Texture analysis on MR images helps predicting non-response to NAC in breast cancer

Efficacy of texture analysis in determining the gene amplification status of HER2 2+ for invasive ductal carcinoma cases

BACKGROUND

Gene amplification of human epidermal growth factor receptor2 (HER2) 2+ is essential to be determined for treatment planning. A search of the PubMed database indicates that the correlation between texture features from dynamic contrast enhanced (DCE)-MRI and HER2 2+ status has not been investigated extensively in invasive ductal carcinoma cases.

METHODS

Seventy-one DCE-MRI cases of HER2 2+ status verified using fluorescence in-situ hybridization (FISH) were selected, including 36 positive and 35 negative cases. Overall, 279 texture features were derived from lesion regions of interest manually drawn onto the subtraction images between pre- and post-contrast agent. Fisher coefficient, mutual information, minimization of both classification error probability and average correlation coefficients as well as a combination of all three methods (MPF) were independently used to reduce the dimensionality of texture parameters. A popular machine learning algorithm, the Support Vector Machine, was further applied to determine HER2 2+ status. Receiver operating characteristic (ROC) analysis was conducted to evaluate the classification performance.

RESULTS

Diagnostic accuracy was optimal when the most significant discriminatory features were selected using MPF. The area under ROC curve reached 0.863 with corresponding accuracy, sensitivity and specificity rates of 81.80%, 85.71% and 77.78%, respectively.

CONCLUSIONS

Texture analysis based on breast MRI delivered consistently high performance with FISH detection and may serve as a useful supplementary tool for determining the gene amplification status of HER2 2+ for cases with invasive ductal carcinoma.

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.

Machine learning on MRI radiomic features: identification of molecular subtype alteration in breast cancer after neoadjuvant therapy

OBJECTIVES

Recent studies have revealed the change of molecular subtypes in breast cancer (BC) after neoadjuvant therapy (NAT). This study aims to construct a non-invasive model for predicting molecular subtype alteration in breast cancer after NAT.

METHODS

Eighty-two estrogen receptor (ER)-negative/ human epidermal growth factor receptor 2 (HER2)-negative or ER-low-positive/HER2-negative breast cancer patients who underwent NAT and completed baseline MRI were retrospectively recruited between July 2010 and November 2020. Subtype alteration was observed in 21 cases after NAT. A 2D-DenseUNet machine-learning model was built to perform automatic segmentation of breast cancer. 851 radiomic features were extracted from each MRI sequence (T2-weighted imaging, ADC, DCE, and contrast-enhanced T1-weighted imaging), both in the manual and auto-segmentation masks. All samples were divided into a training set (n = 66) and a test set (n = 16). XGBoost model with 5-fold cross-validation was performed to predict molecular subtype alterations in breast cancer patients after NAT. The predictive ability of these models was subsequently evaluated by the AUC of the ROC curve, sensitivity, and specificity.

RESULTS

A model consisting of three radiomics features from the manual segmentation of multi-sequence MRI achieved favorable predictive efficacy in identifying molecular subtype alteration in BC after NAT (cross-validation set: AUC = 0.908, independent test set: AUC = 0.864); whereas an automatic segmentation approach of BC lesions on the DCE sequence produced good segmentation results (Dice similarity coefficient = 0.720).

CONCLUSIONS

A machine learning model based on baseline MRI is proven useful for predicting molecular subtype alterations in breast cancer after NAT.

KEY POINTS

• Machine learning models using MRI-based radiomics signature have the ability to predict molecular subtype alterations in breast cancer after neoadjuvant therapy, which subsequently affect treatment protocols. • The application of deep learning in the automatic segmentation of breast cancer lesions from MRI images shows the potential to replace manual segmentation..

Artificial intelligence reveals features associated with breast cancer neoadjuvant chemotherapy responses from multi-stain histopathologic images

Advances in computational algorithms and tools have made the prediction of cancer patient outcomes using computational pathology feasible. However, predicting clinical outcomes from pre-treatment histopathologic images remains a challenging task, limited by the poor understanding of tumor immune micro-environments. In this study, an automatic, accurate, comprehensive, interpretable, and reproducible whole slide image (WSI) feature extraction pipeline known as, IMage-based Pathological REgistration and Segmentation Statistics (IMPRESS), is described. We used both H&E and multiplex IHC (PD-L1, CD8+, and CD163+) images, investigated whether artificial intelligence (AI)-based algorithms using automatic feature extraction methods can predict neoadjuvant chemotherapy (NAC) outcomes in HER2-positive (HER2+) and triple-negative breast cancer (TNBC) patients. Features are derived from tumor immune micro-environment and clinical data and used to train machine learning models to accurately predict the response to NAC in breast cancer patients (HER2+ AUC = 0.8975; TNBC AUC = 0.7674). The results demonstrate that this method outperforms the results trained from features that were manually generated by pathologists. The developed image features and algorithms were further externally validated by independent cohorts, yielding encouraging results, especially for the HER2+ subtype.

Applying artificial intelligence technology to assist with breast cancer diagnosis and prognosis prediction

Breast cancer remains the most diagnosed cancer in women. Advances in medical imaging modalities and technologies have greatly aided in the early detection of breast cancer and the decline of patient mortality rates. However, reading and interpreting breast images remains difficult due to the high heterogeneity of breast tumors and fibro-glandular tissue, which results in lower cancer detection sensitivity and specificity and large inter-reader variability. In order to help overcome these clinical challenges, researchers have made great efforts to develop computer-aided detection and/or diagnosis (CAD) schemes of breast images to provide radiologists with decision-making support tools. Recent rapid advances in high throughput data analysis methods and artificial intelligence (AI) technologies, particularly radiomics and deep learning techniques, have led to an exponential increase in the development of new AI-based models of breast images that cover a broad range of application topics. In this review paper, we focus on reviewing recent advances in better understanding the association between radiomics features and tumor microenvironment and the progress in developing new AI-based quantitative image feature analysis models in three realms of breast cancer: predicting breast cancer risk, the likelihood of tumor malignancy, and tumor response to treatment. The outlook and three major challenges of applying new AI-based models of breast images to clinical practice are also discussed. Through this review we conclude that although developing new AI-based models of breast images has achieved significant progress and promising results, several obstacles to applying these new AI-based models to clinical practice remain. Therefore, more research effort is needed in future studies.

Prediction of the Pathological Response to Neoadjuvant Chemotherapy in Breast Cancer Patients With MRI-Radiomics: A Systematic Review and Meta-analysis

We performed a systematic review and a meta-analysis of studies using MRI-radiomics for predicting the pathological complete response in breast cancer patients undergoing neoadjuvant therapy , and we evaluated their methodological quality using the radiomics-quality-score (RQS). Random effects meta-analysis was performed pooling area under the receiver operating characteristics curves. Publication-bias was assessed using the Egger's test and visually inspecting the funnel plot. Forty-three studies were included in the qualitative review and 34 in the meta-analysis. Summary area under the receiver operating characteristics curve was 0,78 (95%CI:0,74-0,81). Heterogeneity according to the I² statistic was substantial (71%) and there was no evidence of publication bias (P-value = 0,2). The average RQS was 12,7 (range:-1-26), with an intra-class correlation coefficient of 0.93 (95%CI:0.61-0.97). Year of publication, field intensity and synthetic RQS score do not appear to be moderators of the effect (P-value = 0.36, P-value = 0.28 and P-value = 0.92, respectively). MRI-radiomics may predict response to neoadjuvant therapy in breast cancer patients but the heterogeneity of the current studies is still substantial.

Changes in kinetic heterogeneity of breast cancer via computer-aided diagnosis on MRI predict the pathological response to neoadjuvant systemic therapy

OBJECTIVE

To evaluate whether the computer-aided diagnosis (CAD)-extracted kinetic heterogeneity of breast cancer on MRI and changes therein during treatment were associated with the pathological response to neoadjuvant systemic therapy (NST).

