Evaluation of an artificial neural network to predict urea nitrogen appearance for critically III multiple-trauma patients

Roland Dickerson, Darius L. Mason, Martin Croce, Gayle Minard, Rex Brown

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Background: Computer-based simulated biologic neural network models have made significant strides in clinical medicine. Methods: To determine the predictive performance of a conventional regression model and an artificial neural network for estimating urea nitrogen appearance (UNA) during critical illness, 125 adult patients admitted to the trauma intensive care unit who required specialized nutrition support were studied. The first 100 consecutive patients were used to develop the 2 models. The first model used stepwise multivariate regression analysis. The second model entailed the use of a feeding-forward, back-propagation, supervised neural network. Bias and precision of both methods were evaluated in 25 separate patients. Results: Multivariate regression analysis revealed a significant highly correlative relationship (r2 = .918, p ≤ .01): Predicted UNA (g/d) = (0.29 × WT) + (1.20 × WBC) + (0.44 × SUN) with WT as current body weight in kg, WBC as white blood cell count in cells/mm3, and SUN as serum urea nitrogen concentration (mg/dL). The regression method was biased toward overestimating measured UNA, whereas the neural network was unbiased. Precision (95% confidence interval) of the neural network was significantly better than the regression (3.3-7.2 g vs 7.3-11.6 g, respectively, p < .01). Regression analysis successfully predicted UNA within 3 g of measured UNA in 16% (4 of 25) of patients, whereas the neural network successfully predicted UNA in 44% (11 out of 25) of patients (p < .06). Conclusions: These preliminary data indicate that use of an artificial neural network may be superior to conventional regression modeling techniques for estimating UNA in critically ill adult multiple-trauma patients receiving specialized nutrition support.

Original languageEnglish (US)
Pages (from-to)429-435
Number of pages7
JournalJournal of Parenteral and Enteral Nutrition
Volume29
Issue number6
DOIs
StatePublished - Nov 1 2005

Fingerprint

Multiple Trauma
Urea
Nitrogen
Neural Networks (Computer)
Regression Analysis
Critical Illness
Multivariate Analysis
Clinical Medicine
Leukocyte Count
Intensive Care Units
Body Weight
Confidence Intervals
Wounds and Injuries
Serum

All Science Journal Classification (ASJC) codes

  • Medicine (miscellaneous)
  • Nutrition and Dietetics

Cite this

Evaluation of an artificial neural network to predict urea nitrogen appearance for critically III multiple-trauma patients. / Dickerson, Roland; Mason, Darius L.; Croce, Martin; Minard, Gayle; Brown, Rex.

In: Journal of Parenteral and Enteral Nutrition, Vol. 29, No. 6, 01.11.2005, p. 429-435.

Research output: Contribution to journalArticle

@article{052d6cba79114aab97ae88524019fcf0,
title = "Evaluation of an artificial neural network to predict urea nitrogen appearance for critically III multiple-trauma patients",
abstract = "Background: Computer-based simulated biologic neural network models have made significant strides in clinical medicine. Methods: To determine the predictive performance of a conventional regression model and an artificial neural network for estimating urea nitrogen appearance (UNA) during critical illness, 125 adult patients admitted to the trauma intensive care unit who required specialized nutrition support were studied. The first 100 consecutive patients were used to develop the 2 models. The first model used stepwise multivariate regression analysis. The second model entailed the use of a feeding-forward, back-propagation, supervised neural network. Bias and precision of both methods were evaluated in 25 separate patients. Results: Multivariate regression analysis revealed a significant highly correlative relationship (r2 = .918, p ≤ .01): Predicted UNA (g/d) = (0.29 × WT) + (1.20 × WBC) + (0.44 × SUN) with WT as current body weight in kg, WBC as white blood cell count in cells/mm3, and SUN as serum urea nitrogen concentration (mg/dL). The regression method was biased toward overestimating measured UNA, whereas the neural network was unbiased. Precision (95{\%} confidence interval) of the neural network was significantly better than the regression (3.3-7.2 g vs 7.3-11.6 g, respectively, p < .01). Regression analysis successfully predicted UNA within 3 g of measured UNA in 16{\%} (4 of 25) of patients, whereas the neural network successfully predicted UNA in 44{\%} (11 out of 25) of patients (p < .06). Conclusions: These preliminary data indicate that use of an artificial neural network may be superior to conventional regression modeling techniques for estimating UNA in critically ill adult multiple-trauma patients receiving specialized nutrition support.",
author = "Roland Dickerson and Mason, {Darius L.} and Martin Croce and Gayle Minard and Rex Brown",
year = "2005",
month = "11",
day = "1",
doi = "10.1177/0148607105029006429",
language = "English (US)",
volume = "29",
pages = "429--435",
journal = "Journal of Parenteral and Enteral Nutrition",
issn = "0148-6071",
publisher = "SAGE Publications Inc.",
number = "6",

