Abstract
AIMS: To advise a system of neuronal networks which can classify the densitometric patterns of serum electrophoresis. METHODS: Digitised data containing 83 normal and 132 pathological serum protein electrophoresis patterns were presented to four neuronal networks containing 1900 neurons. Network 1 evaluates the integrated values of the albumin, alpha 1, alpha 2, beta and gamma fractions together with total protein (Biuret method). Networks 2, 3, and 4 analyse the shape of the albumin, beta and gamma fractions. To increase the sensitivity for the detection of monoclonal gammopathies a Fourier transformation was applied to the beta and gamma fractions. RESULTS: After a learning period of 20 minutes (back-propagation learning algorithm) the system was tested with a set of electrophoresis patterns comprising 446 routinely collected samples. It differentiated between physiological and pathological curves with a sensitivity of 97.5% and a specificity of 98.8%, with 86% correct diagnoses. All monoclonal gammopathies were recognised by the Fourier detector. CONCLUSIONS: Neuronal networks could be useful for certain medical uses. Unlike rule based systems, neuronal networks do not have to be programmed but have the capacity to "learn" quickly.
Full text
PDF
Selected References
These references are in PubMed. This may not be the complete list of references from this article.
- Heathfield H., Bose D., Kirkham N. Knowledge-based computer system to aid in the histopathological diagnosis of breast disease. J Clin Pathol. 1991 Jun;44(6):502–508. doi: 10.1136/jcp.44.6.502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Knüppel W., Neumeier D., Fateh-Moghadam A., Knedel M. Rechnerunterstützte Befundung von Eiweisselektrophoresen auf Celluloseacetatfolie. J Clin Chem Clin Biochem. 1984 Jun;22(6):407–417. [PubMed] [Google Scholar]