MATERIALS AND METHODS

Consecutive patients with invasive breast cancer, who underwent NST followed by surgery between 2014 and 2020, were retrospectively evaluated. Using a commercial CAD system, kinetic features (angiovolume, peak enhancement, delayed enhancement profiles, and kinetic heterogeneity) of breast cancer were assessed with pre- and mid-treatment MRI. Multivariate logistic regression was used to identify the associations between CAD-extracted kinetic features and pathological complete response (pCR).

RESULTS

A total of 130 patients (mean age, 55 years) were included, 37 (28.5%) of whom achieved a pCR. When the pre- and mid-treatment MRI data were compared, the pCR group exhibited greater changes in kinetic heterogeneity (86.14 ± 32.05% vs. 8.50 ± 141.01%, p < 0.001) and angiovolume (95.20 ± 14.29% vs. 19.89 ± 320.16%; p < 0.001) than the non-pCR group. Multivariate regression analysis showed that a large change in kinetic heterogeneity (odds ratio (OR) = 1.030, p < 0.001), age (OR = 0.931, p = 0.005), progesterone receptor negativity (OR = 7.831, p = 0.001), and HER2 positivity (OR = 3.455, p = 0.017) were associated with pCR.

CONCLUSIONS

A greater change in the CAD-extracted kinetic heterogeneity of breast cancer between pre- and mid-treatment MRI was associated with a pCR in patients on NST.

KEY POINTS

A greater change in kinetic heterogeneity was associated with a pathological complete response. Computer-aided diagnosis-extracted kinetic heterogeneity might serve as a quantitative biomarker of therapeutic efficacy.

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.

Automated machine learning for differentiation of hepatocellular carcinoma from intrahepatic cholangiocarcinoma on multiphasic MRI

With modern management of primary liver cancer shifting towards non-invasive diagnostics, accurate tumor classification on medical imaging is increasingly critical for disease surveillance and appropriate targeting of therapy. Recent advancements in machine learning raise the possibility of automated tools that can accelerate workflow, enhance performance, and increase the accessibility of artificial intelligence to clinical researchers. We explore the use of an automated Tree-Based Optimization Tool that leverages a genetic programming algorithm for differentiation of the two common primary liver cancers on multiphasic MRI. Manual and automated analyses were performed to select an optimal machine learning model, with an accuracy of 73–75% (95% CI 0.59–0.85), sensitivity of 70–75% (95% CI 0.48–0.89), and specificity of 71–79% (95% CI 0.52–0.90) on manual optimization, and an accuracy of 73–75% (95% CI 0.59–0.85), sensitivity of 65–75% (95% CI 0.43–0.89) and specificity of 75–79% (95% CI 0.56–0.90) for automated machine learning. We found that automated machine learning performance was similar to that of manual optimization, and it could classify hepatocellular carcinoma and intrahepatic cholangiocarcinoma with an sensitivity and specificity comparable to that of radiologists. However, automated machine learning performance was poor on a subset of scans that met LI-RADS criteria for LR-M. Exploration of additional feature selection and classifier methods with automated machine learning to improve performance on LR-M cases as well as prospective validation in the clinical setting are needed prior to implementation.

Prediction for pathological and immunohistochemical characteristics of triple-negative invasive breast carcinomas: the performance comparison between quantitative and qualitative sonographic feature analysis

OBJECTIVE

Sonographic features are associated with pathological and immunohistochemical characteristics of triple-negative breast cancer (TNBC). To predict the biological property of TNBC, the performance using quantitative high-throughput sonographic feature analysis was compared with that using qualitative feature assessment.

METHODS

We retrospectively reviewed ultrasound images, clinical, pathological, and immunohistochemical (IHC) data of 252 female TNBC patients. All patients were subgrouped according to the histological grade, Ki67 expression level, and human epidermal growth factor receptor 2 (HER2) score. Qualitative sonographic feature assessment included shape, margin, posterior acoustic pattern, and calcification referring to the Breast Imaging Reporting and Data System (BI-RADS). Quantitative sonographic features were acquired based on the computer-aided radiomics analysis. Breast cancer masses were manually segmented from the surrounding breast tissues. For each ultrasound image, 1688 radiomics features of 7 feature classes were extracted. The principal component analysis (PCA), least absolute shrinkage and selection operator (LASSO), and support vector machine (SVM) were used to determine the high-throughput radiomics features that were highly correlated to biological properties. The performance using both quantitative and qualitative sonographic features to predict biological properties of TNBC was represented by the area under the receiver operating characteristic curve (AUC).

RESULTS

In the qualitative assessment, regular tumor shape, no angular or spiculated margin, posterior acoustic enhancement, and no calcification were used as the independent sonographic features for TNBC. Using the combination of these four features to predict the histological grade, Ki67, HER2, axillary lymph node metastasis (ALNM), and lymphovascular invasion (LVI), the AUC was 0.673, 0.680, 0.651, 0.587, and 0.566, respectively. The number of high-throughput features that closely correlated with biological properties was 34 for histological grade (AUC 0.942), 27 for Ki67 (AUC 0.732), 25 for HER2 (AUC 0.730), 34 for ALNM (AUC 0.804), and 34 for LVI (AUC 0.795).

CONCLUSION

High-throughput quantitative sonographic features are superior to traditional qualitative ultrasound features in predicting the biological behavior of TNBC.

KEY POINTS

• Sonographic appearances of TNBCs showed a great variety in accordance with its biological and clinical characteristics. • Both qualitative and quantitative sonographic features of TNBCs are associated with tumor biological characteristics. • The quantitative high-throughput feature analysis is superior to two-dimensional sonographic feature assessment in predicting tumor biological property.

Diagnosing breast cancer tumors using stacked ensemble model

Breast cancer is the most common cancer that progresses from cells in the breast tissue among women. Early-stage detection could reduce death rates significantly, and the detection-stage determines the treatment process. Mammography is utilized to discover breast cancer at an early stage prior to any physical sign. However, mammography might return false-negative, in which case, if it is suspected that lesions might have cancer of chance greater than two percent, a biopsy is recommended. About 30 percent of biopsies result in malignancy that means the rate of unnecessary biopsies is high. So to reduce unnecessary biopsies, recently, due to its excellent capability in soft tissue imaging, Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) has been utilized to detect breast cancer. Nowadays, DCE-MRI is a highly recommended method not only to identify breast cancer but also to monitor its development, and to interpret tumorous regions. However, in addition to being a time-consuming process, the accuracy depends on radiologists’ experience. Radiomic data, on the other hand, are used in medical imaging and have the potential to extract disease characteristics that can not be seen by the naked eye. Radiomics are hard-coded features and provide crucial information about the disease where it is imaged. Conversely, deep learning methods like convolutional neural networks(CNNs) learn features automatically from the dataset. Especially in medical imaging, CNNs’ performance is better than compared to hard-coded features-based methods. However, combining the power of these two types of features increases accuracy significantly, which is especially critical in medicine. Herein, a stacked ensemble of gradient boosting and deep learning models were developed to classify breast tumors using DCE-MRI images. The model makes use of radiomics acquired from pixel information in breast DCE-MRI images. Prior to train the model, radiomics had been applied to the factor analysis to refine the feature set and eliminate unuseful features. The performance metrics, as well as the comparisons to some well-known machine learning methods, state the ensemble model outperforms its counterparts. The ensembled model’s accuracy is 94.87% and its AUC value is 0.9728. The recall and precision are 1.0 and 0.9130, respectively, whereas F1-score is 0.9545.

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.

Classification of MR-Detected Additional Lesions in Patients With Breast Cancer Using a Combination of Radiomics Analysis and Machine Learning

Objective

This study was conducted in order to investigate the feasibility of using radiomics analysis (RA) with machine learning algorithms based on breast magnetic resonance (MR) images for discriminating malignant from benign MR-detected additional lesions in patients with primary breast cancer.

Materials and Methods

One hundred seventy-four MR-detected additional lesions (benign, n = 86; malignancy, n = 88) from 158 patients with ipsilateral primary breast cancer from a tertiary medical center were included in this retrospective study. The entire data were randomly split to training (80%) and independent test sets (20%). In addition, 25 patients (benign, n = 21; malignancy, n = 15) from another tertiary medical center were included for the external test. Radiomics features that were extracted from three regions-of-interest (ROIs; intratumor, peritumor, combined) using fat-saturated T1-weighted images obtained by subtracting pre- from postcontrast images (SUB) and T2-weighted image (T2) were utilized to train the support vector machine for the binary classification. A decision tree method was utilized to build a classifier model using clinical imaging interpretation (CII) features assessed by radiologists. Area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, and specificity were used to compare the diagnostic performance.

Results

The RA models trained using radiomics features from the intratumor-ROI showed comparable performance to the CII model (accuracy, AUROC: 73.3%, 69.6% for the SUB RA model; 70.0%, 75.1% for the T2 RA model; 73.3%, 72.0% for the CII model). The diagnostic performance increased when the radiomics and CII features were combined to build a fusion model. The fusion model that combines the CII features and radiomics features from multiparametric MRI data demonstrated the highest performance with an accuracy of 86.7% and an AUROC of 91.1%. The external test showed a similar pattern where the fusion models demonstrated higher levels of performance compared with the RA- or CII-only models. The accuracy and AUROC of the SUB+T2 RA+CII model in the external test were 80.6% and 91.4%, respectively.

Conclusion

Our study demonstrated the feasibility of using RA with machine learning approach based on multiparametric MRI for quantitatively characterizing MR-detected additional lesions. The fusion model demonstrated an improved diagnostic performance over the models trained with either RA or CII alone.

MRI texture features from tumor core and margin in the prediction of response to neoadjuvant chemotherapy in patients with locally advanced breast cancer

BACKGROUND

Radiomics involving quantitative analysis of imaging has shown promises in oncology to serve as non-invasive biomarkers. We investigated whether pre-treatment T2-weighted magnetic resonance imaging (MRI) can be used to predict response to neoadjuvant chemotherapy (NAC) in breast cancer.

MATERIALS AND METHODS

MRI scans were obtained for 102 patients with locally advanced breast cancer (LABC). All patients were treated with standard regimens of NAC as decided by the treating oncologist, followed by surgery and adjuvant treatment according to standard institutional practice. The primary tumor was segmented, and 11 texture features were extracted using the grey-level co-occurrence matrices analysis of the T2W-images from tumor cores and margins. Response assessment was done using clinical-pathological responses with patients classified into binary groups: responders and non-responders. Machine learning classifiers were used to develop a radiomics model, and a leave-one-out cross-validation technique was used to assess the performance.

RESULTS

7 features were significantly (p < 0.05) different between the two response groups. The best classification accuracy was obtained using a k-nearest neighbor (kNN) model with sensitivity, specificity, accuracy, and area under curve of 63, 93, 87, and 0.78, respectively.

CONCLUSIONS

Pre-treatment T2-weighted MRI texture features can predict NAC response with reasonable accuracy.

Value of Machine Learning with Multiphases CE-MRI Radiomics for Early Prediction of Pathological Complete Response to Neoadjuvant Therapy in HER2-Positive Invasive Breast Cancer

Background

To assess the value of radiomics based on multiphases contrast-enhanced magnetic resonance imaging (CE-MRI) for early prediction of pathological complete response (pCR) to neoadjuvant therapy (NAT) in patients with human epithelial growth factor receptor 2 (HER2) positive invasive breast cancer.

Methods

A total of 127 patients with newly diagnosed primary HER2 positive invasive breast cancer underwent CE-MRI before NAT and performed surgery after NAT. Radiomic features were extracted from the 1^st postcontrast CE-MRI phase (CE₁) and multi-phases CE-MRI (CE_m),respectively. With selected features using a forward stepwise regression, 23 machine learning classifiers based on CE₁ and CE_m were constructed respectively for differentiating pCR and non-pCR patients. The performances of classifiers were assessed and compared by their accuracy, sensitivity, specificity and AUC (area under curve). The optimal machine learning classification was used to discriminate pCR vs non-pCR in mass and non-mass groups, uni-focal and unilateral multi-focal groups, respectively.

Results

For the task of pCR classification, 6 radiomic features from CE₁ and 6 from CE_m were selected for the construction of machine learning models, respectively. The linear SVM based on CE_m outperformed the logistic regression model using CE₁ with an AUC of 0.84 versus 0.69. In mass and non-mass enhancement groups, the accuracy of linear SVM achieved 84% and 76%. Whereas in unifocal and unilateral multifocal cases, 79% and 75% accuracy were achieved by linear SVM.

Conclusion

Multiphases CE-MRI imaging may offer more heterogeneity information in the tumor and provide a non-invasive approach for early prediction of pCR to NAT in patients with HER2-positive invasive breast cancer.

Texture analysis for chemotherapy response evaluation in osteosarcoma using MR imaging

The efficacy of MRI-based statistical texture analysis (TA) in predicting chemotherapy response among patients with osteosarcoma was assessed. Forty patients (male: female = 31:9; age = 17.2 ± 5.7 years) with biopsy-proven osteosarcoma were analyzed in this prospective study. Patients were scheduled for three cycles of neoadjuvant chemotherapy (NACT) and diffusion-weighted MRI acquisition at three time points: at baseline (t0), after the first NACT (t1) and after the third NACT (t2) using a 1.5 T scanner. Eight patients (nonsurvivors) died during NACT while 34 patients (survivors) completed the NACT regimen followed by surgery. Histopathological evaluation was performed in the resected tumor to assess NACT response (responder [≤50% viable tumor] and nonresponder [>50% viable tumor]) and revealed nonresponder: responder = 20:12. Apparent diffusion coefficient (ADC) and intravoxel incoherent motion (IVIM) parameters, diffusion coefficient (D), perfusion coefficient (D*) and perfusion fraction (f) were evaluated. A total of 25 textural features were evaluated on ADC, D, D* and f parametric maps and structural T1-weighted (T1W) and T2-weighted (T2W) images in the entire tumor volume using 3D TA methods gray-level cooccurrence matrix (GLCM), neighborhood gray-tone-difference matrix (NGTDM) and run-length matrix (RLM). Receiver-operating-characteristic curve analysis was performed on the selected textural feature set to assess the role of TA features (a) as marker(s) of tumor aggressiveness leading to mortality at baseline and (b) in predicting the NACT response among survivors in the course of treatment. Findings showed that the NGTDM features coarseness, busyness and strength quantifying tumor heterogeneity in D, D* and f maps and T1W and T2W images were useful markers of tumor aggressiveness in identifying the nonsurvivor group (area-under-the-curve [AUC] = 0.82-0.88) at baseline. The GLCM features contrast and correlation, NGTDM features contrast and complexity and RLM feature short-run-low-gray-level-emphasis quantifying homogeneity/terogeneity in tumor were effective markers for predicting chemotherapeutic response using D (AUC = 0.80), D* (AUC = 0.80) and T2W (AUC = 0.70) at t0, and D* (AUC = 0.80) and f (AUC = 0.70) at t1. 3D statistical TA features might be useful as imaging-based markers for characterizing tumor aggressiveness and predicting chemotherapeutic response in patients with osteosarcoma.

Feasibility of contrast-enhanced MRI derived textural features to predict overall survival in locally advanced breast cancer

BACKGROUND

The prognosis for women with locally advanced breast cancer (LABC) is poor and there is a need for better treatment stratification. Gray-level co-occurrence matrix (GLCM) texture analysis of magnetic resonance (MR) images has been shown to predict pathological response and could become useful in stratifying patients to more targeted treatments.

