Abstract
GEORGIA is a modular Medical Decision Support System (MDSS) applied separately to the fields of Pulmonology and Haematology: the first introduced that treats whole classes of diseases. Its core is an integrated Knowledge/Information Base (KIB) that encapsulates the necessary medical knowledge in the form of rules, experience on the field, and actual clinical patient cases, while being evaluated by the inference properties of a Medical Data Evaluator (MDE) that handles approved subjective and objective criteria. Both KIB’s data and MDE’s evaluation process follow the medical standard of the Clinical Differential Diagnosis Methodology and are built in the synaptic weights of a novel architecture of feed-forward Artificial Neural Networks (ANNs). ANNs have been taught by means of many experimentally tested learning algorithms and a newly introduced methodology for generating Learning Patterns (LPs). These LPs can use incomplete data, independently boost their particular characteristics, and they do not grow into inhibiting numbers. GEORGIA resulted in a generalisation performance of 88%-95% correct classification of unknown medical inputs when first taught. Moreover, a `user-friendly’ human computer interface has been supplied to the MDSS (compiled by the MS-Visual Basic v5.0), to enhance its usability by physicians. Its ANNs can also be implemented in hardware by means of FPGAs and they are also coded in VHDL to assure their future portability to any VLSI design rules technology. After a four (4)-year period of its extensive testing in the Regional University General Hospital of Patras, Hellas, GEORGIA is currently under a thorough re-assessment and updating phase.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Mulsant, B. H. A neural network as an approach to clinical diagnosis. M. D. Computing, 7:25–36, 1990.
Poli, R., Cagnoni, S., Livi, R., Coppini, G., and Valli, G. An NN expert system for diagnosing and treating hypertension. IEEE Comp., 24:64–71, 1991.
Economou, G.-P. K., Goumas, P. D., and Spiropoulos, K. A novel medical decision supporting system. Invited paper in Computing & Control Engineering Journal, 7:177–183, IEE Press, 1966.
Hush, D. R., Horne, B. G. Progress in supervised neural networks. IEEE Sig. Proc. Mag., 10:8–39, 1993.
Haykin, S. Neural Networks: A Comprehensive Foundation. IEEE Press, 1994.
Larson, D. E., editor-in-chief. Mayo Clinic Family Health Book. W. Morrow and Co., Inc., NY, 1990.
Scalero, R. S., and Tepedelenlioglu, N. A fast new algorithm for training feedforward NN. IEEE Trans. on Sig. Proc., 40:202–210, 1992.
XILINX Inc. The Programmable Gate Array Data Book. 1998.
Economou, G. - P. K., Hallas, J. A., Mariatos, E. P., and Goutis, C. E. ANNs in Medical Decision Making Systems: An Application to Pulmonary Diseases’ Diagnosis through VHDL Synthesis. EDTC & EuroASIC Exhibition, Paris, France, 1995.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag London
About this chapter
Cite this chapter
Economou, G .K. (2000). Georgia: An Overview. In: Lisboa, P.J.G., Ifeachor, E.C., Szczepaniak, P.S. (eds) Artificial Neural Networks in Biomedicine. Perspectives in Neural Computing. Springer, London. https://doi.org/10.1007/978-1-4471-0487-2_9
Download citation
DOI: https://doi.org/10.1007/978-1-4471-0487-2_9
Publisher Name: Springer, London
Print ISBN: 978-1-85233-005-7
Online ISBN: 978-1-4471-0487-2
eBook Packages: Springer Book Archive