}

TY - JOUR

T1 - Evaluation of an artificial neural network to predict urea nitrogen appearance for critically III multiple-trauma patients

AU - Dickerson, Roland

AU - Mason, Darius L.

AU - Croce, Martin

AU - Minard, Gayle

AU - Brown, Rex

PY - 2005/11/1

Y1 - 2005/11/1

N2 - Background: Computer-based simulated biologic neural network models have made significant strides in clinical medicine. Methods: To determine the predictive performance of a conventional regression model and an artificial neural network for estimating urea nitrogen appearance (UNA) during critical illness, 125 adult patients admitted to the trauma intensive care unit who required specialized nutrition support were studied. The first 100 consecutive patients were used to develop the 2 models. The first model used stepwise multivariate regression analysis. The second model entailed the use of a feeding-forward, back-propagation, supervised neural network. Bias and precision of both methods were evaluated in 25 separate patients. Results: Multivariate regression analysis revealed a significant highly correlative relationship (r2 = .918, p ≤ .01): Predicted UNA (g/d) = (0.29 × WT) + (1.20 × WBC) + (0.44 × SUN) with WT as current body weight in kg, WBC as white blood cell count in cells/mm3, and SUN as serum urea nitrogen concentration (mg/dL). The regression method was biased toward overestimating measured UNA, whereas the neural network was unbiased. Precision (95% confidence interval) of the neural network was significantly better than the regression (3.3-7.2 g vs 7.3-11.6 g, respectively, p < .01). Regression analysis successfully predicted UNA within 3 g of measured UNA in 16% (4 of 25) of patients, whereas the neural network successfully predicted UNA in 44% (11 out of 25) of patients (p < .06). Conclusions: These preliminary data indicate that use of an artificial neural network may be superior to conventional regression modeling techniques for estimating UNA in critically ill adult multiple-trauma patients receiving specialized nutrition support.

AB - Background: Computer-based simulated biologic neural network models have made significant strides in clinical medicine. Methods: To determine the predictive performance of a conventional regression model and an artificial neural network for estimating urea nitrogen appearance (UNA) during critical illness, 125 adult patients admitted to the trauma intensive care unit who required specialized nutrition support were studied. The first 100 consecutive patients were used to develop the 2 models. The first model used stepwise multivariate regression analysis. The second model entailed the use of a feeding-forward, back-propagation, supervised neural network. Bias and precision of both methods were evaluated in 25 separate patients. Results: Multivariate regression analysis revealed a significant highly correlative relationship (r2 = .918, p ≤ .01): Predicted UNA (g/d) = (0.29 × WT) + (1.20 × WBC) + (0.44 × SUN) with WT as current body weight in kg, WBC as white blood cell count in cells/mm3, and SUN as serum urea nitrogen concentration (mg/dL). The regression method was biased toward overestimating measured UNA, whereas the neural network was unbiased. Precision (95% confidence interval) of the neural network was significantly better than the regression (3.3-7.2 g vs 7.3-11.6 g, respectively, p < .01). Regression analysis successfully predicted UNA within 3 g of measured UNA in 16% (4 of 25) of patients, whereas the neural network successfully predicted UNA in 44% (11 out of 25) of patients (p < .06). Conclusions: These preliminary data indicate that use of an artificial neural network may be superior to conventional regression modeling techniques for estimating UNA in critically ill adult multiple-trauma patients receiving specialized nutrition support.

UR - http://www.scopus.com/inward/record.url?scp=33645735163&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645735163&partnerID=8YFLogxK

U2 - 10.1177/0148607105029006429

DO - 10.1177/0148607105029006429

M3 - Article

VL - 29

SP - 429

EP - 435

JO - Journal of Parenteral and Enteral Nutrition

JF - Journal of Parenteral and Enteral Nutrition

SN - 0148-6071

IS - 6

ER -