Jove
Visualize
Contact Us

Related Experiment Videos

Using artificial intelligence to predict the equilibrated postdialysis blood urea concentration.

E A Fernández1, R Valtuille, P Willshaw

  • 1Favaloro University, Buenos Aires, Argentina. elmerfer@favaloro.edu.ar

Blood Purification
|March 13, 2001
PubMed
Summary

Accurately estimating hemodialysis dose (Kt/V) is vital for patient outcomes. A new neural network model predicts post-dialysis blood urea (eqU) more effectively, improving hemodialysis prescription accuracy.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Ipsilateral proximal and shaft femoral fractures: Results of two treatment strategies.

Revista espanola de cirugia ortopedica y traumatologia·2023
Same author

Usefulness of the Nasal Allergen Provocation Test in the Diagnosis of Shellfish Allergy.

Journal of investigational allergology & clinical immunology·2021
Same author

Shellfish Allergy: Unmet Needs in Diagnosis and Treatment.

Journal of investigational allergology & clinical immunology·2020
Same author

A patient-specific method for the evaluation of wall shear stress in human coronary arteries.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2010
Same author

Investigation of uremic analytes in hemodialysate and their structural elucidation from accurate mass maps generated by a multi-dimensional liquid chromatography/mass spectrometry approach.

Rapid communications in mass spectrometry : RCM·2009
Same author

Partial least squares regression: a valuable method for modeling molecular behavior in hemodialysis.

Annals of biomedical engineering·2008
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Area of Science:

  • Nephrology
  • Biomedical Engineering
  • Artificial Intelligence

Background:

  • Hemodialysis dose (Kt/V) significantly impacts patient morbidity and mortality.
  • Urea rebound, the rise in blood urea post-dialysis, complicates accurate dose assessment.
  • Miscalculating equilibrated Kt/V leads to inadequate dialysis prescriptions and poor patient outcomes.

Purpose of the Study:

  • To develop and validate a supervised neural network for predicting equilibrated postdialysis blood urea (eqU) at 60 minutes.
  • To compare the neural network's performance against existing methods for eqU and equilibrated Kt/V estimation.
  • To enhance the accuracy of hemodialysis prescription by improving the estimation of true dialysis dose.

Main Methods:

  • A supervised neural network model was trained to predict eqU.

Related Experiment Videos

  • The model's predictions were compared with established methods like the Smye formula for eqU and Daugirdas formula for eqKt/V.
  • Performance was evaluated using mean difference error and mean percentage error for eqU and eqKt/V.

    • Main Results:

    • The neural network achieved a mean difference error of 0.22 +/- 7.71 mg/ml for eqU prediction.
    • The model demonstrated a mean difference error of -0.01 +/- 0.15 for eqKt/V.
    • The Smye formula showed significant dispersion in eqU estimation, deeming it less appropriate.
    • The Daugirdas double-pool formula for Kt/V estimation showed accuracy consistent with HEMO study results.

    Conclusions:

    • The proposed neural network offers a more accurate method for predicting equilibrated postdialysis blood urea (eqU).
    • This improved prediction can lead to more precise hemodialysis prescriptions and better patient management.
    • The neural network approach represents a novel and superior alternative to current methods for assessing true hemodialysis dose.