health care. However, methodological pitfalls may affect diagnostic test accuracy studies used to appraise such tools. We aimed to evaluate the prevalence and reporting of design characteristics within the literature. Further, we sought to empirically assess whether design features may be associated with different estimates of diagnostic accuracy.

MATERIALS AND METHODS

We systematically retrieved 2 × 2 tables (n = 281) describing the performance of ML diagnostic tools, derived from 114 publications in 38 meta-analyses, from PubMed. Data extracted included test performance, sample sizes, and design features. A mixed-effects metaregression was run to quantify the association between design features and diagnostic accuracy.

RESULTS

Participant ethnicity and blinding in test interpretation was unreported in 90% and 60% of studies, respectively. Reporting was occasionally lacking for rudimentary characteristics such as study design (28% unreported). Internal validation without appropriate safeguards was used in 44% of studies. Several design features were associated with larger estimates of accuracy, including having unreported (relative diagnostic odds ratio [RDOR], 2.11; 95% confidence interval [CI], 1.43-3.1) or case-control study designs (RDOR, 1.27; 95% CI, 0.97-1.66), and recruiting participants for the index test (RDOR, 1.67; 95% CI, 1.08-2.59).

DISCUSSION

Significant underreporting of experimental details was present. Study design features may affect estimates of diagnostic performance in the ML diagnostic test accuracy literature.

CONCLUSIONS

The present study identifies pitfalls that threaten the validity, generalizability, and clinical value of ML diagnostic tools and provides recommendations for improvement.

Artificial Intelligence for Breast MRI in 2008-2018: A Systematic Mapping Review

OBJECTIVE

The purpose of this study is to review literature from the past decade on applications of artificial intelligence (AI) to breast MRI.

MATERIALS AND METHODS

In June 2018, a systematic search of the literature was performed to identify articles on the use of AI in breast MRI. For each article identified, the surname of the first author, year of publication, journal of publication, Web of Science Core Collection journal category, country of affiliation of the first author, study design, dataset, study aim(s), AI methods used, and, when available, diagnostic performance were recorded.

RESULTS

Sixty-seven studies, 58 (87%) of which had a retrospective design, were analyzed. When journal categories were considered, 36% of articles were identified as being included in the radiology and imaging journal category. Contrast-enhanced sequences were used for most AI applications (n = 50; 75%) and, on occasion, were combined with other MRI sequences (n = 8; 12%). Four main clinical aims were addressed: breast lesion classification (n = 36; 54%), image processing (n = 14; 21%), prognostic imaging (n = 9; 13%), and response to neoadjuvant therapy (n = 8; 12%). Artificial neural networks, support vector machines, and clustering were the most frequently used algorithms, accounting for 66%. The performance achieved and the most frequently used techniques were then analyzed according to specific clinical aims. Supervised learning algorithms were primarily used for lesion characterization, with the AUC value from ROC analysis ranging from 0.74 to 0.98 (median, 0.87) and with that from prognostic imaging ranging from 0.62 to 0.88 (median, 0.80), whereas unsupervised learning was mainly used for image processing purposes.

CONCLUSION

Interest in the application of advanced AI methods to breast MRI is growing worldwide. Although this growth is encouraging, the current performance of AI applications in breast MRI means that such applications are still far from being incorporated into clinical practice.

Correlation Between Tumor Metabolism and Semiquantitative Perfusion Magnetic Resonance Imaging Metrics in Non-Small Cell Lung Cancer

PURPOSE

To correlate semiquantitative parameters derived from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG-PET) for non-small cell lung cancer (NSCLC).

METHODS AND MATERIALS

Twenty-four NSCLC patients who underwent pretreatment ¹⁸F-FDG-PET and DCE-MRI were analyzed. The maximum standardized uptake value (SUVmax) was measured from 18F-FDG-PET. Dynamic contrast-enhanced MRI was obtained on a 3T MRI scanner using 4-dimensional T1-weighted high-resolution imaging with a volume excitation sequence. The DCE-MRI parameters, consisting of mean, median, standard deviation (SD), and median absolute deviation (MAD) of peak enhancement, time to peak (TTP), time to half peak (TTHP), wash-in slope (WIS), wash-out slope (WOS), initial gradient, wash-out gradient, signal enhancement ratio, and initial area under the relative signal enhancement curve taken up to 30, 60, 90, 120, 150, and 180 seconds, TTP, and TTHP (IAUCtthp), were calculated for each lesion. Univariate analysis (UVA) was performed using Spearman correlation. A linear regression model to predict SUVmax from DCE-MRI parameters was developed by multivariate analysis (MVA) using least absolute shrinkage selection operator in combination with leave-one-out cross-validation (LOOCV).

