Dataset Information

Novel approach to estimate kidney and cyst volumes using mid-slice magnetic resonance images in polycystic kidney disease.

ABSTRACT: To evaluate whether kidney and cyst volumes can be accurately estimated based on limited area measurements from magnetic resonance (MR) images of patients with autosomal dominant polycystic kidney disease (ADPKD).MR coronal images of 178 ADPKD participants from the Consortium for Radiologic Imaging Studies of ADPKD (CRISP) were analyzed. For each MR image slice, we measured kidney and renal cyst areas using stereology and region-based thresholding methods, respectively. The kidney and cyst 'observed' volumes were calculated by summing up the area measurements of all the slices covering the kidney. To estimate the volume, we selected a coronal mid-slice in each kidney and multiplied its area by the total number of slices ('PANK2' for kidney and 'PANC2' for cyst). We then compared the kidney and cyst volumes predicted from PANK2 and PANC2, respectively, to the corresponding observed volumes, using a linear regression analysis.The kidney volume predicted from PANK2 correlated extremely well with the observed kidney volume (R(2) = 0.994 for the right kidney and 0.991 for the left kidney). The linear regression coefficient multiplier to PANK2 that best fit the kidney volume was 0.637 (95% CI: 0.629-0.644) for the right kidney and 0.624 (95% CI: 0.616-0.633) for the left kidney. The correlation between the cyst volume predicted from PANC2 and the observed cyst volume was also very high (R(2) = 0.984 for the right kidney and 0.967 for the left kidney). The least squares linear regression coefficient for PANC2 was 0.637 (95% CI: 0.624-0.649) for the right kidney and 0.608 (95% CI: 0.591-0.625) for the left kidney.Kidney and cyst volumes can be closely approximated by multiplying the product of the mid-slice area measurement and the total number of slices in the coronal MR images of ADPKD kidneys by 0.61-0.64. This information will help save processing time needed to estimate total kidney and cyst volumes of ADPKD kidneys.

SUBMITTER: Bae KT

PROVIDER: S-EPMC3863613 | biostudies-literature | 2013

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Novel approach to estimate kidney and cyst volumes using mid-slice magnetic resonance images in polycystic kidney disease.

Bae Kyongtae T KT Tao Cheng C Wang Jinhong J Kaya Diana D Wu Zhiyuan Z Bae Junu T JT Chapman Arlene B AB Torres Vicente E VE Grantham Jared J JJ Mrug Michal M Bennett William M WM Flessner Michael F MF Landsittel Doug P DP

American journal of nephrology 20131005 4

<h4>Objective</h4>To evaluate whether kidney and cyst volumes can be accurately estimated based on limited area measurements from magnetic resonance (MR) images of patients with autosomal dominant polycystic kidney disease (ADPKD).<h4>Materials and methods</h4>MR coronal images of 178 ADPKD participants from the Consortium for Radiologic Imaging Studies of ADPKD (CRISP) were analyzed. For each MR image slice, we measured kidney and renal cyst areas using stereology and region-based thresholding ...[more]

PMID: 24107679

Similar Datasets

Project description:IntroductionIn the pediatric context, most children with autosomal dominant polycystic kidney disease (ADPKD) maintain a normal glomerular filtration rate (GFR) despite underlying structural kidney damage, highlighting the critical need for early intervention and predictive markers. Due to the inverse relationship between kidney volume and kidney function, risk assessments have been presented on the basis of kidney volume. The aim of this study was to use magnetic resonance imaging (MRI)-based kidney volume assessment for risk stratification in pediatric ADPKD and to investigate clinical and genetic differences among risk groups.MethodsThis multicenter, cross-sectional, and case-control study included 75 genetically confirmed pediatric ADPKD patients (5-18 years) and 27 controls. Kidney function was assessed by eGFR calculated from serum creatinine and cystatin C using the CKiD-U25 equation. Blood pressure was assessed by both office and 24-hour ambulatory measurements. Kidney volume was calculated from MRI using the stereological method. Total kidney volume was adjusted for the height (htTKV). Patients were stratified from A to E classes according to the Leuven Imaging Classification (LIC) using MRI-derived htTKV.ResultsMedian (Q1-Q3) age of the patients was 6.0 (2.0-10.0) years, 56% were male. There were no differences in sex, age, height-SDS, or GFR between the patient and control groups. Of the patients, 89% had PKD1 and 11% had PKD2 mutations. Non-missense mutations were 73% in PKD1 and 75% in PKD2. Twenty patients (27%) had hypertension based on ABPM. Median htTKV of the patients was significantly higher than controls (141 vs. 117 ml/m, p = 0.0003). LIC stratification revealed Classes A (38.7%), B (28%), C (24%), and D + E (9.3%). All children in class D + E and 94% in class C had PKD1 variants. Class D + E patients had significantly higher blood pressure values and hypertension compared to other classes (p > 0.05 for all).DiscussionThis study distinguishes itself by using MRI-based measurements of kidney volume to stratify pediatric ADPKD patients into specific risk groups. It is important to note that PKD1 mutation and elevated blood pressure were higher in the high-risk groups stratified by age and kidney volume. Our results need to be confirmed in further studies.

Project description:BackgroundSegmentation of cardiovascular magnetic resonance (CMR) images is an essential step for evaluating dimensional and functional ventricular parameters as ejection fraction (EF) but may be limited by artifacts, which represent the major challenge to automatically derive clinical information. The aim of this study is to investigate the accuracy of a deep learning (DL) approach for automatic segmentation of cardiac structures from CMR images characterized by magnetic susceptibility artifact in patient with cardiac implanted electronic devices (CIED).MethodsIn this retrospective study, 230 patients (100 with CIED) who underwent clinically indicated CMR were used to developed and test a DL model. A novel convolutional neural network was proposed to extract the left ventricle (LV) and right (RV) ventricle endocardium and LV epicardium. In order to perform a successful segmentation, it is important the network learns to identify salient image regions even during local magnetic field inhomogeneities. The proposed network takes advantage from a spatial attention module to selectively process the most relevant information and focus on the structures of interest. To improve segmentation, especially for images with artifacts, multiple loss functions were minimized in unison. Segmentation results were assessed against manual tracings and commercial CMR analysis software cvi42(Circle Cardiovascular Imaging, Calgary, Alberta, Canada). An external dataset of 56 patients with CIED was used to assess model generalizability.ResultsIn the internal datasets, on image with artifacts, the median Dice coefficients for end-diastolic LV cavity, LV myocardium and RV cavity, were 0.93, 0.77 and 0.87 and 0.91, 0.82, and 0.83 in end-systole, respectively. The proposed method reached higher segmentation accuracy than commercial software, with performance comparable to expert inter-observer variability (bias ± 95%LoA): LVEF 1 ± 8% vs 3 ± 9%, RVEF - 2 ± 15% vs 3 ± 21%. In the external cohort, EF well correlated with manual tracing (intraclass correlation coefficient: LVEF 0.98, RVEF 0.93). The automatic approach was significant faster than manual segmentation in providing cardiac parameters (approximately 1.5 s vs 450 s).ConclusionsExperimental results show that the proposed method reached promising performance in cardiac segmentation from CMR images with susceptibility artifacts and alleviates time consuming expert physician contour segmentation.

Dataset Information

Novel approach to estimate kidney and cyst volumes using mid-slice magnetic resonance images in polycystic kidney disease.

Publications

Novel approach to estimate kidney and cyst volumes using mid-slice magnetic resonance images in polycystic kidney disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets