Computer-aided quantification of contrast agent spatial distribution within atherosclerotic plaque in contrast-enhanced ultrasound image sequences

Artificially intelligent differential diagnosis of enlarged lymph nodes with random vector functional link network plus

Differential diagnosis of enlarged lymph nodes (ELNs) is essential for the treatment of related patients. Though multi-modal ultrasound including B-mode, Doppler ultrasound, elastography and contrast-enhanced ultrasound (CEUS) can enhance diagnostic performance for ELNs, the scenario of having only single or dual modal data is often encountered. In this study, an artificially intelligent diagnosis model based on the learning using privileged information was proposed to aid in differential diagnosis of ELNs in the case of single or dual modal images. In our model, B-mode, or combined with another modality, was used as the standard information (SI) and other modalities were used as the privileged information (PI). The model was constructed through the combination of the SI and PI in the training stage. By learning from the training samples, a random vector functional link network with privileged information (RVFL+) was obtained, which was used to classify the testing samples of solely the SI. Results showed that the accuracy, precision and Youden's index of the RVFL+ model, using B-mode with elastography as the SI and CEUS as the PI, reached 78.4%, 92.4% and 54.9%, increased by 14.0%, 8.4% and 24.5% compared with the model using B-mode as the SI without the PI. The method based on the LUPI can improve the diagnostic performance for ELNs.

Intelligent diagnosis of left ventricular hypertrophy using transthoracic echocardiography videos

PURPOSE

Left ventricular hypertrophy (LVH) is an independent risk factor for cardiovascular events and mortality. Pathological LVH can be caused by various diseases. In this study, we explored the possibility of using time and frequency domain analysis of myocardial radiomics features for patients with LVH in differentiating hypertrophic cardiomyopathy (HCM), hypertensive heart disease (HHD) and uremic cardiomyopathy (UCM) based on transthoracic echocardiography (TTE). This was the first study to explore TTE myocardial time and frequency domain analyses for multiple LVH etiology differentiation.

MATERIALS AND METHODS

We proposed an artificially intelligent diagnosis system based on radiomics techniques for differentiating HCM, HHD and UCM on TTE videos of the apical four-chamber view, which mainly included interventricular septum (IVS) segmentation, feature extraction and classification. We used two independent cohorts, one with 150 patients, including 50 HHD, 50 HCM and 50 UCM, for segmentation training and testing, and another with 149 patients (namely the main cohort), including 50 HHD, 46 HCM and 53 UCM, for classification training and testing after segmentation and feature extraction. Firstly, the U-Net, Residual U-Net (ResUNet) and nnU-Net were trained and tested to segment the IVS on TTE still images in the first cohort. Then the trained model with the best segmentation performance was further used for IVS prediction of ordered TTE images in video sequences in the main cohort. The post-processing was used to eliminate the noisy debris by selecting the maximum connected region and smoothing the edges of the predicted IVS region. Secondly, static radiomics features were extracted from the IVS of ordered TTE images in each video sequence, and subsequently the time and frequency domain features were further extracted from each time series of a static radiomics feature in the video sequence. Finally, the point-wise gated Boltzmann machine (PGBM) was used to learn and fuse the time and frequency domain features, and the support vector machine was used to classify the learned features for LVH diagnosis. The classification was performed with five-fold cross validation.

RESULTS

The ResUNet showed the best segmentation performance, with Dice coefficient, sensitivity, specificity and accuracy of 0.817, 76.3%, 99.6% and 98.6%, respectively. With post-processing, the Dice coefficient, sensitivity, specificity and accuracy of the ResUNet were further improved to 0.839, 77.0%, 99.8%, and 98.8%, respectively. The classification areas under the receiver operating characteristic curves (AUCs) were 0.838 ± 0.049 for HHD vs. HCM, 0.868 ± 0.042 for HCM vs. UCM and 0.701 ± 0.140 for HHD vs. UCM.

