Employing machine learning to assess the accuracy of near-infrared spectroscopy of spent dialysate fluid in monitoring the blood concentrations of uremic toxins


  • Jasna B. Trbojević-Stanković 1. University of Belgrade, Faculty of Medicine, Doktora Subotića 8, 11000 Belgrade; 2. Clinical Hospital Center “Dr Dragiša Mišović – Dedinje”, Clinic of Urology, Department of Hemodialysis, Heroja Milana Tepića 1, 11000 Belgrade, Serbia https://orcid.org/0000-0002-1940-6188
  • Valentina D. Matović University of Belgrade, Faculty of Mechanical EngineerChalmers University of Technology, Departments of Industrial and Material Science, SE-41296, Gothenburg, Swedening, Department of Biomedical Engineering
  • Branislava D. Jeftić University of Belgrade, Faculty of Mechanical Engineering, Department of Biomedical Engineering, Kraljice Marije 16, 11000 Belgrade, Serbia https://orcid.org/0000-0002-8987-303X
  • Dejan Nešić University of Belgrade, Faculty of Medicine, Institute of Medical Physiology, Doktora Subotića 8, 11000 Belgrade, Serbia https://orcid.org/0000-0002-7794-3520
  • Jadranka V. Odović University of Belgrade, Faculty of Pharmacy, Department of Analytical Chemistry, Vojvode Stepe 450, 11221 Belgrade, Serbia
  • Iva Perović-Blagojević Clinical Hospital Center “Doktor Dragiša Mišović – Dedinje”, Department of Laboratory Diagnostics, Heroja Milana Tepića 1, 11000 Belgrade, Serbia
  • Nikola Topalović University of Belgrade, Faculty of Medicine, Institute of Medical Physiology, Doktora Subotića 8, 11000 Belgrade, Serbia
  • Lidija R. Matija University of Belgrade, Faculty of Mechanical Engineering, Department of Biomedical Engineering, Kraljice Marije 16, 11000 Belgrade, Serbia https://orcid.org/0000-0001-8492-7177




hemodialysis, machine learning, near-infrared spectroscopy, urea, creatinine


Paper description:

  • Near-infrared (NIR) absorption monitoring of the removal of nitrogenous compounds as a reflection of the efficacy of hemodialysis is unexplored.
  • We performed NIR spectroscopy of used dialysis fluid and employed machine learning to assess the correlations between obtained spectra and the levels of uremic toxins in patient blood.
  • Correlations between the NIR-absorbance spectra of the spent dialysate fluid with serum urea and uric acid were very good, and the correlation with serum creatinine was excellent.
  • NIR spectroscopy is an accurate, non-invasive, repetitive diagnostic screening that can help clinicians assess the efficiency and individualize hemodialysis treatments in real time.

Abstract: Hemodialysis (HD) removes nitrogenous waste products from patients’ blood through a semipermeable membrane along a concentration gradient. Near-infrared spectroscopy (NIRS) is an underexplored method of monitoring the concentrations of several molecules that reflect the efficacy of the HD process in dialysate samples. In this study, we aimed to evaluate NIRS as a technique for the non-invasive detection of uremic solutes by assessing the correlations between the spectrum of the spent dialysate and the serum levels of urea, creatinine, and uric acid. Blood and dialysate samples were taken from 35 patients on maintenance HD. The absorption spectrum of each dialysate sample was measured three times in the wavelength range of 700-1700 nm, resulting in a dataset with 315 spectra. The artificial neural network (ANN) learning technique was used to assess the correlations between the recorded NIR-absorbance spectra of the spent dialysate and serum levels of selected uremic toxins. Very good correlations between the NIR-absorbance spectra of the spent dialysate fluid with serum urea (R=0.91) and uric acid (R=0.91) and an excellent correlation with serum creatinine (R=0.97) were obtained. These results support the application of NIRS as a non-invasive, safe, accurate, and repetitive technique for online monitoring of uremic toxins to assist clinicians in assessing HD efficiency and individualization of HD treatments.


Download data is not yet available.