RESULTS

In UVA, mean(WOS) (ρ = -0.456, P = .025), mean(IAUCtthp) (ρ = -0.439, P = .032), median(IAUCtthp) (ρ = -0.543, P = .006), and MAD(IAUCtthp) (ρ = -0.557, P = .005) were statistically significant; all these parameters were negatively correlated with SUVmax. In MVA, a linear combination of SD(WIS), SD(TTP), MAD(TTHP), and MAD(IAUCtthp) was statistically significant for predicting SUVmax (LOOCV-based adjusted R² = 0.298, P = .0006). A decrease in SD(WIS), MAD(TTHP), and MAD(IAUCtthp) and an increase in SD(TTP) were associated with a significant increase in SUVmax.

CONCLUSION

An association was found between SUVmax, the SD, and MAD of DCE-MRI metrics derived during contrast uptake in NSCLC, reflecting that intratumoral heterogeneity in wash-in contrast kinetics is associated with tumor metabolism. Although MAD(IAUCtthp) was a significant feature in both UVA and MVA, the LASSO-based multivariate regression model yielded better predictability of SUVmax than a univariate regression model using MAD(IAUCtthp). This study will facilitate understanding of the complex relationship between tumor vascularization and metabolism and eventually help in guiding targeted therapy.

Computer-Aided Diagnosis Scheme for Distinguishing Between Benign and Malignant Masses in Breast DCE-MRI

Our purpose in this study was to develop a computer-aided diagnosis (CAD) scheme for distinguishing between benign and malignant breast masses in dynamic contrast material-enhanced magnetic resonance imaging (DCE-MRI). Our database consisted 90 DCE-MRI examinations, each of which contained four sequential phase images; this database included 28 benign masses and 62 malignant masses. In our CAD scheme, we first determined 11 objective features of masses by taking into account the image features and the dynamic changes in signal intensity that experienced radiologists commonly use for describing masses in DCE-MRI. Quadratic discriminant analysis (QDA) was employed to distinguish between benign and malignant masses. As the input of the QDA, a combination of four objective features was determined among the 11 objective features according to a stepwise method. These objective features were as follows: (i) the change in signal intensity from 2 to 5 min; (ii) the change in signal intensity from 0 to 2 min; (iii) the irregularity of the shape; and (iv) the smoothness of the margin. Using this approach, the classification accuracy, sensitivity, and specificity were shown to be 85.6 % (77 of 90), 87.1 % (54 of 62), and 82.1 % (23 of 28), respectively. Furthermore, the positive and negative predictive values were 91.5 % (54 of 59) and 74.2 % (23 of 31), respectively. Our CAD scheme therefore exhibits high classification accuracy and is useful in the differential diagnosis of masses in DCE-MRI images.

Textural analysis of early-phase spatiotemporal changes in contrast enhancement of breast lesions imaged with an ultrafast DCE-MRI protocol

PURPOSE

New ultrafast view-sharing sequences have enabled breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to be performed at high spatial and temporal resolution. The aim of this study is to evaluate the diagnostic potential of textural features that quantify the spatiotemporal changes of the contrast-agent uptake in computer-aided diagnosis of malignant and benign breast lesions imaged with high spatial and temporal resolution DCE-MRI.

METHOD

The proposed approach is based on the textural analysis quantifying the spatial variation of six dynamic features of the early-phase contrast-agent uptake of a lesion's largest cross-sectional area. The textural analysis is performed by means of the second-order gray-level co-occurrence matrix, gray-level run-length matrix and gray-level difference matrix. This yields 35 textural features to quantify the spatial variation of each of the six dynamic features, providing a feature set of 210 features in total. The proposed feature set is evaluated based on receiver operating characteristic (ROC) curve analysis in a cross-validation scheme for random forests (RF) and two support vector machine classifiers, with linear and radial basis function (RBF) kernel. Evaluation is done on a dataset with 154 breast lesions (83 malignant and 71 benign) and compared to a previous approach based on 3D morphological features and the average and standard deviation of the same dynamic features over the entire lesion volume as well as their average for the smaller region of the strongest uptake rate.

RESULT

The area under the ROC curve (AUC) obtained by the proposed approach with the RF classifier was 0.8997, which was significantly higher (P = 0.0198) than the performance achieved by the previous approach (AUC = 0.8704) on the same dataset. Similarly, the proposed approach obtained a significantly higher result for both SVM classifiers with RBF (P = 0.0096) and linear kernel (P = 0.0417) obtaining AUC of 0.8876 and 0.8548, respectively, compared to AUC values of previous approach of 0.8562 and 0.8311, respectively.

CONCLUSION

The proposed approach based on 2D textural features quantifying spatiotemporal changes of the contrast-agent uptake significantly outperforms the previous approach based on 3D morphology and dynamic analysis in differentiating the malignant and benign breast lesions, showing its potential to aid clinical decision making.