CONCLUSION

In this work, we proposed an intelligent identification system for LVH etiology classification based on routine TTE video images with good diagnostic performance. This deep learning method is feasible in automatic TTE images interpretation and expected to assist clinicians in detecting the primary cause of LVH.

Ultrasound Radiomics Effective for Preoperative Identification of True and Pseudo Gallbladder Polyps Based on Spatial and Morphological Features

Purpose: To explore the value of ultrasound radiomics in the preoperative identification of true and pseudo gallbladder polyps and to evaluate the associated diagnostic accuracy.

Methods: Totally, 99 pathologically proven gallbladder polyps in 96 patients were enrolled, including 58 cholesterol polyps (55 patients) and 41 gallbladder tubular adenomas (41 patients). Features on preoperative ultrasound images, including spatial and morphological features, were acquired for each lesion. Following this, two-stage feature selection was adopted using Fisher's inter-intraclass variance ratios and Z-scores for the selection of intrinsic features important for differential diagnosis achievement with support vector machine use.

Results: Eighty radiomic features were extracted from each polyp. Eight intrinsic features were identified after two-stage selection. The contrast 14 (Cont14) and entropy 6 (Entr6) values in the cholesterol polyp group were significantly higher than those in the gallbladder adenoma group (4.063 ± 1.682 vs. 2.715 ± 1.867, p < 0.001 for Cont14; 4.712 ± 0.427 vs. 4.380 ± 0.720, p = 0.003 for Entr6); however, the homogeneity 13 (Homo13) and energy 8 (Ener8) values in the cholesterol polyp group were significantly lower (0.500 ± 0.069 vs. 0.572 ± 0.057, p < 0.001 for Homo13; 0.050 ± 0.023 vs. 0.068 ± 0.038, p = 0.002 for Ener8). These results indicate that the pixel distribution of cholesterol polyps was more uneven than that of gallbladder tubular adenomas. The dispersion degree was also significantly lower in the cholesterol polyp group than the gallbladder adenoma group (0.579 ± 0.054 vs. 0.608 ± 0.041, p = 0.005), indicating a lower dispersion of high-intensity areas in the cholesterol polyps. The long axis length of the fitting ellipse (Maj.Len), diameter of a circle equal to the lesion area (Eq.Dia) and perimeter (Per) values in the cholesterol polyp group were significantly lower than those in the gallbladder adenoma group (0.971 ± 0.485 vs. 1.738 ± 0.912, p < 0.001 for Maj.Len; 0.818 ± 0.393 vs. 1.438 ± 0.650, p < 0.001 for Eq.Dia; 2.637 ± 1.281 vs. 5.033 ± 2.353, p < 0.001 for Per), demonstrating that the cholesterol polyps were smaller and more regular in terms of morphology. The classification accuracy, sensitivity, specificity, and area under the curve values were 0.875, 0.885, 0.857, and 0.898, respectively.

Conclusions: Ultrasound radiomic analysis based on the spatial and morphological features extracted from ultrasound images effectively contributed to the preoperative diagnosis of true and pseudo gallbladder polyps and may be valuable in their clinical management.

Multimodal feature learning and fusion on B-mode ultrasonography and sonoelastography using point-wise gated deep networks for prostate cancer diagnosis

B-mode ultrasonography and sonoelastography are used in the clinical diagnosis of prostate cancer (PCa). A combination of the two ultrasound (US) modalities using computer aid may be helpful for improving the diagnostic performance. A technique for computer-aided diagnosis (CAD) of PCa is presented based on multimodal US. Firstly, quantitative features are extracted from both B-mode US images and sonoelastograms, including intensity statistics, regional percentile features, gray-level co-occurrence matrix (GLCM) texture features and binary texture features. Secondly, a deep network named PGBM-RBM2 is proposed to learn and fuse multimodal features, which is composed of the point-wise gated Boltzmann machine (PGBM) and two layers of the restricted Boltzmann machines (RBMs). Finally, the support vector machine (SVM) is used for prostatic disease classification. Experimental evaluation was conducted on 313 multimodal US images of the prostate from 103 patients with prostatic diseases (47 malignant and 56 benign). Under five-fold cross-validation, the classification sensitivity, specificity, accuracy, Youden's index and area under the receiver operating characteristic (ROC) curve with the PGBM-RBM2 were 87.0%, 88.8%, 87.9%, 75.8% and 0.851, respectively. The results demonstrate that multimodal feature learning and fusion using the PGBM-RBM2 can assist in the diagnosis of PCa. This deep network is expected to be useful in the clinical diagnosis of PCa.

