Automatic renal segmentation for MR urography using 3D‐GrabCut and random forests

Yoruk, Umit; Hargreaves, Brian A.; Vasanawala, Shreyas

doi:10.1002/mrm.26806

Cited by 27 publications

(25 citation statements)

References 40 publications

(64 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AI methods have been used for automatic segmentation of the cortex and medulla for renal volumetry based on T 1weighted imaging, with [50] and without gadolinium contrast enhancement [51][52][53]. The ROIs obtained by automatic segmentation of T 1 -weighted images can inform segmentation of the cortex and medulla in a multiparametric MRI protocol.…”

Section: Remaining Challenges For Future Researchmentioning

confidence: 99%

Consensus-based technical recommendations for clinical translation of renal BOLD MRI

Bane

Mendichovszky

Milani

et al. 2019

Magn Reson Mater Phy

View full text Add to dashboard Cite

Harmonization of acquisition and analysis protocols is an important step in the validation of BOLD MRI as a renal biomarker. This harmonization initiative provides technical recommendations based on a consensus report with the aim to move towards standardized protocols that facilitate clinical translation and comparison of data across sites. We used a recently published systematic review paper, which included a detailed summary of renal BOLD MRI technical parameters and areas of investigation in its supplementary material, as the starting point in developing the survey questionnaires for seeking consensus. Survey data were collected via the Delphi consensus process from 24 researchers on renal BOLD MRI exam preparation, data acquisition, data analysis, and interpretation. Consensus was defined as ≥ 75% unanimity in response. Among 31 survey questions, 14 achieved consensus resolution, 12 showed clear respondent preference (65-74% agreement), and 5 showed equal (50/50%) split in opinion among respondents. Recommendations for subject preparation, data acquisition, processing and reporting are given based on the survey results and review of the literature. These technical recommendations are aimed towards increased inter-site harmonization, a first step towards standardization of renal BOLD MRI protocols across sites. We expect this to be an iterative process updated dynamically based on progress in the field.

show abstract

Section: Remaining Challenges For Future Researchmentioning

confidence: 99%

Consensus-based technical recommendations for clinical translation of renal BOLD MRI

Bane

Mendichovszky

Milani

et al. 2019

Magn Reson Mater Phy

View full text Add to dashboard Cite

show abstract

“…Most of ML methods used for kidney image segmentation are automatic, they do not need any user interaction [38], [45], [60]. However, both supervised and unsupervised methods are often supported by various image pre-or postprocessing procedures e.g.…”

Section: Machine Learning and Deep Learning Approachesmentioning

confidence: 99%

“…A few GitHub repositories on kidney segmentation are freely available. Yoruk [38], [107] aimed at evaluation of a fully automated renal segmentation technique for glomerular filtration rate (GFR) assessment in children. Yang et.…”

Section: Software and Github Projects Relevant To Kidney Image Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Kidney Segmentation in Renal Magnetic Resonance Imaging - Current Status and Prospects

et al. 2021

View full text Add to dashboard Cite

Magnetic resonance imaging has achieved an increasingly important role in the clinical workup of renal diseases such chronic kidney disease (CKD). A large panel of parameters have been proposed to diagnose CKD among them total kidney volume (TKV) which recently qualified as biomarker. Volume estimation in renal MRI is based on image segmentation of the kidney and/or its compartments. Beyond volume estimation renal segmentation supports also the quantification of other MR based parameters such as perfusion or filtration. The aim of the present article is to discuss the recent existing literature on renal image segmentation techniques and show today's limitations of the proposed techniques that might hinder clinical translation. We also provide pointers to open source software related to renal image segmentation.

show abstract

“…In other renal MR imaging applications, besides ASL, different segmentation processes were introduced to segment renal tissue compartments, utilizing computer vision tools like shape objects on T1 and T2 high resolution anatomical images, 109 concentric objects on BOLD MR images 110 and a combination of wavelet transform and k-means clustering on DCE MR images. 111,112 However, it is unsure if their translation to renal ASL MRI would succeed due to missing key features, e.g. time course of contrast enhancement, for the respective mathematical process.…”

Section: Introductionmentioning

confidence: 99%

Technical advances paving the way for non-invasive renal perfusion MRI

Bones¹

View full text Add to dashboard Cite

Another important part of renal ASL is the process of signal quantification. Up to this point, it requires manual interaction 25,29,42 and with that introduces inter-observer variability and makes the extraction of cortical RBF an unattractive, lengthy process. Two processes specifically require manual interaction: localization of the kidneys for recommended kidney wise motion correction, 11 as they can move differently, 43,44 and renal tissue differentiation, as it is recommended to report ASL MRI quantification results in terms of cortical renal blood flow (RBF), meaning that only signal from cortical voxels is taken into account.

show abstract

Automatic renal segmentation for MR urography using 3D‐GrabCut and random forests

Abstract: The automatic segmentation method performs as well as the manual segmentation for GFR estimation and reduces the segmentation time from several hours to 45 s. Magn Reson Med 79:1696-1707, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Cited by 27 publications

References 40 publications

Consensus-based technical recommendations for clinical translation of renal BOLD MRI

Consensus-based technical recommendations for clinical translation of renal BOLD MRI

Kidney Segmentation in Renal Magnetic Resonance Imaging - Current Status and Prospects

Technical advances paving the way for non-invasive renal perfusion MRI

Contact Info

Product

Resources

About