PURPOSE

To evaluate the ability of GLCM textural features obtained before neoadjuvant chemotherapy to predict overall survival (OS) seven years after diagnosis of patients with LABC.

MATERIAL AND METHODS

This retrospective study includes data from 55 patients with LABC. GLCM textural features were extracted from segmented tumors in pre-treatment dynamic contrast-enhanced 3-T MR images. Prediction of OS by GLCM textural features was assessed and compared to predictions using traditional clinical variables.

RESULTS

Linear mixed-effect models showed significant differences in five GLCM features (f₁, f₂, f₅, f₁₀, f₁₁) between survivors and non-survivors. Using discriminant analysis for prediction of survival, GLCM features from 2 min post-contrast images achieved a classification accuracy of 73% (P < 0.001), whereas traditional prognostic factors resulted in a classification accuracy of 67% (P = 0.005). Using a combination of both yielded the highest classification accuracy (78%, P < 0.001). Median values for features f₁, f₂, f₁₀, and f₁₁ provided significantly different survival curves in Kaplan-Meier analysis.

CONCLUSION

This study shows a clear association between textural features from post-contrast images obtained before neoadjuvant chemotherapy and OS seven years after diagnosis. Further studies in larger cohorts should be undertaken to investigate how this prognostic information can be used to benefit treatment stratification.

Breast dynamic contrast-enhanced-magnetic resonance imaging and radiomics: State of art

Breast cancer represents the most common malignancy in women, being one of the most frequent cause of cancer-related mortality. Ultrasound, mammography, and magnetic resonance imaging (MRI) play a pivotal role in the diagnosis of breast lesions, with different levels of accuracy. Particularly, dynamic contrast-enhanced MRI has shown high diagnostic value in detecting multifocal, multicentric, or contralateral breast cancers. Radiomics is emerging as a promising tool for quantitative tumor evaluation, allowing the extraction of additional quantitative data from radiological imaging acquired with different modalities. Radiomics analysis may provide novel information through the quantification of lesions heterogeneity, that may be relevant in clinical practice for the characterization of breast lesions, prediction of tumor response to systemic therapies and evaluation of prognosis in patients with breast cancers. Several published studies have explored the value of radiomics with good-to-excellent diagnostic and prognostic performances for the evaluation of breast lesions. Particularly, the integrations of radiomics data with other clinical and histopathological parameters have demonstrated to improve the prediction of tumor aggressiveness with high accuracy and provided precise models that will help to guide clinical decisions and patients management. The purpose of this article in to describe the current application of radiomics in breast dynamic contrast-enhanced MRI.

Texture Analysis of Breast DCE-MRI Based on Intratumoral Subregions for Predicting HER2 2+ Status

Background: Breast tumor heterogeneity is related to risk factors that lead to aggressive tumor growth; however, such heterogeneity has not been thoroughly investigated.

Purpose: To evaluate the performance of texture features extracted from heterogeneity subregions on subtraction MRI images for identifying human epidermal growth factor receptor 2 (HER2) 2+ status of breast cancers.

Materials and Methods: Seventy-six patients with HER2 2+ breast cancer who underwent dynamic contrast-enhanced magnetic resonance imaging were enrolled, including 42 HER2 positive and 34 negative cases confirmed by fluorescence in situ hybridization. The lesion area was delineated semi-automatically on the subtraction MRI images at the second, fourth, and sixth phases (P-1, P-2, and P-3). A regionalization method was used to segment the lesion area into three subregions (rapid, medium, and slow) according to peak arrival time of the contrast agent. We extracted 488 texture features from the whole lesion area and three subregions independently. Wrapper, least absolute shrinkage and selection operator (LASSO), and stepwise methods were used to identify the optimal feature subsets. Univariate analysis was performed as well as support vector machine (SVM) with a leave-one-out-based cross-validation method. The area under the receiver operating characteristic curve (AUC) was calculated to evaluate the performance of the classifiers.

Results: In univariate analysis, the variance from medium subregion at P-2 was the best-performing feature for distinguishing HER2 2+ status (AUC = 0.836); for the whole lesion region, the variance at P-2 achieved the best performance (AUC = 0.798). There was no significant difference between the two methods (P = 0.271). In the machine learning with SVM, the best performance (AUC = 0.929) was achieved with LASSO from rapid subregion at P-2; for the whole region, the highest AUC value was 0.847 obtained at P-2 with LASSO. The difference was significant between the two methods (P = 0.021).

Conclusion: The texture analysis of heterogeneity subregions based on intratumoral regionalization method showed potential value for recognizing HER2 2+ status in breast cancer.

Tumor texture parameters of invasive ductal breast carcinoma in neoadjuvant chemotherapy: early identification of non-responders on breast MRI

PURPOSE

Texture analysis (TA) parameters (variance of SI, mean of gradient, variance of gradient, kurtosis of SI, and entropy) in patients with invasive ductal carcinoma (IDC) contribute to objective assessment of neoadjuvant chemotherapy (NACT) activity. The objective was to assess TA parameters in early identification of non-responders (NR) in NACT, after the 2nd cycle of NACT.

MATERIAL AND METHODS

Fifty patients (N = 50) were included in the retrospective analysis of baseline and MRI following the 2nd cycle of NACT. TA parameters were computed and correlated to the lesion size and DWI-ADC in NR (N1 = 25). Additional matched responders (R, N2 = 25) assessed for the same parameters, served as the control group.

RESULTS

Tumor size and ADC did not change significantly in NR after the 2nd cycle of NACT (2.88 ± 0.38 vs. 2.76 ± 0.36 [cm], p = 0.131; 1.01 ± 0.14 vs. 1.05 ± 0.13 [mm²/s × 10^-3], p = 0.363), but TA parameters changed significantly: variance of gradient (346.5 ± 12.6 vs. 355.6 ± 16.9, p = 0.01), kurtosis of SI (1.47 ± 0.09 vs. 1.54 ± 0.11, p = 0.02), entropy LH (60.39 ± 4.34 vs. 64.42 ± 3.05, p = 0.001) and entropy HL (61.02 ± 5.51 vs. 65.63 ± 3.63, p < 0.00001). TA parameters, particularly entropy (EN LH 64.42 ± 3.05 vs. 61.59 ± 1.76, p < 0.0001; EN HL 65.63 ± 3.63 vs. 62.89 ± 2.05, p < 0.0001), significantly differ between NR and R in early response assessment.

CONCLUSION

Entropy, kurtosis of SI and variance of gradient tend to increase in NR. TA parameters significantly differ between NR and R after the 2nd cycle of NACT. TA parameters, related to morpho-functional parameters may contribute to early NR identification.

Reliability of CT radiomic features reflecting tumour heterogeneity according to image quality and image processing parameters

The reliability of radiomics features (RFs) is crucial for quantifying tumour heterogeneity. We assessed the influence of imaging, segmentation, and processing conditions (quantization range, bin number, signal-to-noise ratio [SNR], and unintended outliers) on RF measurement. Low SNR and unintended outliers increased the standard deviation and mean values of histograms to calculate the first-order RFs. Variations in imaging processing conditions significantly altered the shape of the probability distribution (centre of distribution, extent of dispersion, and segmentation of probability clusters) in second-order RF matrices (i.e. grey-level co-occurrence and grey-level run length), thereby eventually causing fluctuations in RF estimation. Inconsistent imaging and processing conditions decreased the number of reliably measured RFs in terms of individual RF values (intraclass correlation coefficient ≥0.75) and inter-lesion RF ratios (coefficient of variation <15%). No RF could be reliably estimated under inconsistent SNR and inclusion of outlier conditions. By contrast, with high SNR and no outliers, all first-order RFs, 11 (42%) grey-level co-occurrence RFs and five (42%) grey-level run length RFs showed acceptable reliability. Our study suggests that optimization of SNR, exclusion of outliers, and application of relevant quantization range and bin number should be performed to ensure the robustness of radiomics studies assessing tumor heterogeneity.

