Automated Contour Detection in Cardiac MRI Using Active Appearance Models: The Effect of the Composition of the Training Set

Automated Artery Localization and Vessel Wall Segmentation using Tracklet Refinement and Polar Conversion

Quantitative analysis of blood vessel wall structures is important to study atherosclerotic diseases and assess cardiovascular event risks. To achieve this, accurate identification of vessel luminal and outer wall contours is needed. Computer-assisted tools exist, but manual preprocessing steps, such as region of interest identification and/or boundary initialization, are still needed. In addition, prior knowledge of the ring shape of vessel walls has not been fully explored in designing segmentation methods. In this work, a fully automated artery localization and vessel wall segmentation system is proposed. A tracklet refinement algorithm was adapted to robustly identify the artery of interest from a neural network-based artery centerline identification architecture. Image patches were extracted from the centerlines and converted in a polar coordinate system for vessel wall segmentation. The segmentation method used 3D polar information and overcame problems such as contour discontinuity, complex vessel geometry, and interference from neighboring vessels. Verified by a large (>32000 images) carotid artery dataset collected from multiple sites, the proposed system was shown to better automatically segment the vessel wall than traditional vessel wall segmentation methods or standard convolutional neural network approaches. In addition, a segmentation uncertainty score was estimated to effectively identify slices likely to have errors and prompt manual confirmation of the segmentation. This robust vessel wall segmentation system has applications in different vascular beds and will facilitate vessel wall feature extraction and cardiovascular risk assessment.

Pediatric cardiac MRI: automated left-ventricular volumes and function analysis and effects of manual adjustments

BACKGROUND

Cardiac MRI is an accurate and reproducible technique for the assessment of left ventricular volumes and function. The accuracy of automated segmentation and the effects of manual adjustments have not been determined in children.

OBJECTIVE

To evaluate automated segmentation and the effects of manual adjustments for left ventricular parameter quantification in pediatric cardiac MR images.

MATERIALS AND METHODS

Left ventricular parameters were evaluated in 45 children with suspected myocarditis (age 13.4 ± 3.5 years, range 4-17 years) who underwent cardiac MRI. Dedicated software was used to automatically segment and adjust the parameters. Results of end-diastolic volume, end-systolic volume, stroke volume, myocardial mass, and ejection fraction were documented before and after apex/base adjustment and after apex/base/myocardial contour adjustment.

RESULTS

The software successfully detected the left ventricle in 42 of 45 (93.3%) children; failures occurred in the smallest and youngest children. Of those 42 children, automatically segmented end-diastolic volume (EDV) was 151 ± 47 ml, and after apex/base adjustment it was 146 ± 45 ml, after apex/base/myocardial contour adjustment 146 ± 45 ml. The corresponding results for end-systolic volume (ESV) were 66 ± 32 ml, 63 ± 29 ml and 64 ± 28 ml; for stroke volume (SV) they were 85 ± 25 ml, 83 ± 23 ml and 83 ± 23 ml; for ejection fracture (EF) they were 57 ± 10%, 58 ± 9% and 58 ± 9%, and for myocardial mass (MM) they were 104 ± 31 g, 95 ± 31 g and 94 ± 30 g. Statistically significant differences were found when comparing the EDV/ESV/MM results, the EF results after apex/base adjustment and after apex/base/myocardial contour adjustment and the SV results (except for comparing the SVs after apex/base adjustment and after apex/base/myocardial contour adjustment).

CONCLUSION

Automated segmentation for the evaluation of left ventricular parameters in pediatric MR images proved to be feasible. Automated segmentation + apex/base adjustment provided clinically acceptable parameters for the majority of cases.

Delayed improvement of right ventricular diastolic function and regression of right ventricular mass after percutaneous pulmonary valve implantation in patients with congenital heart disease

BACKGROUND

Percutaneous pulmonary valve implantation (PPVI) has been introduced as therapy for right ventricular (RV) to pulmonary artery conduit dysfunction in patients with congenital heart disease. It has been shown that RV systolic function improved early after PPVI. The effects of PPVI on RV diastolic function and RV hypertrophy have not yet been studied.

PURPOSE

The objective of this study is to assess early and late changes in systolic and diastolic RV function and RV mass after PPVI.

MATERIALS AND METHODS

Fourteen patients underwent PPVI (7 male, median age 15 years). Cardiac magnetic resonance imaging was performed before and at 2 time points after PPVI (at 1 and 16 months). Right ventricular volume and systolic and diastolic function as well as RV mass were assessed.

RESULTS

At 1 and 16 months after PPVI, the RV mass decreased from 28.6 +/- 2.1 to 25.6 +/- 2.2 g/m(2) (P = .03) and to 22.3 +/- 2.1 g/m(2) (P = .002). E/A volume ratio increased from 1.91 +/- 0.4 to 2.6 +/- 0.4 (not significant [NS]) and to 3.3 +/- 0.4 (P = .01). E/A peak flow ratio increased from 1.34 +/- 0.14 to 1.48 +/- 0.16 (NS) and to 1.73 +/- 0.14 (P = .04). E-wave deceleration time increased from 142 +/- 25 to 160 +/- 27 milliseconds (NS) and to 211 +/- 26 milliseconds (P = .007). At 1 month, RV end-diastolic volume decreased from 124 +/- 8 to 113 +/- 8 mL (P = .01) and RV ejection fraction increased from 36% +/- 2% to 46% +/- 2% (P = .001) without further improvement at 16 months.

CONCLUSION

After PPVI, in contrast to rapid improvement of RV systolic function, the improvement of RV diastolic function is delayed. The reduction of RV mass appears to be the underlying mechanism for improvement of RV diastolic function. Long follow-up for patients with PPVI is recommended.

Automated frame-by-frame endocardial border detection from cardiac magnetic resonance images for quantitative assessment of left ventricular function: validation and clinical feasibility

PURPOSE

To develop a technique based on image noise distribution for automated endocardial border detection from cardiac magnetic resonance (CMR) images throughout the cardiac cycle, validate it, and test its clinical utility.

MATERIALS AND METHODS

Images obtained in 36 patients were analyzed using custom software to obtain left ventricular (LV) volume throughout the cardiac cycle, end-systolic and end-diastolic LV volumes, and ejection fraction (EF). Validation against manually-traced endocardial boundaries included intertechnique comparisons of LV volumes, slice areas, and border positions. Then, the clinical feasibility of the dynamic automated analysis of LV function was tested in 14 patients with normal LV function, 12 patients with systolic dysfunction, and 10 patients with diastolic dysfunction.

RESULTS

Analysis time for one cardiac cycle was <15 minutes. Intertechnique comparisons resulted in high correlation (r>0.96), small biases (volumes: -6 mL; EF: 4.6%) and narrow limits of agreement (volumes: 17.6 mL; EF: 9.2%). We found significant intergroup differences in multiple quantitative indices of systolic and diastolic function.

CONCLUSION

Fast, automated, dynamic detection of LV endocardial boundaries is feasible and allows accurate quantification of LV size and function, which is potentially clinically useful for objective assessment of systolic and diastolic dysfunction.

Advances in magnetic resonance (2008)

Current advances in magnetic resonance, as a diagnostic modality, are discussed in the context of publications from Investigative Radiology during 2007 and 2008. The articles relating to this topic, published during the past 2 years, are reviewed by anatomic region. The discussion concludes with a consideration of magnetic resonance contrast media, focusing on studies published in the journal, and examining in particular the potential impact of nephrogenic systemic fibrosis.