Stavros Tsantis

Automatic vessel lumen segmentation and stent strut detection in intravascular optical coherence tomography

PURPOSE Optical coherence tomography (OCT) is a catheter-based imaging method that employs near-infrared light to produce high-resolution cross-sectional intravascular images. The authors propose a segmentation technique for automatic lumen area extraction and stent strut detection in intravascular OCT images for the purpose of quantitative analysis of neointimal hyperplasia (NIH). METHODS A clinical dataset of frequency-domain OCT scans of the human femoral artery was analyzed. First, a segmentation method based on the Markov random field (MRF) model was employed for lumen area identification. Second, textural and edge information derived from local intensity distribution and continuous wavelet transform (CWT) analysis were integrated to extract the inner luminal contour. Finally, the stent strut positions were detected via the introduction of each strut wavelet response across scales into a feature extraction and classification scheme in order to optimize the strut position detection. RESULTS The inner lumen contour and the position of stent strut were extracted with very high accuracy. Compared with manual segmentation by an expert vascular physician the automatic segmentation had an average overlap value of 0.937 ± 0.045 for all OCT images included in the study. The strut detection accuracy had an area under the curve (AUC) value of 0.95, together with sensitivity and specificity average values of 0.91 and 0.96, respectively. CONCLUSIONS A robust automatic segmentation technique integrating textural and edge information for vessel lumen border extraction and strut detection in intravascular OCT images was designed and presented. The proposed algorithm may be employed for automated quantitative morphological analysis of in-stent neointimal hyperplasia.

Morphological and wavelet features towards sonographic thyroid nodules evaluation

This paper presents a computer-based classification scheme that utilized various morphological and novel wavelet-based features towards malignancy risk evaluation of thyroid nodules in ultrasonography. The study comprised 85 ultrasound images-patients that were cytological confirmed (54 low-risk and 31 high-risk). A set of 20 features (12 based on nodules boundary shape and 8 based on wavelet local maxima located within each nodule) has been generated. Two powerful pattern recognition algorithms (support vector machines and probabilistic neural networks) have been designed and developed in order to quantify the power of differentiation of the introduced features. A comparative study has also been held, in order to estimate the impact speckle had onto the classification procedure. The diagnostic sensitivity and specificity of both classifiers was made by means of receiver operating characteristics (ROC) analysis. In the speckle-free feature set, the area under the ROC curve was 0.96 for the support vector machines classifier whereas for the probabilistic neural networks was 0.91. In the feature set with speckle, the corresponding areas under the ROC curves were 0.88 and 0.86 respectively for the two classifiers. The proposed features can increase the classification accuracy and decrease the rate of missing and misdiagnosis in thyroid cancer control.

Inter-scale wavelet analysis for speckle reduction in thyroid ultrasound images

A wavelet-based method for speckle suppression in ultrasound images of the thyroid gland is introduced. The classification of image pixels as speckle or part of important image structures is accomplished within the framework of back-propagation tracking and singularity detection of wavelet transform modulus maxima, derived from inter-scale analysis. A comparative study with other de-speckling techniques, employing quantitative indices, demonstrated that our method achieved superior speckle reduction performance and edge preservation properties. Moreover, a questionnaire regarding qualitative imaging parameters, emanating from various visual observations, was employed by two experienced physicians in order to evaluate the algorithms speckle suppression efficiency.

Automatic quantitative analysis of in-stent restenosis using FD-OCT in vivo intra-arterial imaging