Contrast-Enhanced Ultrasound Quantification: From Kinetic Modeling to Machine Learning

Ultrasound contrast agents (UCAs) have opened up immense diagnostic possibilities by combined use of indicator dilution principles and dynamic contrast-enhanced ultrasound (DCE-US) imaging. UCAs are microbubbles encapsulated in a biocompatible shell. With a rheology comparable to that of red blood cells, UCAs provide an intravascular indicator for functional imaging of the (micro)vasculature by quantitative DCE-US. Several models of the UCA intravascular kinetics have been proposed to provide functional quantitative maps, aiding diagnosis of different pathological conditions. This article is a comprehensive review of the available methods for quantitative DCE-US imaging based on temporal, spatial and spatiotemporal analysis of the UCA kinetics. The recent introduction of novel UCAs that are targeted to specific vascular receptors has advanced DCE-US to a molecular imaging modality. In parallel, new kinetic models of increased complexity have been developed. The extraction of multiple quantitative maps, reflecting complementary variables of the underlying physiological processes, requires an integrative approach to their interpretation. A probabilistic framework based on emerging machine-learning methods represents nowadays the ultimate approach, improving the diagnostic accuracy of DCE-US imaging by optimal combination of the extracted complementary information. The current value and future perspective of all these advances are critically discussed.

Respiratory motion correction for liver contrast-enhanced ultrasound by automatic selection of a reference image

PURPOSE

Respiratory motion correction is necessary for the quantitative analysis of liver contrast-enhanced ultrasound (CEUS) image sequences. Most respiratory motion correction methods are based on the dual mode of CEUS image sequences, including contrast and grayscale image sequences. Due to free-breathing motion, the acquired two-dimensional (2D) ultrasound cine might show the in-plane and out-of-plane motion of tumors. The registration of an entire 2D ultrasound contrast image sequence based on out-of-plane images is ineffective. For the respiratory motion correction of CEUS sequences, the reference image is usually considered the standard for the deletion of any out-of-plane images. Most methods used for the selection of the reference image are subjective in nature. Here, a quantitative selection method for an optimal reference image from CEUS image sequences in the B mode and contrast mode was explored.

METHODS

The original high-dimensional ultrasound grayscale image data were mapped into a two-dimensional space using Laplacian Eigenmaps (LE), and K-means clustering was adopted. The center image of the larger cluster with a near-peak contrast intensity was considered the optimal ultrasound reference image. In the ultrasound grayscale image sequence, the images with the maximum correlations to the reference image in the same time interval were selected as the corrected image sequence. The effectiveness of this proposed method was then validated on 18 CEUS cases of VX2 tumors in rabbit livers.

RESULTS

Correction smoothed the time-intensity curves (TICs) extracted from the region of interest of the CEUS image sequences. Before correction, the average of the total mean structural similarity (TMSSIM) and the average of the mean correlation coefficient (MCC) from the image sequences were 0.45 ± 0.11 and 0.67 ± 0.16, respectively, and after correction, the average TMSSIM and MCC increased (P < 0.001) by 31% to 0.59 ± 0.11 and by 21% to 0.81 ± 0.11, respectively. The average deviation value (DV) index of the TICs from the image sequences prior to correction was 92.16 ± 18.12, and correction reduced the average to 31.71 ± 7.31. The average TMSSIM and MCC values after correction using the mean frame of the reference image (MBMFRI) were clearly lower than those after correction using the proposed method (P < 0.001). Moreover, the average DV after correction using the MBMFRI was obviously higher than that after correction using the proposed method (P < 0.001).

