Diagnosis of MRSA with neural networks and logistic regression approach

Antibiotic-resistant pathogens are increasingly prevalent in the hospitals and community. A timely and accurate diagnosis of the infection would greatly help physicians effectively treat patients. In this research we investigate the potential of using neural networks (NN) and logistic regression (LR...

Full description

Saved in:
Bibliographic Details
Published in:Health care management science 2000-09, Vol.3 (4), p.287-297
Main Authors: Shang, J S, Lin, Y S, Goetz, A M
Format: Article
Language:English
Subjects:
Antibiotics
Datasets
Decision Support Techniques
Diagnosis, Computer-Assisted - methods
Drug resistance
Hospitals
Humans
Infections
Laboratories
Logistic Models
Long term health care
Medical diagnosis
Methicillin Resistance
Missing data
Multivariate analysis
Neural networks
Neural Networks (Computer)
Patients
Penicillin
Pennsylvania
Peritoneal dialysis
Physicians
Regression analysis
ROC Curve
Staphylococcal Infections - diagnosis
Staphylococcal Infections - economics
Staphylococcal Infections - microbiology
Staphylococcus aureus
Staphylococcus infections
Studies
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Antibiotic-resistant pathogens are increasingly prevalent in the hospitals and community. A timely and accurate diagnosis of the infection would greatly help physicians effectively treat patients. In this research we investigate the potential of using neural networks (NN) and logistic regression (LR) approach in diagnosing methicillin-resistant Staphylococcus aureus (MRSA). Receiver-Operating Characteristic (ROC) curve and the cross-validation method are used to compare the performances of both systems. We found that NN is better than the logistic regression approach, in terms of both the discriminatory power and the robustness. With modeling flexibility inherent in its techniques, NN is effective in dealing with MRSA and other classification problems involving large numbers of variables and interaction complexity. On the other hand, logistic regression in our case is slightly inferior, offers more clarity and less perplexity. It could be a method of choice when fewer variables are involved and/or justification of the results is desired.
ISSN:1386-9620
1572-9389