Top

Published in:

Open Access 01-12-2023 | Chronic Kidney Disease | Original Article

MRI texture-based machine learning models for the evaluation of renal function on different segmentations: a proof-of-concept study

Authors: Xiaokai Mo, Wenbo Chen, Simin Chen, Zhuozhi Chen, Yuanshu Guo, Yulian Chen, Xuewei Wu, Lu Zhang, Qiuying Chen, Zhe Jin, Minmin Li, Luyan Chen, Jingjing You, Zhiyuan Xiong, Bin Zhang, Shuixing Zhang

Published in: Insights into Imaging | Issue 1/2023

Abstract

Background

To develop and validate an MRI texture-based machine learning model for the noninvasive assessment of renal function.

Methods

A retrospective study of 174 diabetic patients (training cohort, n = 123; validation cohort, n = 51) who underwent renal MRI scans was included. They were assigned to normal function (n = 71), mild or moderate impairment (n = 69), and severe impairment groups (n = 34) according to renal function. Four methods of kidney segmentation on T2-weighted images (T2WI) were compared, including regions of interest covering all coronal slices (All-K), the largest coronal slices (LC-K), and subregions of the largest coronal slices (TLCO-K and PIZZA-K). The speeded-up robust features (SURF) and support vector machine (SVM) algorithms were used for texture feature extraction and model construction, respectively. Receiver operating characteristic (ROC) curve analysis was used to evaluate the diagnostic performance of models.

Results

The models based on LC-K and All-K achieved the nonsignificantly highest accuracy in the classification of renal function (all p values > 0.05). The optimal model yielded high performance in classifying the normal function, mild or moderate impairment, and severe impairment, with an area under the curve of 0.938 (95% confidence interval [CI] 0.935–0.940), 0.919 (95%CI 0.916–0.922), and 0.959 (95%CI 0.956–0.962) in the training cohorts, respectively, as well as 0.802 (95%CI 0.800–0.807), 0.852 (95%CI 0.846–0.857), and 0.863 (95%CI 0.857–0.887) in the validation cohorts, respectively.

Conclusion

We developed and internally validated an MRI-based machine-learning model that can accurately evaluate renal function. Once externally validated, this model has the potential to facilitate the monitoring of patients with impaired renal function.

Available only for authorised users

Collaboration GBDCKD (2020) Global, regional, and national burden of chronic kidney disease, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 395:709–733. https://doi.org/10.1016/S0140-6736(20)30045-3CrossRef

Webster AC, Nagler EV, Morton RL, Masson P (2017) Chronic kidney disease. Lancet 389:1238–1252. https://doi.org/10.1016/S0140-6736(16)32064-5CrossRef

Gigante A, Proietti M, Petrillo E, Mannucci PM, Nobili A, Muscaritoli M (2021) Renal function, cardiovascular diseases, appropriateness of drug prescription and outcomes in hospitalized older patients. Drugs Aging 38:1097–1105. https://doi.org/10.1007/s40266-021-00903-0CrossRef

Kawakami T, Mimura I, Shoji K, Tanaka T, Nangaku M (2014) Hypoxia and fibrosis in chronic kidney disease: crossing at pericytes. Kidney Int Suppl 4:107–112. https://doi.org/10.1038/kisup.2014.20CrossRef

Eckerbom P, Hansell P, Cox E et al (2019) Multiparametric assessment of renal physiology in healthy volunteers using noninvasive magnetic resonance imaging. Am J Physiol Renal Physiol 316:F693–F702. https://doi.org/10.1152/ajprenal.00486.2018CrossRef

Mao W, Zhou J, Zeng M et al (2018) Intravoxel incoherent motion diffusion-weighted imaging for the assessment of renal fibrosis of chronic kidney disease: a preliminary study. Magn Reson Imaging 47:118–124. https://doi.org/10.1016/j.mri.2017.12.010CrossRef

Inoue T, Kozawa E, Okada H et al (2011) Noninvasive evaluation of kidney hypoxia and fibrosis using magnetic resonance imaging. J Am Soc Nephrol 22:1429–1434. https://doi.org/10.1681/ASN.2010111143CrossRef

Khatir DS, Pedersen M, Jespersen B, Buus NH (2015) Evaluation of renal blood flow and oxygenation in CKD using magnetic resonance imaging. Am J Kidney Dis 66:402–411. https://doi.org/10.1053/j.ajkd.2014.11.022CrossRef

Mora-Gutierrez JM, Garcia-Fernandez N, Slon Roblero MF et al (2017) Arterial spin labeling MRI is able to detect early hemodynamic changes in diabetic nephropathy. J Magn Reson Imaging 46:1810–1817. https://doi.org/10.1002/jmri.25717CrossRef

10.

Wang YC, Feng Y, Lu CQ, Ju S (2018) Renal fat fraction and diffusion tensor imaging in patients with early-stage diabetic nephropathy. Eur Radiol 28:3326–3334. https://doi.org/10.1007/s00330-017-5298-6CrossRef

11.

Zhang H, Wang P, Shi D et al (2022) Capability of intravoxel incoherent motion and diffusion tensor imaging to detect early kidney injury in type 2 diabetes. Eur Radiol. https://doi.org/10.1007/s00330-021-08415-6CrossRef

12.

Lambin P, Rios-Velazquez E, Leijenaar R et al (2012) Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 48:441–446. https://doi.org/10.1016/j.ejca.2011.11.036CrossRef

13.

Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577. https://doi.org/10.1148/radiol.2015151169CrossRef

14.

Lubner MG, Pickhardt PJ (2018) Multidetector computed tomography for retrospective, noninvasive staging of liver fibrosis. Gastroenterol Clin North Am 47:569–584. https://doi.org/10.1016/j.gtc.2018.04.012CrossRef

15.

Zhang X, Gao X, Liu BJ et al (2015) Effective staging of fibrosis by the selected texture features of liver: which one is better, CT or MR imaging? Comput Med Imaging Graph 46(Pt 2):227–236. https://doi.org/10.1016/j.compmedimag.2015.09.003CrossRef

16.

Kuo CC, Chang CM, Liu KT et al (2019) Automation of the kidney function prediction and classification through ultrasound-based kidney imaging using deep learning. NPJ Digit Med 2:29. https://doi.org/10.1038/s41746-019-0104-2CrossRef

17.

Huang S, Shi K, Zhang Y et al (2022) Texture analysis of T2-weighted cardiovascular magnetic resonance imaging to discriminate between cardiac amyloidosis and hypertrophic cardiomyopathy. BMC Cardiovasc Disord 22:235. https://doi.org/10.1186/s12872-022-02671-0CrossRef

18.

Yu B, Huang C, Fan X et al (2022) Application of MR imaging features in differentiation of renal changes in patients with stage III type 2 diabetic nephropathy and normal subjects. Front Endocrinol (Lausanne) 13:846407. https://doi.org/10.3389/fendo.2022.846407CrossRef

19.

Grzywińska M, Jankowska M, Banach-Ambroziak E, Szurowska E, Dębska-Ślizień A (2020) Computation of the texture features on T2-weighted images as a novel method to assess the function of the transplanted kidney: primary research. Transplant Proc 52:2062–2066. https://doi.org/10.1016/j.transproceed.2020.02.084CrossRef

20.

Ding J, Xing Z, Jiang Z et al (2019) Evaluation of renal dysfunction using texture analysis based on DWI, BOLD, and susceptibility-weighted imaging. Eur Radiol 29:2293–2301. https://doi.org/10.1007/s00330-018-5911-3CrossRef

21.

Deng Y, Yang BR, Luo JW, Du GX, Luo LP (2020) DTI-based radiomics signature for the detection of early diabetic kidney damage. Abdom Radiol (NY) 45:2526–2531. https://doi.org/10.1007/s00261-020-02576-6CrossRef

22.

Berchtold L, Friedli I, Crowe LA et al (2020) Validation of the corticomedullary difference in magnetic resonance imaging-derived apparent diffusion coefficient for kidney fibrosis detection: a cross-sectional study. Nephrol Dial Transplant 35:937–945. https://doi.org/10.1093/ndt/gfy389CrossRef

23.

Lin HY, Lee YL, Lin KD et al (2017) Association of renal elasticity and renal function progression in patients with chronic kidney disease evaluated by real-time ultrasound elastography. Sci Rep 7:43303. https://doi.org/10.1038/srep43303CrossRef

24.