CONCLUSIONS

The breathing frequency of rabbits is notably faster than that of human beings, but the proposed correction method could reduce the effect of the respiratory motion in the CEUS image sequences. The reference image was selected quantitatively, which could improve the accuracy of the quantitative analysis of rabbit liver CEUS sequences using the reference image method based on the current standard of manual selection and the MBMFRI. This easy-to-operate method can potentially be used in both animal studies and clinical applications.

Dual-mode artificially-intelligent diagnosis of breast tumours in shear-wave elastography and B-mode ultrasound using deep polynomial networks

The main goal of this study is to build an artificial intelligence (AI) architecture for automated extraction of dual-modal image features from both shear-wave elastography (SWE) and B-mode ultrasound, and to evaluate the AI architecture for classification between benign and malignant breast tumors. In this AI architecture, ultrasound images were segmented by the reaction diffusion level set model combined with the Gabor-based anisotropic diffusion algorithm. Then morphological features and texture features were extracted from SWE and B-mode ultrasound images at the contourlet domain. Finally, we employed a framework for feature learning and classification with the deep polynomial network (DPN) on dual-modal features to distinguish between malignant and benign breast tumors. With the leave-one-out cross validation, the DPN method on dual-modal features achieved a sensitivity of 97.8%, a specificity of 94.1%, an accuracy of 95.6%, a Youden's index of 91.9% and an area under the receiver operating characteristic curve of 0.961, which was superior to the classic single-modal methods, and the dual-modal methods using the principal component analysis and multiple kernel learning. These results have demonstrated that the dual-modal AI-based technique with DPN has the potential for breast tumor classification in future clinical practice.

Patients with Achilles Tendon Rupture Have a Degenerated Contralateral Achilles Tendon: An Elastography Study

Purpose

To evaluate differences of Achilles tendon (AT) hardness and morphology between asymptomatic tendons in patients with acute AT ruptures on the contralateral side and asymptomatic tendons in healthy people by using computer-assisted quantification on axial-strain sonoelastography (ASE).

Methods

The study consisted of 33 asymptomatic tendons in 33 patients (study group) and 34 tendons in 19 healthy volunteers (control group). All the tendons were examined by both ASE and conventional ultrasound. Computer-assisted quantification on ASE was applied to extract hardness variables, including the mean (Hmean), 20^th percentile (H20), median (H50) and skewness (Hsk) of the hardness within tendon, and the ratio of the mean hardness within tendon to that outside tendon (Hratio) and three morphological variables: the thickness (THK), cross-sectional area, and eccentricity (ECC) of tendons.

Results

The Hmean, Hsk, H20, H50, and Hratio in the proximal third of the tendon body in study group were significantly smaller than those in control group (Hmean: 0.43±0.09 vs 0.50±0.07, p=0.001; Hsk: -0.53±0.51 vs -1.09±0.51, p<0.001; H20: 0.31±0.10 vs 0.40±0.10, p=0.001; H50: 0.45±0.10 vs 0.53±0.08, p<0.001; Hratio: 1.01±0.25 vs 1.20±0.23, p=0.003). The THK and cross-sectional area of tendons in the study group were larger than those in the control group (p<0.05).

Conclusions

As a quantitative objective method, the computer-assisted ASE reveals that the asymptomatic ATs contralateral to acute rupture are softer than those of healthy control group at the proximal third and the asymptomatic tendons in people with rupture history are thicker, larger, and rounder than those of normal volunteers especially at the middle and distal thirds of AT body.

