Quantification of common carotid artery and descending aorta vessel wall thickness from MR vessel wall imaging using a fully automated processing pipeline.

PURPOSE: To develop and evaluate a method that can fully automatically identify the vessel wall boundaries and quantify the wall thickness for both common carotid artery (CCA) and descending aorta (DAO) from axial magnetic resonance (MR) images. MATERIALS AND METHODS: 3T MRI data acquired with T1 -weighted gradient-echo black-blood imaging sequence from carotid (39 subjects) and aorta (39 subjects) were used to develop and test the algorithm. The vessel wall segmentation was achieved by respectively fitting a 3D cylindrical B-spline surface to the boundaries of lumen and outer wall. The tube-fitting was based on the edge detection performed on the signal intensity (SI) profile along the surface normal. To achieve a fully automated process, Hough Transform (HT) was developed to estimate the lumen centerline and radii for the target vessel. Using the outputs of HT, a tube model for lumen segmentation was initialized and deformed to fit the image data. Finally, lumen segmentation was dilated to initiate the adaptation procedure of outer wall tube. The algorithm was validated by determining: 1) its performance against manual tracing; 2) its interscan reproducibility in quantifying vessel wall thickness (VWT); 3) its capability of detecting VWT difference in hypertensive patients compared with healthy controls. Statistical analysis including Bland-Altman analysis, t-test, and sample size calculation were performed for the purpose of algorithm evaluation. RESULTS: The mean distance between the manual and automatically detected lumen/outer wall contours was 0.00 ± 0.23/0.09 ± 0.21 mm for CCA and 0.12 ± 0.24/0.14 ± 0.35 mm for DAO. No significant difference was observed between the interscan VWT assessment using automated segmentation for both CCA (P = 0.19) and DAO (P = 0.94). Both manual and automated segmentation detected significantly higher carotid (P = 0.016 and P = 0.005) and aortic (P < 0.001 and P = 0.021) wall thickness in the hypertensive patients. CONCLUSION: A reliable and reproducible pipeline for fully automatic vessel wall quantification was developed and validated on healthy volunteers as well as patients with increased vessel wall thickness. This method holds promise for helping in efficient image interpretation for large-scale cohort studies. LEVEL OF EVIDENCE: 4 J. Magn. Reson. Imaging 2017;45:215-228.

Repeatability of in vivo quantification of atherosclerotic carotid artery plaque components by supervised multispectral classification.

OBJECTIVE: To evaluate the agreement and scan-rescan repeatability of automated and manual plaque segmentation for the quantification of in vivo carotid artery plaque components from multi-contrast MRI. MATERIALS AND METHODS: Twenty-three patients with 30-70% stenosis underwent two 3T MR carotid vessel wall exams within a 1 month interval. T1w, T2w, PDw and TOF images were acquired around the region of maximum vessel narrowing. Manual delineation of the vessel wall and plaque components (lipid, calcification, loose matrix) by an experienced observer provided the reference standard for training and evaluation of an automated plaque classifier. Areas of different plaque components and fibrous tissue were quantified and compared between segmentation methods and scan sessions. RESULTS: In total, 304 slices from 23 patients were included in the segmentation experiment, in which 144 aligned slice pairs were available for repeatability analysis. The correlation between manual and automated segmented areas was 0.35 for lipid, 0.66 for calcification, 0.50 for loose matrix and 0.82 for fibrous tissue. For the comparison between scan sessions, the coefficient of repeatability of area measurement obtained by automated segmentation was lower than by manual delineation for lipid (9.9 vs. 17.1 mm(2)), loose matrix (13.8 vs. 21.2 mm(2)) and fibrous tissue (24.6 vs. 35.0 mm(2)), and was similar for calcification (20.0 vs. 17.6 mm(2)). CONCLUSION: Application of an automated classifier for segmentation of carotid vessel wall plaque components from in vivo MRI results in improved scan-rescan repeatability compared to manual analysis.

Evaluation of multicontrast MRI including fat suppression and inversion recovery spin echo for identification of intra-plaque hemorrhage and lipid core in human carotid plaque using the mahalanobis distance measure.