Chunmei X, Mei H, Yan Z, Haiying W (2019) Diagnostic method of liver cirrhosis based on MR image texture feature extraction and classification algorithm. J Med Syst 44:11. https://doi.org/10.1007/s10916-019-1508-xCrossRef

25.

Schawkat K, Ciritsis A, von Ulmenstein S et al (2020) Diagnostic accuracy of texture analysis and machine learning for quantification of liver fibrosis in MRI: correlation with MR elastography and histopathology. Eur Radiol 30:4675–4685. https://doi.org/10.1007/s00330-020-06831-8CrossRef

26.

Niles DJ, Artz NS, Djamali A, Sadowski EA, Grist TM, Fain SB (2016) Longitudinal assessment of renal perfusion and oxygenation in transplant donor-recipient pairs using arterial spin labeling and blood oxygen level-dependent magnetic resonance imaging. Invest Radiol 51:113–120. https://doi.org/10.1097/RLI.0000000000000210CrossRef

27.

Sigmund EE, Vivier PH, Sui D et al (2012) Intravoxel incoherent motion and diffusion-tensor imaging in renal tissue under hydration and furosemide flow challenges. Radiology 263:758–769. https://doi.org/10.1148/radiol.12111327CrossRef

28.

Milani B, Ansaloni A, Sousa-Guimaraes S et al (2017) Reduction of cortical oxygenation in chronic kidney disease: evidence obtained with a new analysis method of blood oxygenation level-dependent magnetic resonance imaging. Nephrol Dial Transplant Off Publ Eur Dial Transpl Assoc Eur Renal Assoc 32:2097–2105. https://doi.org/10.1093/ndt/gfw362CrossRef

29.

Pruijm M, Milani B, Pivin E et al (2018) Reduced cortical oxygenation predicts a progressive decline of renal function in patients with chronic kidney disease. Kidney Int 93:932–940. https://doi.org/10.1016/j.kint.2017.10.020CrossRef

30.

Chaddad A, Daniel P, Desrosiers C, Toews M, Abdulkarim B (2019) Novel radiomic features based on joint intensity matrices for predicting glioblastoma patient survival time. IEEE J Biomed Health Inform 23:795–804. https://doi.org/10.1109/JBHI.2018.2825027CrossRef

31.

Xie C, Yang P, Zhang X et al (2019) Sub-region based radiomics analysis for survival prediction in oesophageal tumours treated by definitive concurrent chemoradiotherapy. EBioMedicine 44:289–297. https://doi.org/10.1016/j.ebiom.2019.05.023CrossRef

32.

Xu H, Lv W, Feng H et al (2020) Subregional radiomics analysis of PET/CT imaging with intratumor partitioning: application to prognosis for nasopharyngeal carcinoma. Mol Imaging Biol 22:1414–1426. https://doi.org/10.1007/s11307-019-01439-xCrossRef

Title: MRI texture-based machine learning models for the evaluation of renal function on different segmentations: a proof-of-concept study
Authors: Xiaokai Mo
Wenbo Chen
Simin Chen
Zhuozhi Chen
Yuanshu Guo
Yulian Chen
Xuewei Wu
Lu Zhang
Qiuying Chen
Zhe Jin
Minmin Li
Luyan Chen
Jingjing You
Zhiyuan Xiong
Bin Zhang
Shuixing Zhang
Publication date: 01-12-2023
Publisher: Springer Vienna
Keyword: Chronic Kidney Disease
Published in: Insights into Imaging / Issue 1/2023
Electronic ISSN: 1869-4101
DOI: https://doi.org/10.1186/s13244-023-01370-4

Keynote webinar | Spotlight on medication adherence

Springer Medicine

MRI texture-based machine learning models for the evaluation of renal function on different segmentations: a proof-of-concept study

Abstract

Background

Methods

Results

Conclusion

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Background

Methods

Results

Conclusion

Please log in to get access to this content

Other articles of this Issue 1/2023

Infection prevention and control in the medical imaging environment: a scoping review

CT analysis of thoracolumbar body composition for estimating whole-body composition

A deep learning model using chest X-ray for identifying TB and NTM-LD patients: a cross-sectional study

Undergraduate radiology education in Europe in 2022: a survey from the European Society of Radiology (ESR)

American College of Radiology Appropriateness Criteria®: a bibliometric analysis of panel members

Core and penumbra estimation using deep learning-based AIF in association with clinical measures in computed tomography perfusion (CTP)