Evaluation of renal dysfunction using texture analysis based on DWI, BOLD, and susceptibility-weighted imaging
To explore the value of texture analysis based on diffusion-weighted imaging (DWI), blood oxygen level–dependent MRI (BOLD), and susceptibility-weighted imaging (SWI) in evaluating renal dysfunction.
Seventy-two patients (mean age 53.72 ± 13.46 years) underwent MRI consisting of DWI, BOLD, and SWI. According to their estimated glomerular filtration rate (eGFR), the patients were classified into either severe renal function impairment (sRI, eGFR < 30 mL/min/1.73 m2), non-severe renal function impairment (non-sRI, eGFR ≥ 30 mL/min/1.73 m2, and < 80 mL/min/1.73 m2), or control (CG, eGFR ≥ 80 mL/min/1.73 m2) groups. Thirteen texture features were extracted and then were analyzed to select the most valuable for discerning the three groups with each imaging method. A ROC curve was performed to compare the capacities of the features to differentiate non-sRI from sRI or CG.
Six features proved to be the most valuable for assessing renal dysfunction: 0.25QuantileDWI, 0.5QuantileDWI, HomogeneityDWI, EntropyBOLD, SkewnessSWI, and CorrelationSWI. Three features derived from DWI (0.25QuantileDWI, 0.5QuantileDWI, and HomogeneityDWI) were smaller in sRI than in non-sRI; EntropyBOLD and CorrelationSWI were smaller in non-sRI than in CG (p < 0.05). 0.25QuantileDWI, 0.5QuantileDWI, and HomogeneityDWI showed similar capacities for differentiating sRI from non-sRI. Similarly, EntropyBOLD and CorrelationSWI showed equal capacities for differentiating non-sRI from CG.
Texture analysis based on DWI, BOLD, and SWI can assist in assessing renal dysfunction, and texture features based on BOLD and SWI may be suitable for assessing renal dysfunction during early stages.
• Texture analysis based on MRI techniques allowed for assessing renal dysfunction.
• Texture features based on BOLD and SWI, but not DWI, may be suitable for assessing renal function impairment during early stages.
• SWI exhibited a similar capacity to BOLD for assessing renal dysfunction.
KeywordsDiffusion magnetic resonance imaging Chronic kidney disease Chronic renal insufficiency Image processing, computer-assisted
Apparent diffusion coefficient
Acute kidney injury
Area under the receiver operating characteristic curve
Blood oxygen level–dependent MRI
Concordance correlation coefficient
Chronic kidney disease
Estimated glomerular filtration rate
Gray-level co-occurrence matrix
Median absolute deviation
Magnetic resonance imaging
Non-severe renal function impairment
Receiver operating characteristic curve
Severe renal function impairment
This work was supported by the National Natural Science Foundation of China (81771798, 81771805); Jiangsu Provincial Medical Youth Talent Program, China (QNRC2016299); General Program of Jiangsu Provincial Commission of Health and Family Planning, China (H2017003); Key Project of Health Commission of Changzhou, Jiangsu, China (ZD201509); Applied and Basic Research Program of Science and Technology Bureau of Changzhou, Jiangsu, China (CJ20160038); and Changzhou Municipal Medical Youth Talent Program, Jiangsu, China (QN201610).
Compliance with ethical standards
The scientific guarantor of this publication is Wei Xing.
Conflict of interest
The authors of this manuscript declare no relationships with any companies, whose products or services may be related to the subject matter of the article.
Statistics and biometry
Bin Xu kindly provided statistical advice for this manuscript.
Written informed consent was obtained from all subjects (patients) in this study.
This study was approved by Ethics Committee of Third Affiliated Hospital of Soochow University.
• cross-sectional study
• performed at one institution
- 2.Saran R, Li Y, Robinson B et al (2015) US renal data system 2014 annual data report: epidemiology of kidney disease in the United States. Am J Kidney Dis 66(Svii):S1–S305Google Scholar
- 11.Ortiz-Ramón R, Larroza A, Ruiz-España S, Arana E, Moratal D (2018) Classifying brain metastases by their primary site of origin using a radiomics approach based on texture analysis: a feasibility study. Eur Radiol. https://doi.org/10.1007/s00330-018-5463-6
- 12.Hocquelet A, Auriac T, Perier C et al (2018) Pre-treatment magnetic resonance-based texture features as potential imaging biomarkers for predicting event free survival in anal cancer treated by chemoradiotherapy. Eur Radiol. https://doi.org/10.1007/s00330-017-5284-z
- 13.Naganawa S, Enooku K, Tateishi R et al (2018) Imaging prediction of nonalcoholic steatohepatitis using computed tomography texture analysis. Eur Radiol. https://doi.org/10.1007/s00330-017-5270-5
- 14.American Diabetes Association (2013) Diagnosis and classification of diabetes mellitus. Diabetes Care 36(Suppl 1):S67–S74Google Scholar
- 15.Zhou HY, Chen TW, Zhang XM (2016) Functional magnetic resonance imaging in acute kidney injury: present status. Biomed Res Int 2016:2027370Google Scholar
- 28.Chen WB, Liang L, Zhang B et al (2015) To evaluate the damage of renal function in CIAKI rats at 3T: using ASL and BOLD MRI. Biomed Res Int 2015:593060Google Scholar