Intra-plaque hemorrhage (IPH) and lipid core, characteristics of rupture prone carotid plaques, are often visualized in vivo with MRI using T1 weighted gradient and spin echo, respectively. Increasing magnetic field strength may help to identify IPH and lipid core better. As a proof of concept, automatic segmentation of plaque components was performed with the Mahalanobis distance (MD) measure derived from image contrast from multicontrast MR images including inversion recovery spin echo and T1 weighted gradient echo with fat suppression. After MRI of nine formaldehyde-fixated autopsy specimens, the MDs and Euclidean Distances between plaque component intensities were calculated for each MR weighting. The distances from the carotid bifurcation and the size and shape of calcification spots were used as landmarks for coregistration of MRI and histology. MD between collagen/cell-rich area and IPH was largest with inversion recovery spin echo (4.2/9.3, respectively), between collagen/cell-rich area/foam cells and lipid core with T1 weighted gradient echo with fat suppression (26.9/38.2/4.6, respectively). The accuracy of detection of IPH, cell-rich area, and collagen increased when the MD classifier was used compared with the Euclidean Distance classifier. The enhanced conspicuity of lipid core and IPH in human carotid artery plaque, using ex vivo T1 weighted gradient echo with fat suppression and inversion recovery spin echo MRI and MD classifiers, demands further in vivo evaluation in patients.

Automatic lumen and outer wall segmentation of the carotid artery using deformable three-dimensional models in MR angiography and vessel wall images.

PURPOSE: To develop and validate an automated segmentation technique for the detection of the lumen and outer wall boundaries in MR vessel wall studies of the common carotid artery. MATERIALS AND METHODS: A new segmentation method was developed using a three-dimensional (3D) deformable vessel model requiring only one single user interaction by combining 3D MR angiography (MRA) and 2D vessel wall images. This vessel model is a 3D cylindrical Non-Uniform Rational B-Spline (NURBS) surface which can be deformed to fit the underlying image data. Image data of 45 subjects was used to validate the method by comparing manual and automatic segmentations. Vessel wall thickness and volume measurements obtained by both methods were compared. RESULTS: Substantial agreement was observed between manual and automatic segmentation; over 85% of the vessel wall contours were segmented successfully. The interclass correlation was 0.690 for the vessel wall thickness and 0.793 for the vessel wall volume. Compared with manual image analysis, the automated method demonstrated improved interobserver agreement and inter-scan reproducibility. Additionally, the proposed automated image analysis approach was substantially faster. CONCLUSION: This new automated method can reduce analysis time and enhance reproducibility of the quantification of vessel wall dimensions in clinical studies.

Automatic radiographic quantification of hand osteoarthritis; accuracy and sensitivity to change in joint space width in a phantom and cadaver study.

OBJECTIVE: To validate a newly developed quantification method that automatically detects and quantifies the joint space width (JSW) in hand radiographs. Repeatability, accuracy and sensitivity to changes in JSW were determined. The influence of joint location and joint shape on the measurements was tested. METHODS: A mechanical micrometer set-up was developed to define and adjust the true JSW in an acrylic phantom joint and in human cadaver-derived phalangeal joints. Radiographic measurements of the JSW were compared to the true JSW. Repeatability, systematic error (accuracy) and sensitivity (defined as the smallest detectable difference (SDD)) were determined. The influence of joint position on the JSW measurement was assessed by varying the location of the acrylic phantom on the X-ray detector with respect to the X-ray beam and the influence of joint shape was determined by using morphologically different human cadaver joints. RESULTS: The mean systematic error was 0.052 mm in the phantom joint and 0.210 mm in the cadaver experiment. In the phantom experiments, the repeatability was high (SDD = 0.028 mm), but differed slightly between joint locations (p = 0.046), and a change in JSW of 0.037 mm could be detected. Dependent of the joint shape in the cadaver hand, a change in JSW between 0.018 and 0.047 mm could be detected. CONCLUSIONS: The automatic quantification method is sensitive to small changes in JSW. Considering the published data of JSW decline in the normal and osteoarthritic population, the first signs of OA progression with this method can be detected within 1 or 2 years.