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.

Precision diagnostics based on machine learning-derived imaging signatures

The complexity of modern multi-parametric MRI has increasingly challenged conventional interpretations of such images. Machine learning has emerged as a powerful approach to integrating diverse and complex imaging data into signatures of diagnostic and predictive value. It has also allowed us to progress from group comparisons to imaging biomarkers that offer value on an individual basis. We review several directions of research around this topic, emphasizing the use of machine learning in personalized predictions of clinical outcome, in breaking down broad umbrella diagnostic categories into more detailed and precise subtypes, and in non-invasively estimating cancer molecular characteristics. These methods and studies contribute to the field of precision medicine, by introducing more specific diagnostic and predictive biomarkers of clinical outcome, therefore pointing to better matching of treatments to patients.

Use of radiomic features and support vector machine to distinguish Parkinson's disease cases from normal controls

Background

Parkinson's disease (PD) is an irreversible neurodegenerative disease. The diagnosis of PD based on neuroimaging is usually with low-level or deep learning features, which results in difficulties in achieving precision classification or interpreting the clinical significance. Herein, we aimed to extract high-order features by using radiomics approach and achieve acceptable diagnosis accuracy in PD.

Methods

In this retrospective multicohort study, we collected ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) images and clinical scale [the Unified Parkinson's Disease Rating Scale (UPDRS) and Hoehn & Yahr scale (H&Y)] from two cohorts. One cohort from Huashan Hospital had 91 normal controls (NC) and 91 PD patients (UPDRS: 22.7±11.7, H&Y: 1.8±0.8), and the other cohort from Wuxi 904 Hospital had 26 NC and 22 PD patients (UPDRS: 20.9±11.6, H&Y: 1.7±0.9). The Huashan cohort was used as the training and test sets by 5-fold cross-validation and the Wuxi cohort was used as another separate test set. After identifying regions of interests (ROIs) based on the atlas-based method, radiomic features were extracted and selected by using autocorrelation and fisher score algorithm. A support vector machine (SVM) was trained to classify PD and NC based on selected radiomic features. In the comparative experiment, we compared our method with the traditional voxel values method. To guarantee the robustness, above processes were repeated in 500 times.

Results

Twenty-six brain ROIs were identified. Six thousand one hundred and ten radiomic features were extracted in total. Among them 30 features were remained after feature selection. The accuracies of the proposed method achieved 90.97%±4.66% and 88.08%±5.27% in Huashan and Wuxi test sets, respectively.

Conclusions

This study showed that radiomic features and SVM could be used to distinguish between PD and NC based on 18F-FDG PET images.

Exploring breast cancer response prediction to neoadjuvant systemic therapy using MRI-based radiomics: A systematic review

PURPOSE

MRI-based tumor response prediction to neoadjuvant systemic therapy (NST) in breast cancer patients is increasingly being studied using radiomics with outcomes that appear to be promising. The aim of this study is to systematically review the current literature and reflect on its quality.

METHODS

PubMed and EMBASE databases were systematically searched for studies investigating MRI-based radiomics for tumor response prediction. Abstracts were screened by two reviewers independently. The quality of the radiomics workflow of eligible studies was assessed using the Radiomics Quality Score (RQS). An overview of the methodologies used in steps of the radiomics workflow and current results are presented.

RESULTS

Sixteen studies were included with cohort sizes ranging from 35 to 414 patients. The RQS scores varied from 0 % to 41.2 %. Methodologies in the radiomics workflow varied greatly, especially region of interest segmentation, features selection, and model development with heterogeneous outcomes as a result. Seven studies applied univariate analysis and nine studies applied multivariate analysis. Most studies performed their analysis on the pretreatment dynamic contrast-enhanced T1-weighted sequence. Entropy was the best performing individual feature with AUC values ranging from 0.83 to 0.85. The best performing multivariate prediction model, based on logistic regression analysis, scored a validation AUC of 0.94.

CONCLUSION

This systematic review revealed large methodological heterogeneity for each step of the MRI-based radiomics workflow, consequently, the (overall promising) results are difficult to compare. Consensus for standardization of MRI-based radiomics workflow for tumor response prediction to NST in breast cancer patients is needed to further improve research.

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

Purpose: To evaluate the ability of texture analysis of breast dynamic contrast enhancement-magnetic resonance (DCE-MR) images in differentiating human epidermal growth factor receptor 2 (HER2) 2+ status of breast tumors.

Methods: A total of 73 cases were retrospectively selected. HER2 2+ status was confirmed by fluorescence in situ hybridization. For each case, 279 textural features were derived. A student's t-test or Mann-Whitney U test was used to select features with statistically significant differences between HER2 2+ positive and negative groups. A principal component analysis was applied to eliminate feature correlation. Three machine learning classifiers, logistic regression (LR), quadratic discriminant analysis (QDA), and a support vector machine (SVM), were trained and tested using a leave-one-out cross-validation method. The area under a receiver operating characteristic curve (AUC) was measured to assess the classifier's performance.

Results: The AUCs for the different classifiers were satisfactory, ranging from 0.808 to 0.865. The classification methods derived with LR and SVM demonstrated similarly high performances, and the accuracy levels were 81.06 and 81.18%, respectively. The AUC for the classifier derived with SVM was the highest (0.865), and a marked specificity (88.90%) was presented. For the classifier with LR, the AUC was 0.851, and the corresponding sensitivity (94.44%) was the highest.

Conclusion: The texture analysis for breast DCE-MRI proposed in this study demonstrated potential utility in HER2 2+ status discrimination.

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.

MR texture analysis: potential imaging biomarker for predicting the chemotherapeutic response of patients with colorectal liver metastases

PURPOSE

The purpose of the study was to determine whether the pre-treated MR texture features of colorectal liver metastases (CRLMs) are predictive of therapeutic response after chemotherapy.

METHODS

The study included twenty-six consecutive patients (a total of 193 liver metastasis) with unrespectable CRLMs at our institution from August 2014 to February 2016. Lesions were categorized into either responding group or non-responding group according to changes in size. Texture analysis was quantified on T2-weighted images by two radiologists with consensus on regions of interest which were manually drawn on the largest cross-sectional area of the lesions. Five histogram features (mean, variance, skewness, kurtosis, and entropy¹) and five gray level co-occurrence matrix features (GLCM; angular second moment (ASM), entropy², contrast, correlation, and inverse difference moment (IDM)) were extracted. The texture parameters were statistically analyzed to identify the differences between the two groups, and the potential predictive parameters to differentiate the responding group from the non-responding group were subsequently tested using multivariable logistic regression analysis.

RESULTS

A total of 107 responding and 86 non-responding lesions were evaluated. A higher variance, entropy¹, contrast, entropy² and a lower ASM, correlation, IDM were independently (P < 0.05) associated with a good response to chemotherapy with the areas under the ROC curves (AUCs) of 0.602-0.784. Variance (P < 0.001) and ASM (P = 0.001) remained potential predictive values to discriminate responding lesions from non-responding lesions when tested using multivariable logistic regression analysis. The highest AUC of the predictors from the association of variance and ASM was 0.814.

CONCLUSION

MR texture features on pre-treated T2 images have the potential to predict the therapeutic response of colorectal liver metastases.

Drug-Homogeneity Index in Mass-Spectrometry Imaging

Enhancing drug penetration in solid tumors is an interesting clinical issue of considerable importance. In preclinical research, mass-spectrometry imaging is a promising technique for visualizing drug distribution in tumors under different treatment conditions and its application in this field is rapidly increasing. However, in view of the huge variability among MSI data sets, drug homogeneity is usually manually assessed by an expert, and this approach is biased by interobserver variability and lacks reproducibility. We propose a new texture-based feature, the drug-homogeneity index (DHI), which provides an objective, automated measure of drug homogeneity in MSI data. A simulation study on synthetic data sets showed that previously known texture features do not give an accurate picture of intratumor drug-distribution patterns and are easily influenced by the tumor-tissue morphology. The DHI has been used to study the distribution profile of the anticancer drug paclitaxel in various xenograft models, which were either pretreated or not pretreated with antiangiogenesis compounds. The conclusion is that drug homogeneity is better in the pretreated condition, which is in agreement with previous experimental findings published by our group. This study shows that DHI could be useful in preclinical studies as a new parameter for the evaluation of protocols for better drug penetration.