Tensor based multichannel reconstruction for breast tumours identification from DCE-MRIs

A new methodology based on tensor algebra that uses a higher order singular value decomposition to perform three-dimensional voxel reconstruction from a series of temporal images obtained using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is proposed. Principal component analysis (PCA) is used to robustly extract the spatial and temporal image features and simultaneously de-noise the datasets. Tumour segmentation on enhanced scaled (ES) images performed using a fuzzy C-means (FCM) cluster algorithm is compared with that achieved using the proposed tensorial framework. The proposed algorithm explores the correlations between spatial and temporal features in the tumours. The multi-channel reconstruction enables improved breast tumour identification through enhanced de-noising and improved intensity consistency. The reconstructed tumours have clear and continuous boundaries; furthermore the reconstruction shows better voxel clustering in tumour regions of interest. A more homogenous intensity distribution is also observed, enabling improved image contrast between tumours and background, especially in places where fatty tissue is imaged. The fidelity of reconstruction is further evaluated on the basis of five new qualitative metrics. Results confirm the superiority of the tensorial approach. The proposed reconstruction metrics should also find future applications in the assessment of other reconstruction algorithms.

Associations between Tumor Vascularity, Vascular Endothelial Growth Factor Expression and PET/MRI Radiomic Signatures in Primary Clear-Cell-Renal-Cell-Carcinoma: Proof-of-Concept Study

Studies have shown that tumor angiogenesis is an essential process for tumor growth, proliferation and metastasis. Also, tumor angiogenesis is an important prognostic factor of clear cell renal cell carcinoma (ccRCC), as well as a factor in guiding treatment with antiangiogenic agents. Here, we attempted to find the associations between tumor angiogenesis and radiomic imaging features from PET/MRI. Specifically, sparse canonical correlation analysis was conducted on 3 feature datasets (i.e., radiomic imaging features, tumor microvascular density (MVD), and vascular endothelial growth factor (VEGF) expression) from 9 patients with primary ccRCC. In order to overcome the potential bias of intratumoral heterogeneity of angiogenesis, this study investigated the relationship between regional expressions of angiogenesis and VEGF, and localized radiomic features from different parts within the tumors. Our study highlighted the significant strong correlations between radiomic features and MVD, and also demonstrated that the spatiotemporal features extracted from DCE-MRI provided stronger radiomic correlation to MVD than the textural features extracted from Dixon sequences and FDG PET. Furthermore, PET/MRI, which takes advantage of the combined functional and structural information, had higher radiomics correlation to MVD than solely utilizing PET or MRI alone.

Exploring the complementarity of THz pulse imaging and DCE-MRIs: Toward a unified multi-channel classification and a deep learning framework

We provide a comprehensive account of recent advances in biomedical image analysis and classification from two complementary imaging modalities: terahertz (THz) pulse imaging and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). The work aims to highlight underlining commonalities in both data structures so that a common multi-channel data fusion framework can be developed. Signal pre-processing in both datasets is discussed briefly taking into consideration advances in multi-resolution analysis and model based fractional order calculus system identification. Developments in statistical signal processing using principal component and independent component analysis are also considered. These algorithms have been developed independently by the THz-pulse imaging and DCE-MRI communities, and there is scope to place them in a common multi-channel framework to provide better software standardization at the pre-processing de-noising stage. A comprehensive discussion of feature selection strategies is also provided and the importance of preserving textural information is highlighted. Feature extraction and classification methods taking into consideration recent advances in support vector machine (SVM) and extreme learning machine (ELM) classifiers and their complex extensions are presented. An outlook on Clifford algebra classifiers and deep learning techniques suitable to both types of datasets is also provided. The work points toward the direction of developing a new unified multi-channel signal processing framework for biomedical image analysis that will explore synergies from both sensing modalities for inferring disease proliferation.

Assessment of Treatment Response With Diffusion-Weighted MRI and Dynamic Contrast-Enhanced MRI in Patients With Early-Stage Breast Cancer Treated With Single-Dose Preoperative Radiotherapy: Initial Results

Single-dose preoperative stereotactic body radiotherapy is a novel radiotherapy technique for the early-stage breast cancer, and the treatment response pattern of this technique needs to be investigated on a quantitative basis. In this work, dynamic contrast-enhanced magnetic resonance imaging and diffusion-weighted magnetic resonance imaging were used to study the treatment response pattern in a unique cohort of patients with early-stage breast cancer treated with preoperative radiation. Fifteen female qualified patients received single-dose preoperative radiotherapy with 1 of the 3 prescription doses: 15 Gy, 18 Gy, and 21 Gy. Magnetic resonance imaging scans including both diffusion-weighted magnetic resonance imaging and dynamic contrast-enhanced magnetic resonance imaging were acquired before radiotherapy for planning and after radiotherapy but before surgical resection. In diffusion-weighted magnetic resonance imaging, the regional averaged apparent diffusion coefficient was calculated. In dynamic contrast-enhanced magnetic resonance imaging, quantitative parameters K (trans) and v e were evaluated using the standard Tofts model based on the average contrast agent concentration within the region of interest, and the semiquantitative initial area under the concentration curve (iAUC6min) was also recorded. These parameters' relative changes after radiotherapy were calculated for gross tumor volume, clinical target volume, and planning target volume. The initial results showed that after radiotherapy, initial area under the concentration curve significantly increased in planning target volume (P < .006) and clinical target volume (P < .006), and v e significantly increased in planning target volume (P < .05) and clinical target volume (P < .05). Statistical studies suggested that linear correlations between treatment dose and the observed parameter changes exist in most examined tests, and among these tests, the change in gross tumor volume regional averaged apparent diffusion coefficient (P < .012) and between treatment dose and planning target volume K (trans) (P < .029) were found to be statistically significant. Although it is still preliminary, this pilot study may be useful to provide insights for future works.

