الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 186 |  |  |
| | | from | 186 | | |
Abstract
Hospital-Acquired Infections (HAIs), or healthcare associated infections, encompass almost all clinically evident infections that do not originate from a patients original admitting diagnosis. Most infections that become clinically evident after 48 hours of hospitalization are considered hospital-acquired. HAIs represent a frequent nonspecific clinical problem with potential consequences for morbidity and mortality. The highest rates of HAIs are observed in intensive care units (ICUs), which are also the units in which the most severely ill patients are treated and in which the highest mortality rates are observed. ICU patients are both at risk of acquiring HAIs and at risk of dying. A non-traditional, but increasingly popular quantitative analysis tool is a neural network. Neural networks are software-based mathematical tools that use a process similar to the human brain to predict, classify, and find patterns in data. Neural networks appeared in the later version of statistical software (SPP 16 and more). The Neural networks are constructed in layers of nodes. Nodes are components of a network that aid in the passage of information between layers. It has little or even no assumptions to conduct compared to traditional statistical methods which have a strict assumption and interpretation. The aims of the study were:
1. To determine the association between the HAIs and the identified risk factors using ANN.
2. To compare the detected relative importance of different risk factors by using different measurement scale for the dependent variable (number of infections or infection status) in ANN.
3. To identify the different risk factors of the HAIs using the different traditional statistical models according to the type of the dependent variable.
4. To evaluate the results obtained from ANN in comparison to that obtained from traditional multivariate analysis using different measurement scales.
5. To compare the utilization of ANN with other traditional statistical methods in:
6. Monitoring the HAIs rate and identifying the rate outbreaks.
7. Identifying seasonality and predicting HAI rate.