Neural Networks: What Are They?

Guerriere, Michael; Detsky, Allan S.

doi:10.7326/0003-4819-115-11-906

Cited by 46 publications

(18 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They work particularly well in situations where the rules relating input variables are not well defined and especially where there are nonlinear relationships between the variables (17). They work particularly well in situations where the rules relating input variables are not well defined and especially where there are nonlinear relationships between the variables (17).…”

Section: Methodology Artificial Neural Networkmentioning

confidence: 99%

Measuring Quality of Care in a Psychiatric Hospital Using Artificial Neural Networks

1997

View full text Add to dashboard Cite

This study investigates a new method of measuring quality of care. Taking place at a tertiary psychiatric hospital with 5,128 admissions from January 1989 through December 1995, this study uses artificial neural networks (ANNs) to predict hospital length-of-stay (LOS) and uses the standard deviation of LOS in a formula to measure quality of care, Q. ANNs are trained with data using unique patient identifiers and are compared with identical ANNs trained without these identifiers. These two types of ANNs make a LOS prediction, P, with a slightly different accuracies, and this fact is exploited in measuring Q. The authors defined U as the standard deviation of the difference between the actual and the predicted LOS of the ANNs with unique patient identifiers, and defined G as the standard deviation of the difference between the actual and the predicted LOS of the ANNs without using these unique identifiers. Dividing U, the variation of individual LOS patterns intertwined with systemic LOS patterns, by G, the variation of predominately systemic LOS patterns, yields the ratio U/G, in which systemic effects are factored out leaving a measure of the average severity of patient illness. Ratios that exceed unity are seen in the patients who are more severely ill. The formula for quality of care, Q, divides the best LOS prediction accuracy, P, which is inversely proportional to overall variation in the delivery system, by U/G, which is inversely proportional to quality of care, written as: Q = P/(U/G). Q reflects the patients' perspective because LOS is concrete and tangible to patients. The study took place during hospital downsizing (political change), a consent decree (policy change), new administrative and medical personnel (staffing change), and the introduction of clozapine and risperidone for schizophrenia (therapeutic change). These events had a predominantly positive impact on Q. The value of Q correlated well with the Joint Commission on Accreditation of Health Care Organizations (JCAHO) triennial evaluations. Some conclusions that emerged from this study: 1) System variation, reflected in the standard deviation of LOS, increased with frequent changes in top management. 2) There was a clear-cut beneficial effect of clozapine, and to a lesser extent of risperidone in schizophrenia, allowing more community placement. 3) With a dedicated professional staff quality of care can prevail despite increasing variation in LOS (systemic problems). 4) The number of hospital employees per Q unit halved when the overall hospital Q ranged from low to high values as a result of policy and staffing improvements, suggesting an increased efficiency of operation. 5) Q can be an objective outcome measure of quality care from the patients' perspective.

show abstract

Section: Methodology Artificial Neural Networkmentioning

confidence: 99%

Measuring Quality of Care in a Psychiatric Hospital Using Artificial Neural Networks

1997

View full text Add to dashboard Cite

show abstract

“…However, this apparent flexibility that has been often cited as an advantage [2] can be a source of problems associated with the estimation of misclassification probabilities [7]. ANN represent a relative bblack boxQ in comparison to other statistical techniques [1] since they do not allow to easily determine which variables contribute mostly to a particular output and therefore may be much interesting when we are not looking for a single but rather for a complex of predictors. Although regression-like techniques have been used to examine the connection weights of various input variables and to determine which variables can be removed from a model without affecting its performance [33], none of these techniques offer the ease of interpretation of more common parametric models like, for instance, the odds ratios associated with the coefficients of a logistic regression model.…”

Section: Comparison With Previous Studiesmentioning

confidence: 98%

“…Artificial neural networks (ANN) are distributed networks of computing elements capable of identifying relations in input data that are not apparent with common analytic techniques [1]. bKnowledgeQ is acquired by ANN through a learning or training process in which connection strengths, known as weights, are adjusted according to the present data.…”

Section: Introductionmentioning

confidence: 99%

Artificial neural networks and robust Bayesian classifiers for risk stratification following uncomplicated myocardial infarction

Bigi

Gregori

Cortigiani

et al. 2005

International Journal of Cardiology

View full text Add to dashboard Cite

“…[1][2][3] These systems are being used to replace traditional mathematical/statistical methods and even expert systems, which often cannot be applied in complex environments. 5,6 However, the successful use of these systems in daily clinical practice has not yet been achieved because of their time-consuming methods, their "black box" character and their problematic re-learning characteristics. Neural networks are the most popular of the artificial intelligence systems currently being used.…”

mentioning

confidence: 99%

Prediction of Cyclosporine Blood Levels in Heart Transplantation Patients Using a Pharmacokinetic Model Identified by Evolutionary Algorithms

Hoda

Grimm

Laufer

2005

The Journal of Heart and Lung Transplantation

View full text Add to dashboard Cite

Neural Networks: What Are They?

Cited by 46 publications

References 6 publications

Measuring Quality of Care in a Psychiatric Hospital Using Artificial Neural Networks

Measuring Quality of Care in a Psychiatric Hospital Using Artificial Neural Networks

Artificial neural networks and robust Bayesian classifiers for risk stratification following uncomplicated myocardial infarction

Prediction of Cyclosporine Blood Levels in Heart Transplantation Patients Using a Pharmacokinetic Model Identified by Evolutionary Algorithms

Contact Info

Product

Resources

About