Automatic Methods for Carotid Contrast-Enhanced Ultrasound Imaging Quantification of Adventitial Vasa Vasorum

Adventitial vasa vasorum are physiologic microvessels that nourish artery walls. In the presence of cardiovascular risk factors, these microvessels proliferate abnormally. Studies have reported that they are the first stage of atheromatous disease. Contrast-enhanced ultrasound (CEUS) of the carotid allows direct, quantitative and non-invasive visualization of the adventitial vasa vasorum. Hence, the development of computer-assisted methods that speed image analysis and eliminate user subjectivity is important. We developed methods for automatic analyses and quantification of vasa vasorum neovascularization in CEUS and tested these methods in a cohort of 186 individuals, 63 of whom were healthy volunteers. We implemented alternative automatic strategies for using the images to stratify patients according to their risk group and compare the strategies with respect to diagnostic performance. An automatic single-parameter strategy performs less effectively than the corresponding Arcidiacono method based on manual interpretation of the images (68 < area under the receiver operating characteristic curve [AUROC] for the manual Arcidiacono method < 82; 60 < AUROC for the automatic single-parameter strategy < 63). However, by use of additional image parameters, an automatic multiparameter strategy has significantly improved performance with respect to the manual Arcidiacono method (78 < AUROC < 90). The automatic multiparameter strategy is a valuable alternative to the manual Arcidiacono method, improving both diagnostic speed and diagnostic accuracy.

Machine Learning in Ultrasound Computer-Aided Diagnostic Systems: A Survey

The ultrasound imaging is one of the most common schemes to detect diseases in the clinical practice. There are many advantages of ultrasound imaging such as safety, convenience, and low cost. However, reading ultrasound imaging is not easy. To support the diagnosis of clinicians and reduce the load of doctors, many ultrasound computer-aided diagnosis (CAD) systems are proposed. In recent years, the success of deep learning in the image classification and segmentation led to more and more scholars realizing the potential of performance improvement brought by utilizing the deep learning in the ultrasound CAD system. This paper summarized the research which focuses on the ultrasound CAD system utilizing machine learning technology in recent years. This study divided the ultrasound CAD system into two categories. One is the traditional ultrasound CAD system which employed the manmade feature and the other is the deep learning ultrasound CAD system. The major feature and the classifier employed by the traditional ultrasound CAD system are introduced. As for the deep learning ultrasound CAD, newest applications are summarized. This paper will be useful for researchers who focus on the ultrasound CAD system.

Dual-modal computer-assisted evaluation of axillary lymph node metastasis in breast cancer patients on both real-time elastography and B-mode ultrasound

PURPOSE

To propose a computer-assisted method for quantifying the hardness of an axillary lymph node on real-time elastography (RTE) and its morphology on B-mode ultrasound; and to combine the dual-modal features for differentiation of metastatic and benign axillary lymph nodes in breast cancer patients.

MATERIALS AND METHODS

A total of 161 axillary lymph nodes (benign, n=69; metastatic, n=92) from 158 patients with breast cancer were examined with both B-mode ultrasound and RTE. With computer assistance, five morphological features describing the hilum, size, shape, and echogenic uniformity of a lymph node were extracted from B-mode, and three hardness features were extracted from RTE. Single-modal and dual-modal features were used to classify benign and metastatic nodes with two computerized classification approaches, i.e., a scoring approach and a support vector machine (SVM) approach. The computerized approaches were also compared with a visual evaluation approach.

RESULTS

All features exhibited significant differences between benign and metastatic nodes (p<0.001), with the highest area under the receiver operating characteristic curve (AUC) of 0.803 and the highest accuracy (ACC) of 75.2% for a single feature. The SVM on dual-modal features achieved the largest AUC (0.895) and ACC (85.7%) among all methods, exceeding the scoring (AUC=0.881; ACC=83.6%) and the visual evaluation methods (AUC=0.830; ACC=84.5%). With the leave-one-out cross validation, the SVM on dual-modal features still obtained an ACC as high as 84.5%.

CONCLUSION

Dual-modal features can be extracted from RTE and B-mode ultrasound with computer assistance, which are valuable for discrimination between benign and metastatic lymph nodes. The SVM on dual-modal features outperforms the scoring and visual evaluation methods, as well as all methods using single-modal features. The computer-assisted dual-modal evaluation of lymph nodes could be potentially used in daily clinical practice for assessing axillary metastasis in breast cancer patients.

Evaluating pathologic response of breast cancer to neoadjuvant chemotherapy with computer-extracted features from contrast-enhanced ultrasound videos

PURPOSE

To extract quantitative perfusion and texture features with computer assistance from contrast-enhanced ultrasound (CEUS) videos of breast cancer before and after neoadjuvant chemotherapy (NAC), and to evaluate pathologic response to NAC with these features.

METHODS

Forty-two CEUS videos with 140,484 images were acquired from 21 breast cancer patients pre- and post-NAC. Time-intensity curve (TIC) features were calculated including the difference between area under TIC within a tumor and that within a computer-detected reference region (AUT_T-R). Four texture features were extracted including Homogeneity and Contrast. All patients were identified as pathologic responders by Miller and Payne criteria. The features between pre- and post-treatment in these responders were statistically compared, and the discrimination between pre- and post-treatment cancers was assessed with a receiver operating characteristic (ROC) curve.

RESULTS

Compared with the pre-treatment cancers, the post-treatment cancers had significantly lower Homogeneity (p<0.001) and AUT_T-R (p=0.014), as well as higher Contrast (p<0.001), indicating the intratumoral contrast enhancement decreased and became more heterogeneous after NAC in responders. The combination of Homogeneity and AUT_T-R achieved an accuracy of 90.5% and area under ROC curve of 0.946 for discrimination between pre- and post-chemotherapy cancers without cross validation. The accuracy still reached as high as 85.7% under leave-one-out cross validation.

CONCLUSIONS

The computer-extracted CEUS features show reduced and more heterogeneous neovascularization of cancer after NAC. The features achieve high accuracy for discriminating between pre- and post-chemotherapy cancers in responders and thus are potentially valuable for tumor response evaluation in clinical practice.

Sonoelastography shows that Achilles tendons with insertional tendinopathy are harder than asymptomatic tendons

PURPOSE

To seek differences of Achilles tendon hardness between insertional tendinopathy (IT) and asymptomatic controls by using computer-assisted quantification on axial-strain sonoelastography (ASE).

METHODS

The study consisted of 37 non-athletic patients presenting with Achilles tendon pain in one or two tendons. Both tendons were examined clinically. Among the 74 tendons, 16 were diagnosed and categorized into an IT group and 29 into an asymptomatic group. The remaining 29 tendons were excluded due to non-insertional tendinopathy, ruptures, previous surgery or mixed disorders. The tendons in the IT and asymptomatic groups were examined with both ASE and conventional ultrasound. Computer-assisted quantification on ASE was conducted to extract parameters of tendon hardness, including the 20th percentile (H20), median (H50) and skewness (Hsk) of the hardness within tendon, as well as the ratio of the mean hardness within tendon to that outside tendon (Hratio).

RESULTS

The H20 (p = 0.003), H50 (p = 0.004) and Hratio (p = 0.002) were larger and Hsk (p = 0.001) was smaller at distal thirds of IT tendons than those of asymptomatic tendons. For differentiation between two groups, the Hsk achieved the best value (0.815) of area under the receiver operating characteristic curve, with a sensitivity of 81.3 %, a specificity of 86.2 % and an accuracy of 84.4 %.

CONCLUSIONS

Computer-assisted quantification on ASE shows that IT tendons are harder than asymptomatic tendons. It might act as a potentially useful technique for identification and risk stratification of IT patients and thus be valuable in day-by-day clinical practice for monitoring IT progression and for evaluating therapeutic effects.

LEVEL OF EVIDENCE

III.

