Estimation of glomerular filtration rate by a radial basis function neural network in patients with type-2 diabetes mellitus

BMC Nephrology

Tan-Qi Lou¹  Xiao-Ming Wu²  Ming Li¹  Lin-Sheng Lv³  Cheng Wang¹  Ning-shan Li⁴  Yan-Ru Chen¹  Xun Liu² 

[1] Division of Nephrology, Department of Internal Medicine, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China;College of Biology Engineering, South China University of Technology, Guangzhou, China;Operating Room, The Third Affiliated Hospital of Sun Yat-sen University, Guangzhou, China;Department of Radiation Oncology, Chengdu International Cancer Treatment Hospital, Chengdu, China

关键词: Artificial neural network;    Glomerular filtration rate;    Chronic kidney disease;    Type 2 diabetes;   

Others  :  1082858
DOI  :  10.1186/1471-2369-14-181

 received in 2012-10-08, accepted in 2013-08-12,  发布年份 2013

【 摘 要 】

Background

Accurate and precise estimates of glomerular filtration rate (GFR) are essential for clinical assessments, and many methods of estimation are available. We developed a radial basis function (RBF) network and assessed the performance of this method in the estimation of the GFRs of 207 patients with type-2 diabetes and CKD.

Methods

Standard GFR (sGFR) was determined by ^99mTc-DTPA renal dynamic imaging and GFR was also estimated by the 6-variable MDRD equation and the 4-variable MDRD equation.

Results

Bland-Altman analysis indicated that estimates from the RBF network were more precise than those from the other two methods for some groups of patients. However, the median difference of RBF network estimates from sGFR was greater than those from the other two estimates, indicating greater bias. For patients with stage I/II CKD, the median absolute difference of the RBF network estimate from sGFR was significantly lower, and the P₅₀ of the RBF network estimate (n = 56, 87.5%) was significantly higher than that of the MDRD-4 estimate (n = 49, 76.6%) (p < 0.0167), indicating that the RBF network estimate provided greater accuracy for these patients.

Conclusions

In patients with type-2 diabetes mellitus, estimation of GFR by our RBF network provided better precision and accuracy for some groups of patients than the estimation by the traditional MDRD equations. However, the RBF network estimates of GFR tended to have greater bias and higher than those indicated by sGFR determined by ^99mTc-DTPA renal dynamic imaging.