Learning-based automated segmentation of the carotid artery vessel wall in dual-sequence MRI using subdivision surface fitting.

PURPOSE: The quantification of vessel wall morphology and plaque burden requires vessel segmentation, which is generally performed by manual delineations. The purpose of our work is to develop and evaluate a new 3D model-based approach for carotid artery wall segmentation from dual-sequence MRI. METHODS: The proposed method segments the lumen and outer wall surfaces including the bifurcation region by fitting a subdivision surface constructed hierarchical-tree model to the image data. In particular, a hybrid segmentation which combines deformable model fitting with boundary classification was applied to extract the lumen surface. The 3D model ensures the correct shape and topology of the carotid artery, while the boundary classification uses combined image information of 3D TOF-MRA and 3D BB-MRI to promote accurate delineation of the lumen boundaries. The proposed algorithm was validated on 25 subjects (48 arteries) including both healthy volunteers and atherosclerotic patients with 30% to 70% carotid stenosis. RESULTS: For both lumen and outer wall border detection, our result shows good agreement between manually and automatically determined contours, with contour-to-contour distance less than 1 pixel as well as Dice overlap greater than 0.87 at all different carotid artery sections. CONCLUSIONS: The presented 3D segmentation technique has demonstrated the capability of providing vessel wall delineation for 3D carotid MRI data with high accuracy and limited user interaction. This brings benefits to large-scale patient studies for assessing the effect of pharmacological treatment of atherosclerosis by reducing image analysis time and bias between human observers.

Determinants of carotid atherosclerotic plaque burden in a stroke-free population.

BACKGROUND AND AIMS: In a large stroke-free population, we sought to identify cardiovascular risk factors and carotid plaque components associated with carotid plaque burden, lumen volume and stenosis. METHODS: The carotid arteries of 1562 stroke-free participants from The Rotterdam Study were imaged on a 1.5-Tesla MRI scanner. Inner and outer wall of the carotid arteries were automatically segmented and lumen volume (mm3), wall volume (outer wall-inner wall) and plaque burden (wall volume/outer wall volume) (%) were quantified. Plaque components were visually determined and luminal stenosis was assessed. We analyzed associations of cardiovascular risk factors and carotid plaque components with plaque burden and lumen volumes using regression analysis. RESULTS: We investigated 2821 carotid plaques and found that women had larger plaque burden (50.7 ± 7.8% vs. 49.2 ± 7.7%, p < 0.0001) and smaller lumen volumes (933 ± 286 mm3vs. 1078 ± 334 mm3, p < 0.0001) than men. In women, age, HDL-cholesterol and systolic blood pressure, and in men, total cholesterol, non-HDL cholesterol and statin use were independently associated with higher plaque burden and lumen volume. Furthermore, smoking and diabetes were associated with lumen volume in men (respectively p = 0.04 and p = 0.002). Intraplaque hemorrhage (IPH) and lipid were related to a larger plaque burden (OR 1.30 [1.05-1.60] and OR 1.28[1.06-1.55]). Finally, within the highest quartile of plaque burden, IPH was strongly associated with luminal stenosis independent of age, sex, plaque burden and composition (Beta = 15.2; [11.8-18.6]). CONCLUSIONS: Several cardiovascular risk factors and plaque components, in particular IPH, are associated with higher plaque burden. Carotid IPH is strongly associated with an increased luminal stenosis.

Identification of factors associated with the development of knee osteoarthritis in a young to middle-aged cohort of patients with knee complaints.