PURPOSE A new segmentation technique is implemented for automatic lumen area extraction and stent strut detection in intravascular optical coherence tomography (OCT) images for the purpose of quantitative analysis of in-stent restenosis (ISR). In addition, a user-friendly graphical user interface (GUI) is developed based on the employed algorithm toward clinical use. METHODS Four clinical datasets of frequency-domain OCT scans of the human femoral artery were analyzed. First, a segmentation method based on fuzzy C means (FCM) clustering and wavelet transform (WT) was applied toward inner luminal contour extraction. Subsequently, stent strut positions were detected by utilizing metrics derived from the local maxima of the wavelet transform into the FCM membership function. RESULTS The inner lumen contour and the position of stent strut were extracted with high precision. Compared to manual segmentation by an expert physician, the automatic lumen contour delineation had an average overlap value of 0.917 ± 0.065 for all OCT images included in the study. The strut detection procedure achieved an overall accuracy of 93.80% and successfully identified 9.57 ± 0.5 struts for every OCT image. Processing time was confined to approximately 2.5 s per OCT frame. CONCLUSIONS A new fast and robust automatic segmentation technique combining FCM and WT for lumen border extraction and strut detection in intravascular OCT images was designed and implemented. The proposed algorithm integrated in a GUI represents a step forward toward the employment of automated quantitative analysis of ISR in clinical practice.PURPOSE A new segmentation technique is implemented for automatic lumen area extraction and stent strut detection in intravascular optical coherence tomography (OCT) images for the purpose of quantitative analysis of in-stent restenosis (ISR). In addition, a user-friendly graphical user interface (GUI) is developed based on the employed algorithm toward clinical use. METHODS Four clinical datasets of frequency-domain OCT scans of the human femoral artery were analyzed. First, a segmentation method based on fuzzy C means (FCM) clustering and wavelet transform (WT) was applied toward inner luminal contour extraction. Subsequently, stent strut positions were detected by utilizing metrics derived from the local maxima of the wavelet transform into the FCM membership function. RESULTS The inner lumen contour and the position of stent strut were extracted with high precision. Compared to manual segmentation by an expert physician, the automatic lumen contour delineation had an average overlap value of 0.917 ± 0.065 for all OCT images included in the study. The strut detection procedure achieved an overall accuracy of 93.80% and successfully identified 9.57 ± 0.5 struts for every OCT image. Processing time was confined to approximately 2.5 s per OCT frame. CONCLUSIONS A new fast and robust automatic segmentation technique combining FCM and WT for lumen border extraction and strut detection in intravascular OCT images was designed and implemented. The proposed algorithm integrated in a GUI represents a step forward toward the employment of automated quantitative analysis of ISR in clinical practice.

A new computer aided diagnosis system for evaluation of chronic liver disease with ultrasound shear wave elastography imaging.

PURPOSE Classify chronic liver disease (CLD) from ultrasound shear-wave elastography (SWE) imaging by means of a computer aided diagnosis (CAD) system. METHODS The proposed algorithm employs an inverse mapping technique (red-green-blue to stiffness) to quantify 85 SWE images (54 healthy and 31 with CLD). Texture analysis is then applied involving the automatic calculation of 330 first and second order textural features from every transformed stiffness value map to determine functional features that characterize liver elasticity and describe liver condition for all available stages. Consequently, a stepwise regression analysis feature selection procedure is utilized toward a reduced feature subset that is fed into the support vector machines (SVMs) classification algorithm in the design of the CAD system. RESULTS With regard to the mapping procedure accuracy, the stiffness map values had an average difference of 0.01 ± 0.001 kPa compared to the quantification results derived from the color-box provided by the built-in software of the ultrasound system. Highest classification accuracy from the SVM model was 87.0% with sensitivity and specificity values of 83.3% and 89.1%, respectively. Receiver operating characteristic curves analysis gave an area under the curve value of 0.85 with [0.77-0.89] confidence interval. CONCLUSIONS The proposed CAD system employing color to stiffness mapping and classification algorithms offered superior results, comparing the already published clinical studies. It could prove to be of value to physicians improving the diagnostic accuracy of CLD and can be employed as a second opinion tool for avoiding unnecessary invasive procedures.

SU‐E‐I‐90: Fast and Robust Algorithm Towards Vessel Lumen and Stent Strut Detection in Optical Coherence Tomography

PURPOSE Optical Coherence Tomography (OCT) is a catheter-based imaging method that employs near-infrared light to produce high-resolution cross-sectional intravascular images. We propose a new segmentation technique for automatic lumen area extraction and stent strut detection in intravascular OCT images for the purpose of quantitative analysis of neointimal hyperplasia (NIH). METHODS Two clinical dataset of frequency-domainOCT scans of the human femoral artery were analyzed. First, a segmentation method based on Fuzzy C-Means (FCM) clustering and Wavelet Transform (WT) was applied towards inner luminal contour extraction. Subsequently, stent strut positions were detected by utilizing metrics derived from the local maxima of the wavelet transform into the FCM membership function. RESULTS The inner lumen contour and the position of stent strut were extracted with very high accuracy. Compared with manual segmentation by an expert physician, the automatic segmentation had an average overlap value of 0.917 ± 0.065 for all OCT images included in the study. The strut detection procedure successfully identified 6.7 ± 0.5 struts for each OCT image. CONCLUSIONS A new fast and robust automatic segmentation technique combining FCM and WT for lumen border extraction and strut detection in intravascular OCT images was designed and implemented. The proposed algorithm may be employed for automated quantitative morphological analysis of in-stent neointimal hyperplasia.