Pattern Recognition Approaches for Breast Cancer DCE-MRI Classification: A Systematic Review

We performed a systematic review of several pattern analysis approaches for classifying breast lesions using dynamic, morphological, and textural features in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Several machine learning approaches, namely artificial neural networks (ANN), support vector machines (SVM), linear discriminant analysis (LDA), tree-based classifiers (TC), and Bayesian classifiers (BC), and features used for classification are described. The findings of a systematic review of 26 studies are presented. The sensitivity and specificity are respectively 91 and 83 % for ANN, 85 and 82 % for SVM, 96 and 85 % for LDA, 92 and 87 % for TC, and 82 and 85 % for BC. The sensitivity and specificity are respectively 82 and 74 % for dynamic features, 93 and 60 % for morphological features, 88 and 81 % for textural features, 95 and 86 % for a combination of dynamic and morphological features, and 88 and 84 % for a combination of dynamic, morphological, and other features. LDA and TC have the best performance. A combination of dynamic and morphological features gives the best performance.

Breast MRI contrast enhancement kinetics of normal parenchyma correlate with presence of breast cancer

Background

We investigated dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) contrast enhancement kinetic variables quantified from normal breast parenchyma for association with presence of breast cancer, in a case-control study.

Methods

Under a Health Insurance Portability and Accountability Act compliant and Institutional Review Board-approved protocol, DCE-MRI scans of the contralateral breasts of 51 patients with cancer and 51 controls (matched by age and year of MRI) with biopsy-proven benign lesions were retrospectively analyzed. Applying fully automated computer algorithms on pre-contrast and multiple post-contrast MR sequences, two contrast enhancement kinetic variables, wash-in slope and signal enhancement ratio, were quantified from normal parenchyma of the contralateral breasts of both patients with cancer and controls. Conditional logistic regression was employed to assess association between these two measures and presence of breast cancer, with adjustment for other imaging factors including mammographic breast density and MRI background parenchymal enhancement (BPE). The area under the receiver operating characteristic curve (AUC) was used to assess the ability of the kinetic measures to distinguish patients with cancer from controls.

Results

When both kinetic measures were included in conditional logistic regression analysis, the odds ratio for breast cancer was 1.7 (95 % CI 1.1, 2.8; p = 0.017) for wash-in slope variance and 3.5 (95 % CI 1.2, 9.9; p = 0.019) for signal enhancement ratio volume, respectively. These odds ratios were similar on respective univariate analysis, and remained significant after adjustment for menopausal status, family history, and mammographic density. While percent BPE was associated with an odds ratio of 3.1 (95 % CI 1.2, 7.9; p = 0.018), in multivariable analysis of the three measures, percent BPE was non-significant (p = 0.897) and the two kinetics measures remained significant. For the differentiation of patients with cancer and controls, the unadjusted AUC was 0.71 using a combination of the two measures, which significantly (p = 0.005) outperformed either measure alone (AUC = 0.65 for wash-in slope variance and 0.63 for signal enhancement ratio volume).

Conclusions

Kinetic measures of wash-in slope and signal enhancement ratio quantified from normal parenchyma in DCE-MRI are jointly associated with presence of breast cancer, even after adjustment for mammographic density and BPE.

Pattern identification of biomedical images with time series: Contrasting THz pulse imaging with DCE-MRIs

OBJECTIVE

We provide a survey of recent advances in biomedical image analysis and classification from emergent imaging modalities such as terahertz (THz) pulse imaging (TPI) and dynamic contrast-enhanced magnetic resonance images (DCE-MRIs) and identification of their underlining commonalities.

METHODS

Both time and frequency domain signal pre-processing techniques are considered: noise removal, spectral analysis, principal component analysis (PCA) and wavelet transforms. Feature extraction and classification methods based on feature vectors using the above processing techniques are reviewed. A tensorial signal processing de-noising framework suitable for spatiotemporal association between features in MRI is also discussed.

VALIDATION

Examples where the proposed methodologies have been successful in classifying TPIs and DCE-MRIs are discussed.

RESULTS

Identifying commonalities in the structure of such heterogeneous datasets potentially leads to a unified multi-channel signal processing framework for biomedical image analysis.

CONCLUSION

The proposed complex valued classification methodology enables fusion of entire datasets from a sequence of spatial images taken at different time stamps; this is of interest from the viewpoint of inferring disease proliferation. The approach is also of interest for other emergent multi-channel biomedical imaging modalities and of relevance across the biomedical signal processing community.