Rees L. Assessment of dialysis adequacy: beyond urea kinetic measurements. Pediatr Nephrol. 2019;34(1):61-9. https://doi.org/10.1007/s00467-018-3914-6

Vanholder R, Van Biesen W, Lameire N. A swan song for Kt/Vurea. Semin Dial. 2019;32(5):424-37. https://doi.org/10.1111/sdi.12811

National Kidney Foundation. KDOQI Clinical Practice Guideline for Hemodialysis Adequacy: 2015 update. Am J Kidney Dis. 2015;66(5):884-930. https://doi.org/10.1053/j.ajkd.2015.07.015

Vanholder R, Glorieux G, Eloot S. Once upon a time in dialysis: the last days of Kt/V?. Kidney Int. 2015;88(3):460-5. https://doi.org/10.1038/ki.2015.155

Pattharanitima P, Chauhan K, El Shamy O, Chaudhary K, Sharma S, Goca SG, Nadkarni GN, Uribarri J, Chan L. The association of standard Kt/V and surface area-normalized standard Kt/V with clinical outcomes in hemodialysis patients. Hemodial Int. 2020;24(4):495-505. https://doi.org/10.1111/hdi.12865

Lacson E Jr, Meyer K. B. Routine Monthly Blood Draws in Hemodialysis: Where Is the Evidence? Am J Kidney Dis. 2020;75(4):465-7. https://doi.org/10.1053/j.ajkd.2019.11.009

Eknoyan G, Beck GJ, Cheung AK, Daugirdas JT, Greene T, Kusek JW, Allon M, Bailey J, Delmez JA, Depner TA, Dwyer JT, Levey AS, Levin NW, Milford E, Ornt DB, Rocco MV, Schulman G, Schwab SJ, Teehan B. P, Toto R; Hemodialysis (HEMO) Study Group. Effect of dialysis dose and membrane flux in maintenance hemodialysis. N Engl J Med. 2002;347:2010-19.

Petitclerc T, Ridel C. Routine online assessment of dialysis dose: Ionic dialysance or UV-absorbance monitoring? Semin Dial. 2021;34(2):116-22. https://doi.org/10.1111/sdi.12949

Zhang L, Liu W, Hao C, He Y, Tao Y, Sun S, Jakob M, Marcelli D, Barth C, Chen X. Ensuring hemodialysis adequacy by dialysis dose monitoring with UV spectroscopy analysis of spent dialysate. Int J Artif Organs. 2022;45(4):351-9. https://doi.org/10.1177/03913988211059841

Aslam S, Saggi SJ, Salifu M, Kossmann RJ. Online measurement of hemodialysis adequacy using effective ionic dialysance of sodium-a review of its principles, applications, benefits, and risks. Hemodial Int. 2018;22(4):425-34. https://doi.org/10.1111/hdi.12623

Ross EA, Paugh-Miller JL, Nappo RW. Interventions to improve hemodialysis adequacy: protocols based on real-time monitoring of dialysate solute clearance. Clin Kidney J. 2018;11(3):394-9. https://doi.org/10.1093/ckj/sfx110

Eddy CV, Flanigan M, Arnold MA. Near-infrared spectroscopic measurement of urea in dialysate samples collected during hemodialysis treatments. Appl Spectrosc. 2003;57(10):1230-5. https://doi.org/10.1366/000370203769699081

Cho DS, Olesberg JT, Flanigan MJ, Arnold MA. On-line near-infrared spectrometer to monitor urea removal in real time during hemodialysis. Appl Spectrosc. 2008;62(8):866-72. https://doi.org/10.1366/000370208785284411

Jayatilake SMDAC, Ganegoda GU. Involvement of Machine Learning Tools in Healthcare Decision Making. J Healthc Eng. 2021:6679512. https://doi.org/10.1155/2021/6679512

Ngiam KY, Khor IW. Big data and machine learning algorithms for health-care delivery. Lancet Oncol. 2019;20(5):e262-73. https://doi.org/10.1016/S1470-2045(19)30149-4

Hastie T, Tibshirani R, Friedman JH, Friedman JH. The elements of statistical learning: data mining, inference, and prediction. Vol. 2. New York: Springer; 2009. 758 p.