Focal liver lesions segmentation and classification in nonenhanced T2‐weighted MRI

Purpose To automatically segment and classify focal liver lesions (FLLs) on nonenhanced T2‐weighted magnetic resonance imaging (MRI) scans using a computer‐aided diagnosis (CAD) algorithm. Methods 71 FLLs (30 benign lesions, 19 hepatocellular carcinomas, and 22 metastases) on T2‐weighted MRI scans were delineated by the proposed CAD scheme. The FLL segmentation procedure involved wavelet multiscale analysis to extract accurate edge information and mean intensity values for consecutive edges computed using horizontal and vertical analysis that were fed into the subsequent fuzzy C‐means algorithm for final FLL border extraction. Texture information for each extracted lesion was derived using 42 first‐ and second‐order textural features from grayscale value histogram, co‐occurrence, and run‐length matrices. Twelve morphological features were also extracted to capture any shape differentiation between classes. Feature selection was performed with stepwise multilinear regression analysis that led to a reduced feature subset. A multiclass Probabilistic Neural Network (PNN) classifier was then designed and used for lesion classification. PNN model evaluation was performed using the leave‐one‐out (LOO) method and receiver operating characteristic (ROC) curve analysis. Results The mean overlap between the automatically segmented FLLs and the manual segmentations performed by radiologists was 0.91 ± 0.12. The highest classification accuracies in the PNN model for the benign, hepatocellular carcinoma, and metastatic FLLs were 94.1%, 91.4%, and 94.1%, respectively, with sensitivity/specificity values of 90%/97.3%, 89.5%/92.2%, and 90.9%/95.6% respectively. The overall classification accuracy for the proposed system was 90.1%. Conclusions Our diagnostic system using sophisticated FLL segmentation and classification algorithms is a powerful tool for routine clinical MRI‐based liver evaluation and can be a supplement to contrast‐enhanced MRI to prevent unnecessary invasive procedures.

Development of a Support Vector Machine - Based Image Analysis System for Focal Liver Lesions Classification in Magnetic Resonance Images

Purpose: The design and implementation of a computer-based image analysis system employing the support vector machine (SVM) classifier system for the classification of Focal Liver Lesions (FLLs) on routine non-enhanced, T2-weighted Magnetic Resonance (MR) images. Materials and Methods: The study comprised 92 patients; each one of them has undergone MRI performed on a Magnetom Concerto (Siemens). Typical signs on dynamic contrast-enhanced MRI and biopsies were employed towards a three class categorization of the 92 cases: 40-benign FLLs, 25-Hepatocellular Carcinomas (HCC) within Cirrhotic liver parenchyma and 27-liver metastases from Non-Cirrhotic liver. Prior to FLLs classification an automated lesion segmentation algorithm based on Marcov Random Fields was employed in order to acquire each FLL Region of Interest. 42 texture features derived from the gray-level histogram, co-occurrence and run-length matrices and 12 morphological features were obtained from each lesion. Stepwise multi-linear regression analysis was utilized to avoid feature redundancy leading to a feature subset that fed the multiclass SVM classifier designed for lesion classification. SVM System evaluation was performed by means of leave-one-out method and ROC analysis. Results: Maximum accuracy for all three classes (90.0%) was obtained by means of the Radial Basis Kernel Function and three textural features (Inverse- Different-Moment, Sum-Variance and Long-Run-Emphasis) that describe lesions contrast, variability and shape complexity. Sensitivity values for the three classes were 92.5%, 81.5% and 96.2% respectively, whereas specificity values were 94.2%, 95.3% and 95.5%. The AUC value achieved for the selected subset was 0.89 with 0.81 - 0.94 confidence interval. Conclusion: The proposed SVM system exhibit promising results that could be utilized as a second opinion tool to the radiologist in order to decrease the time/cost of diagnosis and the need for patients to undergo invasive examination.