Identifying Triple-Negative Breast Cancer Using Background Parenchymal Enhancement Heterogeneity on Dynamic Contrast-Enhanced MRI: A Pilot Radiomics Study

OBJECTIVES

To determine the added discriminative value of detailed quantitative characterization of background parenchymal enhancement in addition to the tumor itself on dynamic contrast-enhanced (DCE) MRI at 3.0 Tesla in identifying "triple-negative" breast cancers.

MATERIALS AND METHODS

In this Institutional Review Board-approved retrospective study, DCE-MRI of 84 women presenting 88 invasive carcinomas were evaluated by a radiologist and analyzed using quantitative computer-aided techniques. Each tumor and its surrounding parenchyma were segmented semi-automatically in 3-D. A total of 85 imaging features were extracted from the two regions, including morphologic, densitometric, and statistical texture measures of enhancement. A small subset of optimal features was selected using an efficient sequential forward floating search algorithm. To distinguish triple-negative cancers from other subtypes, we built predictive models based on support vector machines. Their classification performance was assessed with the area under receiver operating characteristic curve (AUC) using cross-validation.

RESULTS

Imaging features based on the tumor region achieved an AUC of 0.782 in differentiating triple-negative cancers from others, in line with the current state of the art. When background parenchymal enhancement features were included, the AUC increased significantly to 0.878 (p<0.01). Similar improvements were seen in nearly all subtype classification tasks undertaken. Notably, amongst the most discriminating features for predicting triple-negative cancers were textures of background parenchymal enhancement.

CONCLUSIONS

Considering the tumor as well as its surrounding parenchyma on DCE-MRI for radiomic image phenotyping provides useful information for identifying triple-negative breast cancers. Heterogeneity of background parenchymal enhancement, characterized by quantitative texture features on DCE-MRI, adds value to such differentiation models as they are strongly associated with the triple-negative subtype. Prospective validation studies are warranted to confirm these findings and determine potential implications.

Integration of DCE-MRI and DW-MRI Quantitative Parameters for Breast Lesion Classification

OBJECTIVE

The purpose of our study was to evaluate the diagnostic value of an imaging protocol combining dynamic contrast-enhanced MRI (DCE-MRI) and diffusion-weighted MRI (DW-MRI) in patients with suspicious breast lesions.

MATERIALS AND METHODS

A total of 31 breast lesions (15 malignant and 16 benign proved by histological examination) in 26 female patients were included in this study. For both DCE-MRI and DW-MRI model free and model based parameters were computed pixel by pixel on manually segmented ROIs. Statistical procedures included conventional linear analysis and more advanced techniques for classification of lesions in benign and malignant.

RESULTS

Our findings indicated no strong correlation between DCE-MRI and DW-MRI parameters. Results of classification analysis show that combining of DCE parameters or DW-MRI parameter, in comparison of single feature, does not yield a dramatic improvement of sensitivity and specificity of the two techniques alone. The best performance was obtained considering a full combination of all features. Moreover, the classification results combining all features are dominated by DCE-MRI features alone.

CONCLUSION

The combination of DWI and DCE-MRI does not show a potential to dramatically increase the sensitivity and specificity of breast MRI. DCE-MRI alone gave the same performance as in combination with DW-MRI.

Breast DCE-MRI Kinetic Heterogeneity Tumor Markers: Preliminary Associations With Neoadjuvant Chemotherapy Response

The ability to predict response to neoadjuvant chemotherapy for women diagnosed with breast cancer, either before or early on in treatment, is critical to judicious patient selection and tailoring the treatment regimen. In this paper, we investigate the role of contrast agent kinetic heterogeneity features derived from breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for predicting treatment response. We propose a set of kinetic statistic descriptors and present preliminary results showing the discriminatory capacity of the proposed descriptors for predicting complete and non-complete responders as assessed from pre-treatment imaging exams. The study population consisted of 15 participants: 8 complete responders and 7 non-complete responders. Using the proposed kinetic features, we trained a leave-one-out logistic regression classifier that performs with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.84 under the ROC. We compare the predictive value of our features against commonly used MRI features including kinetics of the characteristic kinetic curve (CKC), maximum peak enhancement (MPE), hotspot signal enhancement ratio (SER), and longest tumor diameter that give lower AUCs of 0.71, 0.66, 0.64, and 0.54, respectively. Our proposed kinetic statistics thus outperform the conventional kinetic descriptors as well as the classifier using a combination of all the conventional descriptors (i.e., CKC, MPE, SER, and longest diameter), which gives an AUC of 0.74. These findings suggest that heterogeneity-based DCE-MRI kinetic statistics could serve as potential imaging biomarkers for tumor characterization and could be used to improve candidate patient selection even before the start of the neoadjuvant treatment.

Comparison of dynamic contrast-enhanced MRI parameters of breast lesions at 1.5 and 3.0 T: a pilot study

OBJECTIVE

To compare dynamic contrast-enhanced (DCE) MRI parameters from scans of breast lesions at 1.5 and 3.0 T.