Bilski J, Kowalczyk B, Marchlewska A, Zurada JM. Local Levenberg-Marquardt algorithm for learning feedforwad neural networks. J Artif Intell Soft. 2020;10(4):299-316. https://doi.org/10.2478/jaiscr-2020-0020

Shoaib M, Raja MAZ, Jamshed W, Nisar KS, Khan I, Farhat I. Intelligent computing Levenberg Marquardt approach for entropy optimized single-phase comparative study of second grade nanofluidic system. Int Commun Heat Mass . 2021;127(1):105544.

Galuzio PP, Cherif A. Recent Advances and Future Perspectives in the Use of Machine Learning and Mathematical Models in Nephrology. Adv Chronic Kidney Dis. 2022;29(5):472-9. https://doi.org/10.1053/j.ackd.2022.07.002

Du A, Shi X, Guo X, Pei Q, Ding Y, Zhou W, Lu Q, Shi H. Assessing the Adequacy of Hemodialysis Patients via the Graph-Based Takagi-Sugeno-Kang Fuzzy System. Comput Math Methods Med. 2021;2021:9036322. https://doi.org/10.1155/2021/9036322

Matović V, Jeftić B, Trbojević-Stanković J, Matija L. Predicting anemia using NIR spectrum of spent dialysis fluid in hemodialysis patients. Sci Rep. 2021;11(1):10549. https://doi.org/10.1038/s41598-021-88821-4

Matović V, Trbojević-Stanković J, Matija L, Sarac D, Vasić-Milovanović A, Petrović A. Predicting Hyperglycemia Using NIR Spectrum of Spent Fluid in Hemodialysis Patients. J Appl Spectrosc. 2021;88:662–7. https://doi.org/10.1007/s10812-021-01222-3

Matović V, Trbojević-Stanković JB, Jeftić BD, Matija LR. Glucose concentration monitoring using a near-infrared spectrum of spent dialysis fluid in hemodialysis patients. Srp Arh Celok Lek. 2020;148(11-12):706-10. https://doi.org/10.2298/SARH200215090M

Sandys V, Sexton D, O’Seaghdha C. Artificial intelligence and digital health for volume maintenance in hemodialysis patients. Hemodial Int. 2022;26(4):480-95. https://doi.org/10.1111/hdi.13033

Yang CH, Chen YS, Moi SH, Chen JB, Wang L, Chuang LY. Machine learning approaches for the mortality risk assessment of patients undergoing hemodialysis. Ther Adv Chronic Dis. 2022;13:20406223221119617. https://doi.org/10.1177/20406223221119617

Zhang W, Kasun LC, Wang QJ, Zheng Y, Lin Z. A Review of Machine Learning for Near-Infrared Spectroscopy. Sensors (Basel). 2022;22(24):9764. https://doi.org/10.3390/s22249764

Hall JW, Pollard A. Near-infrared spectroscopic determination of serum total proteins, albumin, globulins, and urea. Clin Biochem. 1993;26(6):483-90. https://doi.org/10.1016/0009-9120(93)80013-k

Berger AJ, Koo TW, Itzkan I, Horowitz G, Feld MS. Multicomponent blood analysis by near-infrared Raman spectroscopy. Appl Opt. 1999;38(13):2916-26. https://doi.org/10.1364/ao.38.002916

Shaw RA, Kotowich S, Mantsch HH, Leroux M. Quantitation of protein, creatinine, and urea in urine by near-infrared spectroscopy. Clin Biochem. 1996;29(1):11-9. https://doi.org/10.1016/0009-9120(95)02011-x

Eddy CV, Arnold MA. Near-infrared spectroscopy for measuring urea in hemodialysis fluids. Clin Chem. 2001;47(7):1279-86.

