Classifying drug-induced epileptic seizures in zebrafish larvae

Zebrafish (Danio rerio) is one of the model organisms used for epilepsy research and have been found to have enormous potential for modelling acute seizures and genetic epilepsies. In this project, a mathematical classification model was created with the aim of correctly classifying zebrafish larvae having an (PTZ induced) epileptic seizure. The model also computes biomedical parameters describing behavioural attributes of the larvae. Several data mining approaches were implemented for the data, of which best results were obtained for support vector machine using second-degree polynomial kernel function. Results indicate 91.33% classification accuracy on training data and when run on the new testing datasets similar classification accuracy was obtained, 90.96%. A future application of this model is to aid in the development of new medication to treat epilepsy.

Author dc:creator

• Lilja Níelsdóttir 1987-

Contributors dc:contributor

• Háskólinn í Reykjavík

• Heilbrigðisverkfræði
• Flogaveiki
• Fiskar
• Reiknilíkön
• Tækni- og verkfræðideild
• Meistaraprófsritgerðir
• Biomedical engineering
• Epilepsy
• Fish
• Mathematical models
• School of Science and Engineering

Language dc:language.iso

en

Handle dc:identifier.uri

http://hdl.handle.net/1946/22317

OAI identifier oai:identifier

oai:skemman.is:1946/22317