NORMA eResearch @NCI Library

Classification of Seizure Disorders using Machine Learning

Bose, Dibyajyoti (2016) Classification of Seizure Disorders using Machine Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration File]
PDF (Configuration File)
Download (483kB) | Preview


The unpredictable nature of seizures poses a risk to patients with epilepsy.The aim of this study is to predict epileptic-seizures from EEG/MRI signals and patient-medical history by using machine-learning classifiers. The dataset is collected from NRS Medical Hospital, India, and comprises patients EEG/MRI findings, details of convulsion and type of seizure disorder. Although many studies have been performed to classify seizures based on para-clinical evidence obtained from conventional MRI and EEG, very little work had been hitherto done to characterize various types of seizures by taking into consideration associated clinical factors such as concomitant diseases, impaired-ADL, etiology of seizures, drug/alcohol abuse and family-history. This paper presents a supervised machine-learning approach that classifies seizure types (partial and generalized) using a dataset containing 150 records. In this paper, multiple machine learning techniques are applied and their performance is examined. The impact of feature engineering and parameter tweaking are explored with the objective of achieving superior predictive performance. Out of the various machine learning algorithms used, namely ANN,kNN and Random Forest, the the kNN classifier showed the best results (90% accuracy and 87% sensitivity). The model is evaluated on the testing data by using k-fold cross validation. The study will help clinicians to make diagnosis of epilepsy and initiate timely treatment; also, it will help a primary physician to decide the next step without the intervention of a trained neurologist.

Item Type: Thesis (Masters)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
R Medicine > Healthcare Industry
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Caoimhe Ní Mhaicín
Date Deposited: 03 Dec 2016 11:25
Last Modified: 03 Dec 2016 14:51

Actions (login required)

View Item View Item