The objective of this study was to identify risk factors for knee osteoarthritis (OA) development in a young to middle-aged population with sub-acute knee complaints. This, in order to define high risk patients who may benefit from early preventive or future disease modifying therapies. Knee OA development visible on radiographs and MR in 319 patients (mean age 41.5 years) 10 years after sub-acute knee complaints and subjective knee function (KOOS score) was studied. Associations between OA development and age, gender, activity level, BMI, meniscal or anterior cruciate ligament (ACL) lesions, OA in first-degree relatives and radiographic hand OA were determined using multivariable logistic regression analysis. OA on radiographs and MR in the TFC is associated with increased age (OR: 1.10, 95 % 1.04-1.16 and OR: 1.07, 95 % 1.02-1.13). TFC OA on radiographs only is associated with ACL and/or meniscal lesions (OR: 5.01, 95 % 2.14-11.73), presence of hand OA (OR: 4.69, 95 % 1.35-16.32) and higher Tegner activity scores at baseline before the complaints (OR: 1.20, 95 % 1.01-1.43). The presence of OA in the TFC diagnosed only on MRI is associated with a family history of OA (OR: 2.44, 95 % 1.18-5.06) and a higher BMI (OR: 1.13, 95 % 1.04-1.23). OA in the PFC diagnosed on both radiographs and MR is associated with an increased age (OR: 1.06, 95 % 1.02-1.12 and OR: 1.05, 95 % 1.00-1.09). PFC OA diagnosed on radiographs only is associated with a higher BMI (OR: 1.12, 95 % 1.02-1.22). The presence of OA in the PFC diagnosed on MR only is associated with the presence of hand OA (OR: 3.39, 95 % 1.10-10.50). Compared to normal reference values, the study population had significantly lower KOOS scores in the different subscales. These results show that knee OA development in young to middle aged patients with a history of sub-acute knee complaints is associated with the presence of known risk factors for knee OA. OA is already visible on radiographs and MRI after 10 years. These high risk patients may benefit from adequate OA management early in life.

Multi-Center MRI Carotid Plaque Component Segmentation Using Feature Normalization and Transfer Learning.

Automated segmentation of plaque components in carotid artery magnetic resonance imaging (MRI) is important to enable large studies on plaque vulnerability, and for incorporating plaque composition as an imaging biomarker in clinical practice. Especially supervised classification techniques, which learn from labeled examples, have shown good performance. However, a disadvantage of supervised methods is their reduced performance on data different from the training data, for example on images acquired with different scanners. Reducing the amount of manual annotations required for each new dataset will facilitate widespread implementation of supervised methods. In this paper we segment carotid plaque components of clinical interest (fibrous tissue, lipid tissue, calcification and intraplaque hemorrhage) in a multi-center MRI study. We perform voxelwise tissue classification by traditional same-center training, and compare results with two approaches that use little or no annotated same-center data. These approaches additionally use an annotated set of different-center data. We evaluate 1) a nonlinear feature normalization approach, and 2) two transfer-learning algorithms that use same and different-center data with different weights. Results showed that the best results were obtained for a combination of feature normalization and transfer learning. While for the other approaches significant differences in voxelwise or mean volume errors were found compared with the reference same-center training, the proposed approach did not yield significant differences from that reference. We conclude that both extensive feature normalization and transfer learning can be valuable for the development of supervised methods that perform well on different types of datasets.

An objective method to optimize the MR sequence set for plaque classification in carotid vessel wall images using automated image segmentation.

A typical MR imaging protocol to study the status of atherosclerosis in the carotid artery consists of the application of multiple MR sequences. Since scanner time is limited, a balance has to be reached between the duration of the applied MR protocol and the quantity and quality of the resulting images which are needed to assess the disease. In this study an objective method to optimize the MR sequence set for classification of soft plaque in vessel wall images of the carotid artery using automated image segmentation was developed. The automated method employs statistical pattern recognition techniques and was developed based on an extensive set of MR contrast weightings and corresponding manual segmentations of the vessel wall and soft plaque components, which were validated by histological sections. Evaluation of the results from nine contrast weightings showed the tradeoff between scan duration and automated image segmentation performance. For our dataset the best segmentation performance was achieved by selecting five contrast weightings. Similar performance was achieved with a set of three contrast weightings, which resulted in a reduction of scan time by more than 60%. The presented approach can help others to optimize MR imaging protocols by investigating the tradeoff between scan duration and automated image segmentation performance possibly leading to shorter scanning times and better image interpretation. This approach can potentially also be applied to other research fields focusing on different diseases and anatomical regions.