Document Detail


Prediction of trauma mortality using a neural network.
MedLine Citation:
PMID:  9036899     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
A neural network is a computerized construct consisting of input neurons (which process input data) connected to hidden neurons (to mathematically manipulate values they receive from all the input neurons) connected to output neurons (to output a prediction). Neural networks are created and trained via multiple iterations over data with known results. In 1993, 897 trauma patients were either declared dead in the emergency room (ER; 76 cases), admitted to the intensive care unit (427 cases, 36 deaths), or taken directly to the operating room (394 cases, 29 deaths). Using only data available from the ER, a neural network was created, and 628 cases were randomly selected for training. After 268 iterations, the network was trained to correctly predict death or survival in all 628 cases. This trained network was then tested on the other 269 cases without our providing the death or survival result. Its overall accuracy was 91 per cent (244 of 269 cases). It was able to predict correctly 60 per cent (12 of 20 cases) of the postoperative or post-intensive care unit admission deaths and 90 per cent (26 of 29 cases) of the deaths in the ER. Computerized neural networks can accurately predict a trauma patient's fate based on inital ER presentation. The theory and use of neural networks in predicting clinical outcome will be presented.
Authors:
S D Izenberg; M D Williams; A Luterman
Related Documents :
14568779 - Sexual asphyxia by hanging--a case report and a review of the literature.
7208759 - Expectation of death: case reports with implications for medical and psychiatric care.
16515879 - Methamphetamine-related sudden death with a concentration which was of a 'toxic level'.
12935649 - Immunohistochemical study of myoglobin and oxidative injury-related markers in the kidn...
17456099 - Traumatic asphyxia in weightlifters.
21029529 - Decrease in shigellosis-related deaths without shigella spp.-specific interventions, asia.
21743369 - Lipoblastomatosis: case report and review of the literature.
12708139 - Work-related reactive airways dysfunction syndrome cases from surveillance in selected ...
2231569 - Amniography for confirming the diagnosis of monoamniotic twinning. a case report.
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  The American surgeon     Volume:  63     ISSN:  0003-1348     ISO Abbreviation:  Am Surg     Publication Date:  1997 Mar 
Date Detail:
Created Date:  1997-03-07     Completed Date:  1997-03-07     Revised Date:  2004-11-17    
Medline Journal Info:
Nlm Unique ID:  0370522     Medline TA:  Am Surg     Country:  UNITED STATES    
Other Details:
Languages:  eng     Pagination:  275-81     Citation Subset:  IM    
Affiliation:
Department of Surgery, University of South Alabama Medical Center, Mobile, USA.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Age Factors
Artificial Intelligence
Female
Hospital Mortality
Humans
Injury Severity Score
Length of Stay
Male
Neural Networks (Computer)*
Prognosis
Wounds and Injuries / classification,  mortality*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Limb ischemia: surgical therapy in acute arterial occlusion.
Next Document:  Blunt pancreatic trauma: experience at a rural referral center.