Role of texture analysis in breast MRI as a cancer biomarker: A review

Breast cancer is a known heterogeneous disease. Current clinically utilized histopathologic biomarkers may undersample tumor heterogeneity, resulting in higher rates of misdiagnosis for breast cancer. MRI can provide a whole-tumor sampling of disease burden and is widely utilized in clinical care. Texture analysis can provide a localized description of breast cancer, with particular emphasis on quantifying breast lesion heterogeneity. The object of this review is to provide an overview of texture analysis applications towards breast cancer diagnosis, prognosis, and treatment response evaluation and review the role of image-based texture features as noninvasive prognostic and predictive biomarkers. Level of Evidence: 5 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:927-938.

Multivariate machine learning models for prediction of pathologic response to neoadjuvant therapy in breast cancer using MRI features: a study using an independent validation set

PURPOSE

To determine whether a multivariate machine learning-based model using computer-extracted features of pre-treatment dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict pathologic complete response (pCR) to neoadjuvant therapy (NAT) in breast cancer patients.

METHODS

Institutional review board approval was obtained for this retrospective study of 288 breast cancer patients at our institution who received NAT and had a pre-treatment breast MRI. A comprehensive set of 529 radiomic features was extracted from each patient's pre-treatment MRI. The patients were divided into equal groups to form a training set and an independent test set. Two multivariate machine learning models (logistic regression and a support vector machine) based on imaging features were trained to predict pCR in (a) all patients with NAT, (b) patients with neoadjuvant chemotherapy (NACT), and (c) triple-negative or human epidermal growth factor receptor 2-positive (TN/HER2+) patients who had NAT. The multivariate models were tested using the independent test set, and the area under the receiver operating characteristics (ROC) curve (AUC) was calculated.

RESULTS

Out of the 288 patients, 64 achieved pCR. The AUC values for predicting pCR in TN/HER+ patients who received NAT were significant (0.707, 95% CI 0.582-0.833, p < 0.002).

CONCLUSIONS

The multivariate models based on pre-treatment MRI features were able to predict pCR in TN/HER2+ patients.

Virtual monoenergetic imaging in rapid kVp-switching dual-energy CT (DECT) of the abdomen: impact on CT texture analysis

PURPOSE

To study the impact of keV levels of virtual monoenergetic images generated from rapid kVp-switching dual-energy CT (rsDECT) on CT texture analysis (CTTA).

METHODS

This study included 30 consecutive patients (59.3 ± 12 years; range 34-77 years; 17M:13F) who underwent portal venous phase abdominal CT on a rsDECT scanner. Axial 5-mm monoenergetic images at 5 energy levels (40/50/60/70/80 keV) were created and CTTA of liver was performed. CTTA comprised a filtration-histogram technique with different spatial scale filter (SSF) values (0-6). CTTA quantification at each SSF value included histogram-based statistical parameters such as mean intensity, standard deviation (SD), entropy, mean of positive pixels (MPP), skewness, and kurtosis. The values were compared using repeated measures ANOVA.

RESULTS

Among the different CTTA metrics, mean intensity (at SSF > 0), skewness, and kurtosis did not show variability whereas entropy, MPP, and SD varied with different keV levels. There was no change in skewness and kurtosis values for all 6 filters (p > 0.05). Mean intensity showed no change for filters 2-6 (p > 0.05). Mean intensity at SSF = 0 i.e., mean attenuations were 91.2 ± 2.9, 108.7 ± 3.6, 136.1 ± 4.7, 179.8 ± 6.9, and 250.5 ± 10.1 HU for 80, 70, 60, 50, and 40 keV images, respectively demonstrating significant variability (decrease) with increasing keV levels (p < 0.001). Entropy, MPP, and SD values showed a statistically significant decrease with increasing keV of monoenergetic images on all 6 filters (p < 0.001).

CONCLUSION

The energy levels of monoenergetic images have variable impact on the different CTTA parameters, with no significant change in skewness, kurtosis, and filtered mean intensity whereas significant decrease in mean attenuation, entropy, MPP, and SD values with increasing energy levels.

Texture analysis of 3D dose distributions for predictive modelling of toxicity rates in radiotherapy

BACKGROUND AND PURPOSE

To explore the use of texture analysis (TA) features of patients' 3D dose distributions to improve prediction modelling of treatment complication rates in prostate cancer radiotherapy.

MATERIAL AND METHODS

Late toxicity scores, dose distributions, and non-treatment related (NTR) predictors for late toxicity, such as age and baseline symptoms, of 351 patients of the hypofractionation arm of the HYPRO randomized trial were used in this study. Apart from DVH parameters, also TA features of rectum and bladder 3D dose distributions were used for predictive modelling of gastrointestinal (GI) and genitourinary (GU) toxicities. Logistic Normal Tissue Complication Probability (NTCP) models were derived, using only NTR parameters, NTR + DVH, NTR + TA, and NTR + DVH + TA.

RESULTS

For rectal bleeding, the area under the curve (AUC) for using only NTR parameters was 0.58, which increased to 0.68, and 0.73, when adding DVH or TA parameters respectively. For faecal incontinence, the AUC went up from 0.63 (NTR only), to 0.68 (+DVH) and 0.73 (+TA). For nocturia, adding TA features resulted in an AUC increase from 0.64 to 0.66, while no improvement was seen when including DVH parameters in the modelling. For urinary incontinence, the AUC improved from 0.68 to 0.71 (+DVH) and 0.73 (+TA). For GI, model improvements resulting from adding TA parameters to NTR instead of DVH were statistically significant (p < 0.04).

CONCLUSION

Inclusion of 3D dosimetric texture analysis features in predictive modelling of GI and GU toxicity rates in prostate cancer radiotherapy improved prediction performance, which was statistically significant for GI.

Computation of reliable textural indices from multimodal brain MRI: suggestions based on a study of patients with diffuse intrinsic pontine glioma

Few methodological studies regarding widely used textural indices robustness in MRI have been reported. In this context, this study aims to propose some rules to compute reliable textural indices from multimodal 3D brain MRI. Diagnosis and post-biopsy MR scans including T1, post-contrast T1, T2 and FLAIR images from thirty children with diffuse intrinsic pontine glioma (DIPG) were considered. The hybrid white stripe method was adapted to standardize MR intensities. Sixty textural indices were then computed for each modality in different regions of interest (ROI), including tumor and white matter (WM). Three types of intensity binning were compared [Formula: see text]: constant bin width and relative bounds; [Formula: see text] constant number of bins and relative bounds; [Formula: see text] constant number of bins and absolute bounds. The impact of the volume of the region was also tested within the WM. First, the mean Hellinger distance between patient-based intensity distributions decreased by a factor greater than 10 in WM and greater than 2.5 in gray matter after standardization. Regarding the binning strategy, the ranking of patients was highly correlated for 188/240 features when comparing [Formula: see text] with [Formula: see text], but for only 20 when comparing [Formula: see text] with [Formula: see text], and nine when comparing [Formula: see text] with [Formula: see text]. Furthermore, when using [Formula: see text] or [Formula: see text] texture indices reflected tumor heterogeneity as assessed visually by experts. Last, 41 features presented statistically significant differences between contralateral WM regions when ROI size slightly varies across patients, and none when using ROI of the same size. For regions with similar size, 224 features were significantly different between WM and tumor. Valuable information from texture indices can be biased by methodological choices. Recommendations are to standardize intensities in MR brain volumes, to use intensity binning with constant bin width, and to define regions with the same volumes to get reliable textural indices.