Sonoelastomics for Breast Tumor Classification: A Radiomics Approach with Clustering-Based Feature Selection on Sonoelastography

A radiomics approach to sonoelastography, called "sonoelastomics," is proposed for classification of benign and malignant breast tumors. From sonoelastograms of breast tumors, a high-throughput 364-dimensional feature set was calculated consisting of shape features, intensity statistics, gray-level co-occurrence matrix texture features and contourlet texture features, which quantified the shape, hardness and hardness heterogeneity of a tumor. The high-throughput features were then selected for feature reduction using hierarchical clustering and three-feature selection metrics. For a data set containing 42 malignant and 75 benign tumors from 117 patients, seven selected sonoelastomic features achieved an area under the receiver operating characteristic curve of 0.917, an accuracy of 88.0%, a sensitivity of 85.7% and a specificity of 89.3% in a validation set via the leave-one-out cross-validation, revealing superiority over the principal component analysis, deep polynomial networks and manually selected features. The sonoelastomic features are valuable in breast tumor differentiation.

Deep learning based classification of breast tumors with shear-wave elastography

This study aims to build a deep learning (DL) architecture for automated extraction of learned-from-data image features from the shear-wave elastography (SWE), and to evaluate the DL architecture in differentiation between benign and malignant breast tumors. We construct a two-layer DL architecture for SWE feature extraction, comprised of the point-wise gated Boltzmann machine (PGBM) and the restricted Boltzmann machine (RBM). The PGBM contains task-relevant and task-irrelevant hidden units, and the task-relevant units are connected to the RBM. Experimental evaluation was performed with five-fold cross validation on a set of 227 SWE images, 135 of benign tumors and 92 of malignant tumors, from 121 patients. The features learned with our DL architecture were compared with the statistical features quantifying image intensity and texture. Results showed that the DL features achieved better classification performance with an accuracy of 93.4%, a sensitivity of 88.6%, a specificity of 97.1%, and an area under the receiver operating characteristic curve of 0.947. The DL-based method integrates feature learning with feature selection on SWE. It may be potentially used in clinical computer-aided diagnosis of breast cancer.

Quantification of carotid plaque elasticity and intraplaque neovascularization using contrast-enhanced ultrasound and image registration-based elastography

It is valuable for evaluation of carotid plaque vulnerability to investigate the relation between intraplaque neovascularization (IPN) and plaque elasticity. The contrast-enhanced ultrasound (CEUS) has been used in IPN measurement, but it cannot assess plaque elasticity. The aim of this study was to develop an ultrasound elastography technique based on registration of CEUS sequential images and to use this technique for direct comparison between IPN and plaque elasticity. We employed a nonrigid image registration method using the free-form deformation model to register a pair of clinical CEUS images at systole and diastole. The 2D displacement field of the plaque was estimated and then utilized to calculate the axial and lateral strain distributions within the plaque, from which quantitative strain parameters were obtained. The IPN was measured semiquantitatively with visual assessment and quantitatively with the time-intensity curve analysis and the analysis of contrast agent spatial distributions. Histopathology with CD34 staining for quantification of microvessel density (MVD) was performed on plaques excised by carotid endarterectomy. Simulation experiments showed that the mean absolute error and the root mean squared error of the displacement estimation were 0.325±0.180 pixel (7.2%±3.8%) and 0.556±0.284 pixel (12.3%±6.1%), respectively, demonstrating high accuracy of the elastography technique. Thirty-eight plaques in 29 patients met the inclusion criteria for the elastography and image analysis, where ten plaques underwent endarterectomy. The 95th percentile (A95) and standard deviation (Asd) of the axial strains exhibited significant differences between the low and high grades of IPN visually assessed (p<0.01). A95 (R=0.579; p<0.001) and Asd (R=0.609; p<0.001) were correlated with the enhanced intensity of plaque, and also correlated with the MVD (R=0.793 and 0.817, respectively; p<0.01), suggesting that plaque became softer and more elastically heterogeneous as IPN increased. These findings provide direct and quantitative evidence for the associations between plaque strains and IPN and might be helpful for evaluation of carotid plaque vulnerability and for plaque risk stratification.

Spatio-temporal Quantification of Carotid Plaque Neovascularization on Contrast Enhanced Ultrasound: Correlation with Visual Grading and Histopathology

OBJECTIVE/BACKGROUND

To evaluate whether carotid intraplaque neovascularization (IPN) can be accurately assessed by two types of quantitative analysis on contrast enhanced ultrasound (CEUS), the time intensity curve analysis and the analysis of contrast agent spatial distributions, and whether the quantitative analysis correlates with semiquantitative visual interpretation and histopathology.