METHODS

11 patients underwent paired MRI examinations in both Philips 1.5 and 3.0 T systems (Best, Netherlands) using a standard clinical fat-suppressed, T1 weighted DCE-MRI protocol, with 70-76 s temporal resolution. Signal intensity vs time curves were fit with an empirical mathematical model to obtain semi-quantitative measures of uptake and washout rates as well as time-to-peak enhancement (TTP). Maximum percent enhancement and signal enhancement ratio (SER) were also measured for each lesion. Percent differences between parameters measured at the two field strengths were compared.

RESULTS

TTP and SER parameters measured at 1.5 and 3.0 T were similar; with mean absolute differences of 19% and 22%, respectively. Maximum percent signal enhancement was significantly higher at 3 T than at 1.5 T (p = 0.006). Qualitative assessment showed that image quality was significantly higher at 3 T (p = 0.005).

CONCLUSION

Our results suggest that TTP and SER are more robust to field strength change than other measured kinetic parameters, and therefore measurements of these parameters can be more easily standardized than measurements of other parameters derived from DCE-MRI. Semi-quantitative measures of overall kinetic curve shape showed higher reproducibility than do discrete classification of kinetic curve early and delayed phases in a majority of the cases studied.

ADVANCES IN KNOWLEDGE

Qualitative measures of curve shape are not consistent across field strength even when acquisition parameters are standardized. Quantitative measures of overall kinetic curve shape, by contrast, have higher reproducibility.

Breast Cancer: Radiogenomic Biomarker Reveals Associations among Dynamic Contrast-enhanced MR Imaging, Long Noncoding RNA, and Metastasis

PURPOSE

To perform a radiogenomic analysis of women with breast cancer to study the multiscale relationships among quantitative computer vision-extracted dynamic contrast material-enhanced (DCE) magnetic resonance (MR) imaging phenotypes, early metastasis, and long noncoding RNA (lncRNA) expression determined by means of high-resolution next-generation RNA sequencing.

MATERIALS AND METHODS

In this institutional review board-approved study, an automated image analysis platform extracted 47 computational quantitative features from DCE MR imaging data in a training set (n = 19) to screen for MR imaging biomarkers indicative of poor metastasis-free survival (MFS). The lncRNA molecular landscape of the candidate feature was defined by using an RNA sequencing-specific negative binomial distribution differential expression analysis. Then, this radiogenomic biomarker was applied prospectively to a validation set (n = 42) to allow prediction of MFS and lncRNA expression by using quantitative polymerase chain reaction analysis.

RESULTS

The quantitative MR imaging feature, enhancing rim fraction score, was predictive of MFS in the training set (P = .007). RNA sequencing analysis yielded an average of 55.7 × 10(6) reads per sample and identified 14 880 lncRNAs from a background of 189 883 transcripts per sample. Radiogenomic analysis allowed identification of three previously uncharacterized and five named lncRNAs significantly associated with high enhancing rim fraction, including Homeobox transcript antisense intergenic RNA (HOTAIR) (P < .05), a known predictor of poor MFS in patients with breast cancer. Independent validation confirmed the association of the enhancing rim fraction phenotype with both MFS (P = .002) and expression of four of the top five differentially expressed lncRNAs (P < .05), including HOTAIR.

CONCLUSION

The enhancing rim fraction score, a quantitative DCE MR imaging lncRNA radiogenomic biomarker, is associated with early metastasis and expression of the known predictor of metastatic progression, HOTAIR.

Review of treatment assessment using DCE-MRI in breast cancer radiation therapy

As a noninvasive functional imaging technique, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is being used in oncology to measure properties of tumor microvascular structure and permeability. Studies have shown that parameters derived from certain pharmacokinetic models can be used as imaging biomarkers for tumor treatment response. The use of DCE-MRI for quantitative and objective assessment of radiation therapy has been explored in a variety of methods and tumor types. However, due to the complexity in imaging technology and divergent outcomes from different pharmacokinetic approaches, the method of using DCE-MRI in treatment assessment has yet to be standardized, especially for breast cancer. This article reviews the basic principles of breast DCE-MRI and recent studies using DCE-MRI in treatment assessment. Technical and clinical considerations are emphasized with specific attention to assessment of radiation treatment response.

Heterogeneity wavelet kinetics from DCE-MRI for classifying gene expression based breast cancer recurrence risk

Breast tumors are heterogeneous lesions. Intra-tumor heterogeneity presents a major challenge for cancer diagnosis and treatment. Few studies have worked on capturing tumor heterogeneity from imaging. Most studies to date consider aggregate measures for tumor characterization. In this work we capture tumor heterogeneity by partitioning tumor pixels into subregions and extracting heterogeneity wavelet kinetic (HetWave) features from breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) to obtain the spatiotemporal patterns of the wavelet coefficients and contrast agent uptake from each partition. Using a genetic algorithm for feature selection, and a logistic regression classifier with leave one-out cross validation, we tested our proposed HetWave features for the task of classifying breast cancer recurrence risk. The classifier based on our features gave an ROC AUC of 0.78, outperforming previously proposed kinetic, texture, and spatial enhancement variance features which give AUCs of 0.69, 0.64, and 0.65, respectively.