Shaw RA, Kotowich S, Leroux M, Mantsch HH. Multianalyte serum analysis using mid-infrared spectroscopy. Ann Clin Biochem. 1998;35(Pt 5):624-32. https://doi.org/10.1177/000456329803500505

Pezzaniti JL, Jeng TW, McDowell L, Oosta GM. Preliminary investigation of near-infrared spectroscopic measurements of urea, creatinine, glucose, protein, and ketone in urine. Clin Biochem. 2001;34(3):239-46. https://doi.org/10.1016/s0009-9120(01)00198-9

Alzubaidi L, Zhang J, Humaidi AJ, Al-Dujaili A, Duan Y, Al-Shamma O, Santamaria J, Fadhel MA, Al-Amidie M, Farhan L. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. J Big Data. 2021;8(1):53. https://doi.org/10.1186/s40537-021-00444-8

Goodfellow I, McDaniel P, Papernot N. Making machine learning robust against adversarial inputs. Commun ACM. 2018;61(7):57-66.

Amann J, Blasimme A, Vayena E, Frey D, Madai VI. Precise4Q consortium. Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak. 2020;20(1):310. https://doi.org/10.1186/s12911-020-01332-6

Jang EC, Park YM, Han HW, Lee CS, Kang ES, Lee YH, Nam SM. Machine-learning enhancement of urine dipstick tests for chronic kidney disease detection. J Am Med Inform Assoc. 2023;30(6):1114-24. https://doi.org/10.1093/jamia/ocad051

Lee H, Moon SJ, Kim SW, Min JW, Park HS, Yoon HE, Kim YS, Yang CW, Chung S, Koh ES, Chung BH. Prediction of intradialytic hypotension using pre-dialysis features - a deep learning-based artificial intelligence model. Nephrol Dial Transplant. 2023;gfad064. https://doi.org/10.1093/ndt/gfad064

Li X, Wu R, Zhao W, Shi R, Zhu Y, Wang Z, Pan H, Wang D. Machine learning algorithm to predict mortality in critically ill patients with sepsis-associated acute kidney injury. Sci Rep. 2023;13(1):5223. https://doi.org/10.1038/s41598-023-32160-z

Ravindhran B, Chandak P, Schafer N, Kundalia K, Hwang W, Antoniadis S, Haroon U, Zakri RH. Machine learning models in predicting graft survival in kidney transplantation: meta-analysis. BJS Open. 2023;7(2):zrad011. https://doi.org/10.1093/bjsopen/zrad011

Yoo KD, Noh J, Bae W, An JN, Oh HJ, Rhee H, Seong EY, Baek SH, Ahn SY, Cho J-H, Kim DK, Ryu D-R, Kim S, Lim CS, Lee JP; Korean Association for the Study of Renal Anemia and Artificial Intelligence (KARAI). Predicting outcomes of continuous renal replacement therapy using body composition monitoring: a deep-learning approach. Sci Rep. 2023;13(1):4605. https://doi.org/10.1038/s41598-023-30074-4

Du YC, Stephanus A. Levenberg-Marquardt Neural Network Algorithm for Degree of Arteriovenous Fistula Stenosis Classification Using a Dual Optical Photoplethysmography Sensor. Sensors (Basel). 2018;18(7):2322. https://doi.org/10.3390/s18072322

Henn R, Kirchler CG, Schirmeister ZL, Roth A, Mäntele W, Huck CW. Hemodialysis monitoring using mid- and near-infrared spectroscopy with partial least squares regression. J Biophotonics. 2018;11(7):e201700365. https://doi.org/10.1002/jbio.201700365

Canaud B. Recent advances in dialysis membranes. Curr Opin Nephrol Hypertens. 2021;30(6):613-22. https://doi.org/10.1097/MNH.0000000000000744

Burton JO, Corbett RW, Kalra PA, Vas P, Yiu V, Chrysochou C, Kirmizis D. Recent advances in treatment of haemodialysis. J R Soc Med. 2021;114(1):30-7. https://doi.org/10.1177/0141076820972669




How to Cite

Trbojević-Stanković JB, Matović VD, Jeftić BD, Nešić D, Odović JV, Perović-Blagojević I, Topalović N, Matija LR. Employing machine learning to assess the accuracy of near-infrared spectroscopy of spent dialysate fluid in monitoring the blood concentrations of uremic toxins. Arch Biol Sci [Internet]. 2023Oct.26 [cited 2024Feb.29];75(3):309-17. Available from: https://www.serbiosoc.org.rs/arch/index.php/abs/article/view/8683




Most read articles by the same author(s)