METHODS

Forty-four plaques in 34 patients were included for CEUS examination. A three point score system (absent, moderate, and extensive) was used for semiquantitative grading of IPN. Eight spatial quantitative parameters were derived, including the IPN area ratio in plaque (AR) and the AR in plaque core (AR13). Two temporal quantitative parameters were obtained, namely the enhanced intensity in plaque (EI) and the enhanced intensity ratio (EIR). Histopathology with CD34 staining for quantification of microvessel density (MVD) was performed on 12 plaques excised by carotid endarterectomy.

RESULTS

Both spatial and temporal parameters were correlated with MVD on histology (AR: r = .854; AR13: r = .858; EI: r = .767; EIR: r = .750 [p < .01]), as well as with semiquantitative grading (p < .01). Five mutually independent factors were condensed from 10 interrelated parameters by using factor analysis, and they significantly predicted MVD with an radj value as high as .932 (p = .01).

CONCLUSION

Both spatial and temporal analysis on CEUS can accurately assess IPN. Combining them provides better IPN assessment and may be useful for plaque vulnerability evaluation and risk stratification.

Quantification of elastic heterogeneity using contourlet-based texture analysis in shear-wave elastography for breast tumor classification

Ultrasound shear-wave elastography (SWE) has become a valuable tool for diagnosis of breast tumors. The purpose of this study was to quantify the elastic heterogeneity of breast tumors in SWE by using contourlet-based texture features and evaluating their diagnostic performance for classification of benign and malignant breast tumors, with pathologic results as the gold standard. A total of 161 breast tumors in 125 women who underwent B-mode and SWE ultrasonography before biopsy were included. Five quantitative texture features in SWE images were extracted from the directional subbands after the contourlet transform, including the mean (Tmean), maximum (Tmax), median (Tmed), third quartile (Tqt), and standard deviation (Tsd) of the subbands. Diagnostic performance of the texture features and the classic features was compared using the area under the receiver operating characteristic curve (AUC) and the leave-one-out cross validation with Fisher classifier. The feature Tmean achieved the highest AUC (0.968) among all features and it yielded a sensitivity of 89.1%, a specificity of 94.3% and an accuracy of 92.5% for differentiation between benign and malignant tumors via the leave-one-out cross validation. Compared with the best classic feature, i.e., the maximum elasticity, Tmean improved the AUC, sensitivity, specificity and accuracy by 3.5%, 12.7%, 2.8% and 6.2%, respectively. The Tmed, Tqt and Tsd were also superior to the classic features in terms of the AUC and accuracy. The results demonstrated that the contourlet-based texture features captured the tumor's elastic heterogeneity and improved diagnostic performance contrasted with the classic features.

Fully automated carotid plaque segmentation in combined contrast-enhanced and B-mode ultrasound

Carotid plaque segmentation in B-mode ultrasound (BMUS) and contrast-enhanced ultrasound (CEUS) is crucial to the assessment of plaque morphology and composition, which are linked to plaque vulnerability. Segmentation in BMUS is challenging because of noise, artifacts and echo-lucent plaques. CEUS allows better delineation of the lumen but contains artifacts and lacks tissue information. We describe a method that exploits the combined information from simultaneously acquired BMUS and CEUS images. Our method consists of non-rigid motion estimation, vessel detection, lumen-intima segmentation and media-adventitia segmentation. The evaluation was performed in training (n = 20 carotids) and test (n = 28) data sets by comparison with manually obtained ground truth. The average root-mean-square errors in the training and test data sets were comparable for media-adventitia (411 ± 224 and 393 ± 239 μm) and for lumen-intima (362 ± 192 and 388 ± 200 μm), and were comparable to inter-observer variability. To the best of our knowledge, this is the first method to perform fully automatic carotid plaque segmentation using combined BMUS and CEUS.