Prediction of Chemoresistance in Women Undergoing Neo-Adjuvant Chemotherapy for Locally Advanced Breast Cancer: Volumetric Analysis of First-Order Textural Features Extracted from Multiparametric MRI

Purpose

To assess correlations between volumetric first-order texture parameters on baseline MRI and pathological response after neoadjuvant chemotherapy (NAC) for locally advanced breast cancer (BC).

Materials and Methods

69 patients with locally advanced BC candidate to neoadjuvant chemotherapy underwent MRI within 4 weeks from the start of therapeutic regimen. T2, DWI, and DCE sequences were analyzed and maps were generated for Apparent Diffusion Coefficient (ADC), T2 signal intensity, and the following dynamic parameters: k-trans, peak enhancement, area under curve (AUC), time to maximal enhancement (TME), wash-in rate, and washout rate. Volumetric analysis of these parameters was performed, yielding a histogram analysis including first-order texture kinetics (percentiles, maximum value, minimum value, range, standard deviation, mean, median, mode, skewness, and kurtosis). Finally, correlations between these values and response to NAC (evaluated on the surgical specimen according to RECIST 1.1 criteria) were assessed.

Results

Out of 69 tumors, 33 (47.8%) achieved complete pathological response, 26 (37.7%) partial response, and 10 (14.5%) no response. Higher levels of AUCmax (p value = 0.0338), AUCrange (p value = 0.0311), and TME75 (p value = 0.0452) and lower levels of washout10 (p value = 0.0417), washout20 (p value = 0.0138), washout25 (p value = 0.0114), and washout30 (p value = 0.05) were predictive of noncomplete response.

Conclusion

Histogram-derived texture analysis of MRI images allows finding quantitative parameters predictive of nonresponse to NAC in women affected by locally advanced BC.

Promises and challenges for the implementation of computational medical imaging (radiomics) in oncology

Medical image processing and analysis (also known as Radiomics) is a rapidly growing discipline that maps digital medical images into quantitative data, with the end goal of generating imaging biomarkers as decision support tools for clinical practice. The use of imaging data from routine clinical work-up has tremendous potential in improving cancer care by heightening understanding of tumor biology and aiding in the implementation of precision medicine. As a noninvasive method of assessing the tumor and its microenvironment in their entirety, radiomics allows the evaluation and monitoring of tumor characteristics such as temporal and spatial heterogeneity. One can observe a rapid increase in the number of computational medical imaging publications-milestones that have highlighted the utility of imaging biomarkers in oncology. Nevertheless, the use of radiomics as clinical biomarkers still necessitates amelioration and standardization in order to achieve routine clinical adoption. This Review addresses the critical issues to ensure the proper development of radiomics as a biomarker and facilitate its implementation in clinical practice.

MRI texture analysis in differentiating luminal A and luminal B breast cancer molecular subtypes - a feasibility study

Background

The aim of this study was to use texture analysis (TA) of breast magnetic resonance (MR) images to assist in differentiating estrogen receptor (ER) positive breast cancer molecular subtypes.

Methods

Twenty-seven patients with histopathologically proven invasive ductal breast cancer were selected in preliminary study. Tumors were classified into molecular subtypes: luminal A (ER-positive and/or progesterone receptor (PR)-positive, human epidermal growth factor receptor type 2 (HER2) -negative, proliferation marker Ki-67 < 20 and low grade (I)) and luminal B (ER-positive and/or PR-positive, HER2-positive or HER2-negative with high Ki-67 ≥ 20 and higher grade (II or III)). Co-occurrence matrix -based texture features were extracted from each tumor on T1-weighted non fat saturated pre- and postcontrast MR images using TA software MaZda. Texture parameters and tumour volumes were correlated with tumour prognostic factors.

Results

Textural differences were observed mainly in precontrast images. The two most discriminative texture parameters to differentiate luminal A and luminal B subtypes were sum entropy and sum variance (p = 0.003). The AUCs were 0.828 for sum entropy (p = 0.004), and 0.833 for sum variance (p = 0.003), and 0.878 for the model combining texture features sum entropy, sum variance (p = 0.001). In the LOOCV, the AUC for model combining features sum entropy and sum variance was 0.876. Sum entropy and sum variance showed positive correlation with higher Ki-67 index. Luminal B types were larger in volume and moderate correlation between larger tumour volume and higher Ki-67 index was also observed (r = 0.499, p = 0.008).

Conclusions

Texture features which measure randomness, heterogeneity or smoothness and homogeneity may either directly or indirectly reflect underlying growth patterns of breast tumours. TA and volumetric analysis may provide a way to evaluate the biologic aggressiveness of breast tumours and provide aid in decisions regarding therapeutic efficacy.

Identification of ghost artifact using texture analysis in pediatric spinal cord diffusion tensor images

PURPOSE

Ghost artifacts are a major contributor to degradation of spinal cord diffusion tensor images. A multi-stage post-processing pipeline was designed, implemented and validated to automatically remove ghost artifacts arising from reduced field of view diffusion tensor imaging (DTI) of the pediatric spinal cord.

METHOD

A total of 12 pediatric subjects including 7 healthy subjects (mean age=11.34years) with no evidence of spinal cord injury or pathology and 5 patients (mean age=10.96years) with cervical spinal cord injury were studied. Ghost/true cords, labeled as region of interests (ROIs), in non-diffusion weighted b0 images were segmented automatically using mathematical morphological processing. Initially, 21 texture features were extracted from each segmented ROI including 5 first-order features based on the histogram of the image (mean, variance, skewness, kurtosis and entropy) and 16s-order feature vector elements, incorporating four statistical measures (contrast, correlation, homogeneity and energy) calculated from co-occurrence matrices in directions of 0°, 45°, 90° and 135°. Next, ten features with a high value of mutual information (MI) relative to the pre-defined target class and within the features were selected as final features which were input to a trained classifier (adaptive neuro-fuzzy interface system) to separate the true cord from the ghost cord.

RESULTS

The implemented pipeline was successfully able to separate the ghost artifacts from true cord structures. The results obtained from the classifier showed a sensitivity of 91%, specificity of 79%, and accuracy of 84% in separating the true cord from ghost artifacts.

CONCLUSION

The results show that the proposed method is promising for the automatic detection of ghost cords present in DTI images of the spinal cord. This step is crucial towards development of accurate, automatic DTI spinal cord post processing pipelines.

Features from Computerized Texture Analysis of Breast Cancers at Pretreatment MR Imaging Are Associated with Response to Neoadjuvant Chemotherapy

Purpose To evaluate whether features from texture analysis of breast cancers were associated with pathologic complete response (pCR) after neoadjuvant chemotherapy and to explore the association between texture features and tumor subtypes at pretreatment magnetic resonance (MR) imaging. Materials and Methods Institutional review board approval was obtained. This retrospective study included 85 patients with 85 breast cancers who underwent breast MR imaging before neoadjuvant chemotherapy between April 10, 2008, and March 12, 2015. Two-dimensional texture analysis was performed by using software at T2-weighted MR imaging and contrast material-enhanced T1-weighted MR imaging. Quantitative parameters were compared between patients with pCR and those with non-pCR and between patients with triple-negative breast cancer and those with non-triple-negative cancer. Multiple logistic regression analysis was used to determine independent parameters. Results Eighteen tumors (22%) were triple-negative breast cancers. pCR was achieved in 30 of the 85 tumors (35%). At univariate analysis, mean pixel intensity with spatial scaling factor (SSF) of 2 and 4 on T2-weighted images and kurtosis on contrast-enhanced T1-weighted images showed a significant difference between triple-negative breast cancer and non-triple-negative breast cancer (P = .009, .003, and .001, respectively). Kurtosis (SSF, 2) on T2-weighted images showed a significant difference between pCR and non-pCR (P = .015). At multiple logistic regression, kurtosis on T2-weighted images was independently associated with pCR in non-triple-negative breast cancer (P = .033). A multivariate model incorporating T2-weighted and contrast-enhanced T1-weighted kurtosis showed good performance for the identification of triple-negative breast cancer (area under the receiver operating characteristic curve, 0.834). Conclusion At pretreatment MR imaging, kurtosis appears to be associated with pCR to neoadjuvant chemotherapy in non-triple-negative breast cancer and may be a promising biomarker for the identification of triple-negative breast cancer. ^© RSNA, 2017.