【 授权许可】

   
2013 Liu et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

【 参考文献 】

• [1]Barnett AH: Preventing renal complications in diabetic patients: the Diabetics Exposed to Telmisartan And enalaprIL (DETAIL) study. Acta Diabetol 2005, 42:S42-S49.
• [2]National Kidney Foundation: K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Am J Kidney Dis 2002, 39:S1-266.
• [3]American Diabetes Association: Standards of medical care in diabetes-2012. Diabetes Care 2012, 35:S11-S63.
• [4]Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D: A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999, 130:461-470.
• [5]Liu X, Wang C, Tang H: Assessing glomerular filtration rate (GFR) in elderly Chinese patients with chronic kidney disease (CKD): A comparison of various predictive equations. Arch Gerontol Geriatr 2010, 51:13-20.
• [6]Silveiro SP, Araújo GN, Ferreira MN, Souza FD, Yamaguchi HM, Camargo EG: Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation pronouncedly underestimates glomerular filtration rate in type 2 diabetes. Diabetes Care 2011, 34:2353-2355.
• [7]Fontseré N, Bonal J, Salinas I, de Arellano MR, Rios J, Torres F, Sanmartí A, Romero R: Is the new Mayo Clinic Quadratic equation useful for the estimation of glomerular filtration rate in type 2 diabetic patients? Diabetes Care 2008, 31:2265-2267.
• [8]Nair S, Mishra V, Hayden K, Lisboa PJ, Pandya B, Vinjamuri S, Hardy KJ, Wilding JP: The four-variable modification of diet in renal disease formula underestimates glomerular filtration rate in obese type 2 diabetic individuals with chronic kidney disease. Diabetologia 2011, 54:1304-1307.
• [9]Joshy G, Porter T, Le Lievre C, Lane J, Williams M, Lawrenson R: Implication of using estimated glomerular filtration rate (GFR) in a multi ethnic population of diabetes patients in general practice. N Z Med J 2010, 123:9-18.
• [10]Rigalleau V, Beauvieux MC, Le Moigne F, Lasseur C, Chauveau P, Raffaitin C, Perlemoine C, Barthe N, Combe C, Gin H: Cystatin C improves the diagnosis and stratification of chronic kidney disease, and the estimation of glomerular filtration rate in diabetes. Diabetes Metab 2008, 34:482-489.
• [11]Stevens LA, Claybon MA, Schmid CH, Chen J, Horio M, Imai E, Nelson RG, Van Deventer M, Wang HY, Zuo L, Zhang YL, Levey AS: Evaluation of the Chronic Kidney Disease Epidemiology Collaboration equation for estimating the glomerular filtration rate in multiple ethnicities. Kidney Int 2011, 79:555-562.
• [12]Zuo L, Ma YC, Zhou YH, Wang M, Xu GB, Wang HY: Application of GFR-estimating equations in Chinese patients with chronic kidney disease. Am J Kidney Dis 2005, 45:463-472.
• [13]Levey AS, Coresh J, Greene T, Stevens LA, Zhang YL, Hendriksen S, Kusek JW, Van Lente F: Using standardized serum creatinine values in the modification of diet in renal disease study equation for estimating glomerular filtration rate. Ann Intern Med 2006, 145:247-254.
• [14]Levey AS, Stevens LA, Schmid CH: CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration). A New Equation to Estimate Glomerular Filtration Rate. Ann Intern Med 2009, 150:604-612.
• [15]Goldfarb-Rumyantzev AS, Pappas L: Prediction of renal insufficiency in Pima Indians with nephropathy of type 2 diabetes mellitus. Am J Kidney Dis 2002, 40:252-264.
• [16]Baxt WG, Skora J: Prospective validation of artificial neural network trained to identify acute myocardial infarction. Lancet 1996, 347:12-15.
• [17]Dybowski R, Weller P, Chang R, Gant V: Prediction of outcome in critically ill patients using artificial neural network synthesised by genetic algorithm. Lancet 1996, 347:1146-1150.
• [18]Song Q, Kasabov N, Ma T, Marshall MR: Integrating regression formulas and kernel functions into locally adaptive knowledge-based neural networks: a case study on renal function evaluation. Artif Intell Med 2006, 36:235-244.
• [19]Marshall MR, Song Q, Ma TM, MacDonell SG, Kasabov NK: Evolving connectionist system versus algebraic formulas for prediction of renal function from serum creatinine. Kidney Int 2005, 67:1944-1954.
• [20]Liu X, Wu XM, Li LS, Lou TQ: Application of radial basis function neural network to estimate glomerular filtration rate in Chinese patients with chronic kidney disease. ICCASM 2010, 15:332-335. [The International Conference on Computer Application and System Modeling]
• [21]Du X, Liu L, Hu B, Wang F, Wan X, Jiang L, Zhang R, Cao C: Is the Chronic Kidney Disease Epidemiology Collaboration four-level race equation better than the cystatin C equation? Nephrology (Carlton) 2012, 17:407-414.
• [22]Liu X, Lv L, Wang C, Shi C, Cheng C, Tang H, Chen Z, Ye Z, Lou T: Comparison of prediction equations to estimate glomerular filtration rate in Chinese patients with chronic kidney disease. Intern Med J 2012, 42:e59-67.
• [23]Li Q, Zhang CL, Fu ZL, Wang RF, Ma YC, Zuo L: Development of formulae for accurate measurement of the glomerular filtration rate by renal dynamic imaging. Nucl Med Commun 2007, 28:407-413.
• [24]Assadi M, Eftekhari M, Hozhabrosadati M, Saghari M, Ebrahimi A, Nabipour I, Abbasi MZ, Moshtaghi D, Abbaszadeh M, Assadi S: Comparison of methods for determination of glomerular filtration rate: low and high-dose Tc-99m-DTPA renography, predicted creatinine clearance method, and plasma sample method. Int Urol Nephrol 2008, 40:1059-1065.
• [25]Blaufox MD, Aurell M, Bubeck B, Fommei E, Piepsz A, Russell C, Taylor A, Thomsen HS, Volterrani D: Report of the Radionuclides in Nephrourology Committee on renal clearance. J Nucl Med 1996, 37:1883-1890.
• [26]Stevens LA, Zhang Y, Schmid CH: Evaluating the performance of equations for estimating glomerular filtration rate. J Nephrol 2008, 21:797-807.
• [27]Bland JM, Altman DG: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986, 1:307-310.
• [28]Bevc S, Hojs R, Ekart R, Gorenjak M, Puklavec L: Simple cystatin C formula compared to sophisticated CKD-EPI formulas for estimation of glomerular filtration rate in the elderly. Ther Apher Dial 2011, 15:261-268.
• [29]Peake M, Whiting M: Measurement of serum creatinine– current status and future goals. Clin Biochem Rev 2006, 27:173-184.
• [30]Hallan S, Astor B, Lydersen S: Estimating glomerular filtration rate in the general population: the second Health Survey of Nord-Trondelag (HUNT II). Nephrol Dial Transplant 2006, 21:1525-1533.
• [31]Ma YC, Zuo L, Zhang CL, Wang M, Wang RF, Wang HY: Comparison of 99mTc-DTPA renal dynamic imaging with modified MDRD equation for glomerular filtration rate estimation in Chinese patients in different stages of chronic kidney disease. Nephrol Dial Transplant 2007, 22:417-23.
• [32]Xie P, Huang JM, Liu XM, Wu WJ, Pan LP, Lin HY: (99m)Tc-DTPA renal dynamic imaging method may be unsuitable to be used as the reference method in investigating the validity of CKD-EPI equation for determining glomerular filtration rate. PLoS One 2013, 8:e62328.
• [33]Rehling M, Møller ML, Thamdrup B, Lund JO, Trap-Jensen J: Reliability of a 99mTc-DTPA gamma camera technique for determination of single kidney glomerular filtration rate. A comparison to plasma clearance of 51Cr-EDTA in one-kidney patients, using the renal clearance of inulin as a reference. Scand J Urol Nephrol 1986, 20:57-62.
• [34]Yang W, Lu J, Weng J, Jia W, Ji L, Xiao J, Shan Z, Liu J, Tian H, Ji Q, Zhu D, Ge J, Lin L, Chen L, Guo X, Zhao Z, Li Q, Zhou Z, Shan G, He J, China National Diabetes and Metabolic Disorders Study Group: Prevalence of diabetes among men and women in China. N Engl J Med 2010, 362:1090-1101.