TU-H-CAMPUS-IeP3-05: Computer Aided Diagnosis Employing Automatically Segmented Color-Specific Regions in Ultrasound Shear Wave Elastography for the Assessment of Chronic Liver Disease

PURPOSE To assess the role of an automated Computer Aided Diagnosis (CAD) system in differentiation of Healthy Subjects to Chronic Liver Disease (CLD) Patients in terms of liver fibrosis (F0-F4), using Ultrasound Shear Wave Elastography (SWE). METHODS Clinical Dataset consisted of 125 subjects, 55 Healthy (F0) whom condition was validated with Normal Biochemical Markers and clear Clinical History, and 70 with CLD (F1-F4) whom condition was validated with Liver Biopsy. For each subject an SWE examination on their liver right lobe was performed using Supersonic Imagines Aixplorer ultrasound system, and an SC6-1 transducer. The examination was performed using preset default settings for Abdomen/Liver, and according to literatures suggested guidelines. An SWE image was then acquired, and a separation of the colored areas from the grayscale ones was performed. An RGB-to-Stiffness algorithm was consequently applied on the resulting image (Stiffness Box), using the nbuilt-in colormap, to replace RGB values to Stiffness values. A segmentation procedure was employed using thresholds that segment the image to 5 stiffness range clusters (0-6, 6-12, 12-18, 18-24, and 24-30 kPa), and extract 7 features from each cluster (Mean, Median, Standard Deviation, Sample Kurtosis, 10th Percentile, 90th Percentile, and Clusters Pixel Percentage on the Stiffness Box) deriving 35 features for each SWE image. The computed feature set was then fed to a designed SVM classifier. SVM-model evaluation was performed by means of exhaustive search, and leave-one-out methods. RESULTS Maximum classification accuracy (85.6%) in distinguishing Healthy from chronic liver disease patients was obtained employing three features (Blue mean and Cyan median values, and Red maximum pixel number) with sensitivity and specificity values of 91.9% and 79.3% respectively. CONCLUSION The proposed CAD system accurately differentiates Healthy to CLD patients assisting in evaluating important factors for performing SWE examination, enriching its guidelines, and aiding clinicians in a more accurate diagnosis.

SU‐E‐U‐01: Automatic Quantitative Analysis of Chronic Liver Disease Employing Shear Wave Ultrasound Elastography

Purpose: The purpose of this study was to quantify liver elastic heterogeneity in Shear Wave Elastography (SWE) by using textural features and evaluating their diagnostic performance on differentiating healthy from chronic liver disease patients, taking biopsy results as the gold standard. Methods: Clinical material includes 16 healthy (F0) and 15 with Chronic Liver Disease (F1,F2,F3,F4) patients according to the Metavir staging system. All exams were performed using the Aixplorer ultrasound system with a SuperCurved SC6-1 transducer. From the SWE-QBox the RGB displayed elasticity data of Young’s modulus were transformed from RGB color space into an elasticity matrix of gray tones, whose values varied from zero to the maximum elasticity measurement. Every pixel with no RGB-values was set to ‘−1’ due to non-valid elasticity value for that pixel. From the elastogram map 185 textural features were computed (5 from the gray-tone histogram, 26 second order statistic features, extracted from the co-occurrence matrices and 10 features extracted from the run-length matrices over four directions (00,450,900,1350) and distances of (d=1,3,5,7,9) pixels). Stepwise multi-linear regression analysis was utilized to avoid feature redundancy leading to a feature subset feeding a Support Vector Machine (SVM) classifier. SVM-model evaluation was performed by means of the leave-one-out method. Results: Maximum classification accuracy (93.5%) in distinguishing healthy from chronic liver disease patients was obtained employing three textural features (Standard Deviation, Sum-Variance, Contrast) that describe the elastogram’s contrast, variability and complexity. Sensitivity and specificity values were 93.3% and 94.0% respectively. Conclusion: The proposed classification scheme can provide reproducibility and reliability of liver SWE application in clinical practice. It can also assist the interpretation of SWE measurements which is considered as a difficult task due to absence of guidelines in the literature and to decrease the time/cost of diagnosis and the need for patients to undergo invasive examination. This research has been co-financed by the European Union (European Social Fund ESF) and Greek national funds through the Operational Program ’Education and Lifelong Learning’ of the National Strategic Reference Framework (NSRF) Research Funding Program: ARCHIMEDES III. Investing in knowledge society through the European Social Fund.