Training a learning vector quantization network using the pattern electroretinography signals

• Published in: Computers in biology and medicine Vol. 37; no. 1; pp. 77 - 82
• Main Authors: Kara, Sadık, Güven, Ayşegül
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.01.2007; Elsevier Limited
• Subjects: Adult; Aged; Artificial Intelligence; Artificial neural network; Computer Simulation; Diagnosis, Computer-Assisted; Electroretinography - statistics & numerical data; Female; Humans; Internal Medicine; Learning vector quantization network; Male; Middle Aged; Models, Neurological; Neural Networks (Computer); Optic nerve disease; Optic Nerve Diseases - diagnosis; Other; Pattern electroretinography; Reference Values; Signal Processing, Computer-Assisted; Optic nerve disease; Artificial neural network; Pattern electroretinography; Learning vector quantization network
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2005.10.005

In this study, the pattern electroretinography (PERG) signals derived from evoked potential across retinal cells of subjects after visual stimulation were analyzed using artificial neural network (ANN) with 172 healthy and 148 diseased subjects. ANN was employed to PERG signals to distinguish between healthy eye and diseased eye. Supervised network examined was a competitive learning vector quantization network. The designed classification structure has about 94% sensitivity, 90.32% specifity, 5.94% false negative, 9.67% false positive and correct classification is calculated to be 92%. Testing results were found to be compliant with the expected results that are derived from the physician's direct diagnosis. The end benefit would be to assist the physician to make the final decision without hesitation.