Texture analysis as a radiomic marker for differentiating renal tumors

PURPOSE

To evaluate the utility of texture analysis for the differentiation of renal tumors, including the various renal cell carcinoma subtypes and oncocytoma.

MATERIALS AND METHODS

Following IRB approval, a retrospective analysis was performed, including all patients with pathology-proven renal tumors and an abdominal computed tomography (CT) examination. CT images of the tumors were manually segmented, and texture analysis of the segmented tumors was performed. A support vector machine (SVM) method was also applied to classify tumor types. Texture analysis results were compared to the various tumors and areas under the curve (AUC) were calculated. Similar calculations were performed with the SVM data.

RESULTS

One hundred nineteen patients were included. Excellent discriminators of tumors were identified among the histogram-based features noting features skewness and kurtosis, which demonstrated AUCs of 0.91 and 0.93 (p < 0.0001), respectively, for differentiating clear cell subtype from oncocytoma. Histogram feature median demonstrated an AUC of 0.99 (p < 0.0001) for differentiating papillary subtype from oncocytoma and an AUC of 0.92 for differentiating oncocytoma from other tumors. Machine learning further improved the results achieving very good to excellent discrimination of tumor subtypes. The ability of machine learning to distinguish clear cell subtype from other tumors and papillary subtype from other tumors was excellent with AUCs of 0.91 and 0.92, respectively.

CONCLUSION

Texture analysis is a promising non-invasive tool for distinguishing renal tumors on CT images. These results were further improved upon application of machine learning, and support the further development of texture analysis as a quantitative biomarker for distinguishing various renal tumors.

A computer-aided diagnosis (CAD) scheme for pretreatment prediction of pathological response to neoadjuvant therapy using dynamic contrast-enhanced MRI texture features

OBJECTIVE

To assess whether a computer-aided, diagnosis (CAD) system can predict pathological Complete Response (pCR) to neoadjuvant chemotherapy (NAC) prior to treatment using texture features.

METHODS

Response to treatment of 44 patients was defined according to the histopatology of resected tumour and extracted axillary nodes in two ways: (a) pCR+ (Smith's Grade = 5) vs pCR- (Smith's Grade < 5); (b) pCRN+ (pCR+ and absence of residual lymph node metastases) vs pCRN - . A CAD system was developed to: (i) segment the breasts; (ii) register the DCE-MRI sequence; (iii) detect the lesion and (iv) extract 27 3D texture features. The role of individual texture features, multiparametric models and Bayesian classifiers in predicting patients' response to NAC were evaluated.

RESULTS

A cross-validated Bayesian classifier fed with 6 features was able to predict pCR with a specificity of 72% and a sensitivity of 67%. Conversely, 2 features were used by the Bayesian classifier to predict pCRN, obtaining a sensitivity of 69% and a specificity of 61%.

CONCLUSION

A CAD scheme, that extracts texture features from an automatically segmented 3D mask of the tumour, could predict pathological response to NAC. Additional research should be performed to validate these promising results on a larger cohort of patients and using different classification strategies. Advances in knowledge: This is the first study assessing the role of an automatic CAD system in predicting the pathological response to NAC before treatment. Fully automatic methods represent the backbone of standardized analysis and may help in timely managing patients candidate to NAC.

Interim heterogeneity changes measured using entropy texture features on T2-weighted MRI at 3.0 T are associated with pathological response to neoadjuvant chemotherapy in primary breast cancer

Objectives

To investigate whether interim changes in hetereogeneity (measured using entropy features) on MRI were associated with pathological residual cancer burden (RCB) at final surgery in patients receiving neoadjuvant chemotherapy (NAC) for primary breast cancer.

Methods

This was a retrospective study of 88 consenting women (age: 30–79 years). Scanning was performed on a 3.0 T MRI scanner prior to NAC (baseline) and after 2–3 cycles of treatment (interim). Entropy was derived from the grey-level co-occurrence matrix, on slice-matched baseline/interim T2-weighted images. Response, assessed using RCB score on surgically resected specimens, was compared statistically with entropy/heterogeneity changes and ROC analysis performed. Association of pCR within each tumour immunophenotype was evaluated.

Results

Mean entropy percent differences between examinations, by response category, were: pCR: 32.8%, RCB-I: 10.5%, RCB-II: 9.7% and RCB-III: 3.0%. Association of ultimate pCR with coarse entropy changes between baseline/interim MRI across all lesions yielded 85.2% accuracy (area under ROC curve: 0.845). Excellent sensitivity/specificity was obtained for pCR prediction within each immunophenotype: ER+: 100%/100%; HER2+: 83.3%/95.7%, TNBC: 87.5%/80.0%.

Conclusions

Lesion T2 heterogeneity changes are associated with response to NAC using RCB scores, particularly for pCR, and can be useful across all immunophenotypes with good diagnostic accuracy.

Key Points

• Texture analysis provides a means of measuring lesion heterogeneity on MRI images.

• Heterogeneity changes between baseline/interim MRI can be linked with ultimate pathological response.

• Heterogeneity changes give good diagnostic accuracy of pCR response across all immunophenotypes.

• Percentage reduction in heterogeneity is associated with pCR with good accuracy and NPV.

Computer-aided heterogeneity analysis in breast MR imaging assessment of ductal carcinoma in situ: Correlating histologic grade and receptor status

PURPOSE

To identify breast MR imaging biomarkers to predict histologic grade and receptor status of ductal carcinoma in situ (DCIS).

MATERIALS AND METHODS

Informed consent was waived in this Health Insurance Portability and Accountability Act-compliant Institutional Review Board-approved study. Case inclusion was conducted from 7332 consecutive breast MR studies from January 1, 2009, to December 31, 2012. Excluding studies with benign diagnoses, studies without visible abnormal enhancement, and pathology containing invasive disease yielded 55 MR-imaged pathology-proven DCIS seen on 54 studies. Twenty-eight studies (52%) were performed at 1.5 Tesla (T); 26 (48%) at 3T. Regions-of-interest representing DCIS were segmented for precontrast, first and fourth postcontrast, and subtracted first and fourth postcontrast images on the open-source three-dimensional (3D) Slicer software. Fifty-seven metrics of each DCIS were obtained, including distribution statistics, shape, morphology, Renyi dimensions, geometrical measure, and texture, using the 3D Slicer HeterogeneityCAD module. Statistical correlation of heterogeneity metrics with DCIS grade and receptor status was performed using univariate Mann-Whitney test.

RESULTS

Twenty-four of the 55 DCIS (44%) were high nuclear grade (HNG); 44 (80%) were estrogen receptor (ER) positive. Human epidermal growth factor receptor-2 (HER2) was amplified in 10/55 (18%), nonamplified in 34/55 (62%), unknown/equivocal in 8/55 (15%). Surface area-to-volume ratio showed significant difference (P < 0.05) between HNG and non-HNG DCIS. No metric differentiated ER status (0.113 < p ≤ 1.000). Seventeen metrics showed significant differences between HER2-positive and HER2-negative DCIS (0.016 < P < 0.050).

CONCLUSION

Quantitative heterogeneity analysis of DCIS suggests the presence of MR imaging biomarkers in classifying DCIS grade and HER2 status. Validation with larger samples and prospective studies is needed to translate these results into clinical applications.

LEVEL OF EVIDENCE

3 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1748-1759.