Computerized image analysis for identifying triple-negative breast cancers and differentiating them from other molecular subtypes of breast cancer on dynamic contrast-enhanced MR images: a feasibility study

PURPOSE

To determine the feasibility of using a computer-aided diagnosis (CAD) system to differentiate among triple-negative breast cancer, estrogen receptor (ER)-positive cancer, human epidermal growth factor receptor type 2 (HER2)-positive cancer, and benign fibroadenoma lesions on dynamic contrast material-enhanced (DCE) magnetic resonance (MR) images.

MATERIALS AND METHODS

This is a retrospective study of prospectively acquired breast MR imaging data collected from an institutional review board-approved, HIPAA-compliant study between 2002 and 2007. Written informed consent was obtained from all patients. The authors collected DCE MR images from 65 women with 76 breast lesions who had been recruited into a larger study of breast MR imaging. The women had triple-negative (n = 21), ER-positive (n = 25), HER2-positive (n = 18), or fibroadenoma (n = 12) lesions. All lesions were classified as Breast Imaging Reporting and Data System category 4 or higher on the basis of previous imaging. Images were subject to quantitative feature extraction, feed-forward feature selection by means of linear discriminant analysis, and lesion classification by using a support vector machine classifier. The area under the receiver operating characteristic curve (Az) was calculated for each of five lesion classification tasks involving triple-negative breast cancers.

RESULTS

For each pair-wise lesion type comparison, linear discriminant analysis helped identify the most discriminatory features, which in conjunction with a support vector machine classifier yielded an Az of 0.73 (95% confidence interval [CI]: 0.59, 0.87) for triple-negative cancer versus all non-triple-negative lesions, 0.74 (95% CI: 0.60, 0.88) for triple-negative cancer versus ER- and HER2-positive cancer, 0.77 (95% CI: 0.63, 0.91) for triple-negative versus ER-positive cancer, 0.74 (95% CI: 0.58, 0.89) for triple-negative versus HER2-positive cancer, and 0.97 (95% CI: 0.91, 1.00) for triple-negative cancer versus fibroadenoma.

CONCLUSION

Triple-negative cancers possess certain characteristic features on DCE MR images that can be captured and quantified with CAD, enabling good discrimination of triple-negative cancers from non-triple-negative cancers, as well as between triple-negative cancers and benign fibroadenomas. Such CAD algorithms may provide added diagnostic benefit in identifying the highly aggressive triple-negative cancer phenotype with DCE MR imaging in high-risk women.

Max-AUC feature selection in computer-aided detection of polyps in CT colonography

We propose a feature selection method based on a sequential forward floating selection (SFFS) procedure to improve the performance of a classifier in computerized detection of polyps in CT colonography (CTC). The feature selection method is coupled with a nonlinear support vector machine (SVM) classifier. Unlike the conventional linear method based on Wilks' lambda, the proposed method selected the most relevant features that would maximize the area under the receiver operating characteristic curve (AUC), which directly maximizes classification performance, evaluated based on AUC value, in the computer-aided detection (CADe) scheme. We presented two variants of the proposed method with different stopping criteria used in the SFFS procedure. The first variant searched all feature combinations allowed in the SFFS procedure and selected the subsets that maximize the AUC values. The second variant performed a statistical test at each step during the SFFS procedure, and it was terminated if the increase in the AUC value was not statistically significant. The advantage of the second variant is its lower computational cost. To test the performance of the proposed method, we compared it against the popular stepwise feature selection method based on Wilks' lambda for a colonic-polyp database (25 polyps and 2624 nonpolyps). We extracted 75 morphologic, gray-level-based, and texture features from the segmented lesion candidate regions. The two variants of the proposed feature selection method chose 29 and 7 features, respectively. Two SVM classifiers trained with these selected features yielded a 96% by-polyp sensitivity at false-positive (FP) rates of 4.1 and 6.5 per patient, respectively. Experiments showed a significant improvement in the performance of the classifier with the proposed feature selection method over that with the popular stepwise feature selection based on Wilks' lambda that yielded 18.0 FPs per patient at the same sensitivity level.

3D lacunarity in multifractal analysis of breast tumor lesions in dynamic contrast-enhanced magnetic resonance imaging

Dynamic contrast-enhanced magnetic resonance (DCE-MR) of the breast is especially robust for the diagnosis of cancer in high-risk women due to its high sensitivity. Its specificity may be, however, compromised since several benign masses take up contrast agent as malignant lesions do. In this paper, we propose a novel method of 3D multifractal analysis to characterize the spatial complexity (spatial arrangement of texture) of breast tumors at multiple scales. Self-similar properties are extracted from the estimation of the multifractal scaling exponent for each clinical case, using lacunarity as the multifractal measure. These properties include several descriptors of the multifractal spectra reflecting the morphology and internal spatial structure of the enhanced lesions relatively to normal tissue. The results suggest that the combined multifractal characteristics can be effective to distinguish benign and malignant findings, judged by the performance of the support vector machine classification method evaluated by receiver operating characteristics with an area under the curve of 0.96. In addition, this paper confirms the presence of multifractality in DCE-MR volumes of the breast, whereby multiple degrees of self-similarity prevail at multiple scales. The proposed feature extraction and classification method have the potential to complement the interpretation of the radiologists and supply a computer-aided diagnosis system.

