Prediction of Rib Fracture Injury Outcome by an Artificial Neural Network

Abstract

Outcome-based therapy is becoming the standard for assessing patient care efficacy. This study examines the ability of an artificial neural network to predict rib fracture injury outcome based on 20 intake variables determined within 1 hour of admission. The data base contained 580 patient records with four outcome variables: Length of hospital stay (LOS), ICU days, Lived, and Died. A 522-patient training set and a 58-patient test set were randomly selected. Nine networks were set up in a feed-forward, back-propagating design with each trained under different initial conditions. These networks predicted the test set outcome variables with an accuracy as high as 98% at the 80% testing level. Internal weight matrix examination indicated that age, ventilatory support, and high trauma scores were strongly associated with both ICU days and mortality. Being female, injury severity, and injury type were associated with increased LOS. Smoking and rib fracture number were low-level predictors of the four outcome variables.

Smoking as a risk factor for accident death: a meta-analysis of cohort studies

“…Perhaps more importantly, the smoking of tobacco and/or nicotine, and/or the withdrawals that result, cause physiologic performance decrements (Hirsch et al, 1985;Morton and Holmik, 1985;Perkins et al, 1989;Sandvik et al, 1995) that may reduce the smoker's ability to avoid or survive accidents. Smoking is associated with impaired performance on 'tests of muscle strength, agility and coordination, gait and balance, and self-reported functional status… after adjusting for age, … activity, and alcohol use (PB 0.05)' Smoking may increase the risk of serious complications (Warner et al, 1989;Forrest et al, 1992;Underwood and Bailey, 1993;Griepp et al, 1996)) and death (Dombi et al 1995;Raff et al, 1996) from accidents that do occur. Smoking impairs the metabolic efficiency (Perkins et al, 1989) and fitness (Sandvik et al, 1995) that may be needed to survive the acute physiologic stress of accidents.…”

The percentage of bacterial genes on leading versus lagging strands is influenced by multiple balancing forces

“…Out of the initial set of 57 input variables, we have conducted a variable selection process based on the idea of mean impact value (MIV) (29). Based on the ranks of the MIVs of the input variables, input variables with insignificant MIV will be eliminated from the neural network model.…”

Prediction and Analysis of Tensile Properties of Austenitic Stainless Steel Using Artificial Neural Network

“…Using machine learning cannot only make accurate predictions, but also further analyze the effect of each single variable on the corresponding properties. The mean impact value (MIV) method has been widely used for quantitative feature analysis in the machine learning application to explore the relative importance of each input variable for the improvement of the prediction performance [45]. The algorithm process of MIV is as follow:…”