Related Experiment Videos
Artificial neural networks within medical decision support systems
1Bristol Transputer Centre, Faculty of Computer Studies and Mathematics, University of the West of England, Frenchay, UK.
Summary
Artificial neural networks (ANNs) can analyze clinical data for correlations and inferences. These ANNs show promise for medical decision support and extracting diagnostic utility from data.
Area of Science:
- Biomedical Informatics
- Computational Biology
Background:
- Artificial neural networks (ANNs) offer advanced data assimilation and inference capabilities relevant to clinical laboratories.
- Traditional methods may overlook valuable correlations within complex datasets.
Purpose of the Study:
- To describe the backpropagation technique in artificial neural networks.
- To explore practical considerations for implementing ANNs in clinical settings.
- To highlight the application of ANNs in medicine, particularly clinical chemistry and medical decision support.
Main Methods:
- Description of the backpropagation algorithm for artificial neural networks.
- Review of practical implementation considerations for ANNs.
- Presentation of case examples in clinical chemistry and medical decision support.
Main Results:
- Artificial neural networks are effective multivariate techniques for clinical data analysis.
- ANNs demonstrate potential as function approximators in clinical chemistry.
- The pattern recognition capabilities of ANNs can yield diagnostic insights from underutilized data.
Conclusions:
- Artificial neural networks are valuable tools for medical decision support systems.
- ANNs can enhance diagnostic utility by extracting information from complex clinical data.
- Further application of ANNs in clinical chemistry is warranted for improved data interpretation.