A multichannel Markov random field framework for tumor segmentation with an application to classification of gene expression-based breast cancer recurrence risk

We present a methodological framework for multichannel Markov random fields (MRFs). We show that conditional independence allows loopy belief propagation to solve a multichannel MRF as a single channel MRF. We use conditional mutual information to search for features that satisfy conditional independence assumptions. Using this framework we incorporate kinetic feature maps derived from breast dynamic contrast enhanced magnetic resonance imaging as observation channels in MRF for tumor segmentation. Our algorithm based on multichannel MRF achieves an receiver operating characteristic area under curve (AUC) of 0.97 for tumor segmentation when using a radiologist's manual delineation as ground truth. Single channel MRF based on the best feature chosen from the same pool of features as used by the multichannel MRF achieved a lower AUC of 0.89. We also present a comparison against the well established normalized cuts segmentation algorithm along with commonly used approaches for breast tumor segmentation including fuzzy C-means (FCM) and the more recent method of running FCM on enhancement variance features (FCM-VES). These previous methods give a lower AUC of 0.92, 0.88, and 0.60, respectively. Finally, we also investigate the role of superior segmentation in feature extraction and tumor characterization. Specifically, we examine the effect of improved segmentation on predicting the probability of breast cancer recurrence as determined by a validated tumor gene expression assay. We demonstrate that an support vector machine classifier trained on kinetic statistics extracted from tumors as segmented by our algorithm gives a significant improvement in distinguishing between women with high and low recurrence risk, giving an AUC of 0.88 as compared to 0.79, 0.76, 0.75, and 0.66 when using normalized cuts, single channel MRF, FCM, and FCM-VES, respectively, for segmentation.

Characterization of spatiotemporal changes for the classification of dynamic contrast-enhanced magnetic-resonance breast lesions

OBJECTIVE

The early detection of breast cancer is one of the most important predictors in determining the prognosis for women with malignant tumours. Dynamic contrast-enhanced magnetic-resonance imaging (DCE-MRI) is an important imaging modality for detecting and interpreting the different breast lesions from a time sequence of images and has proved to be a very sensitive modality for breast-cancer diagnosis. However, DCE-MRI exhibits only a moderate specificity, thus leading to a high rate of false positives, resulting in unnecessary biopsies that are stressful and physically painful for the patient and lead to an increase in the cost of treatment. There is a strong medical need for a DCE-MRI computer-aided diagnosis tool that would offer a reliable support to the physician's decision providing a high level of sensitivity and specificity.

METHODS

In our study we investigated the possibility of increasing differentiation between the malignant and the benign lesions with respect to the spatial variation of the temporal enhancements of three parametric maps, i.e., the initial enhancement (IE) map, the post-initial enhancement (PIE) map and the signal enhancement ratio (SER) map, by introducing additional methods along with the grey-level co-occurrence matrix, i.e., a second-order statistical method already applied for quantifying the spatiotemporal variations. We introduced the grey-level run-length matrix and the grey-level difference matrix, representing two additional, second-order statistical methods, and the circular Gabor as a frequency-domain-based method. Each of the additional methods is for the first time applied to the DCE-MRI data to differentiate between the malignant and the benign breast lesions. We applied the least-square minimum-distance classifier (LSMD), logistic regression and least-squares support vector machine (LS-SVM) classifiers on a total of 115 (78 malignant and 37 benign) breast DCE-MRI cases. The performances were evaluated using ten experiments of a ten-fold cross-validation.

RESULTS

Our experimental analysis revealed the PIE map, together with the feature subset in which the discriminating ability of the co-occurrence features was increased by adding the newly introduced features, to be the most significant for differentiation between the malignant and the benign lesions. That diagnostic test - the aforementioned combination of parametric map and the feature subset achieved the sensitivity of 0.9193 which is statistically significantly higher compared to other diagnostic tests after ten-experiments of a ten-fold cross-validation and gave a statistically significantly higher specificity of 0.7819 for the fixed 95% sensitivity after the receiver operating characteristic (ROC) curve analysis. Combining the information from all the three parametric maps significantly increased the area under the ROC curve (AUC) of the aforementioned diagnostic test for the LSMD and logistic regression; however, not for the LS-SVM. The LSMD classifier yielded the highest area under the ROC curve when using the combined information, increasing the AUC from 0.9651 to 0.9755.

CONCLUSION

Introducing new features to those of the grey-level co-occurrence matrix significantly increased the differentiation between the malignant and the benign breast lesions, thus resulting in a high sensitivity and improved specificity.