Duboko učenje za procjenu stanja pacijenata s neurodegenerativnim bolestima na temelju snimaka elektroencefalografom

master's thesis

• Title (english): Deep Learning for EEG Assessment of Neurocognitive Disease Progression
• Author: Ena Džanko
• Mentor: Vladimir Čeperić (mentor)
• Committee member: Vladimir Čeperić (predsjednik povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Committee member: Tomislav Marković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Fokus rada je implementacija dubokog učenja u svrhu procjene stanja pacijenata s neurodegenerativnim bolestima. Rad sadrži osnovne koncepte u području neuroznanosti i bioelektričnog potencijala mozga, te objašnjava procese elektroencefalografskog snimanja mozga i implementira obrade biosignala s izvorom u mozgu. Podaci su dobiveni mjerenjem elektroencefalografom prilikom istraživanja reakcije na podražaj po 3-Oddball paradigmi. Nakon procesiranja se radi ekstrakcija značajki valićnom transformacijom te je u svrhu dubokog učenja implementirano par modela neuronskih mreža, s fokusom na CNN i LSTM modele mreža. Procjena stanja vrši se na temelju reakcije na podražaj gdje se u konačnom modeli koriste kao klasifikatori u svrhu predikcije i uznapredovalosti bolesti.

Abstract (english)

The focus of the paper is the implementation of deep learning algorithms for the purpose of assessing the condition of patients with neurodegenerative diseases. The paper contains basic concepts in the field of neuroscience and bioelectrical potential of the brain, explains the processes of electroencephalographic imaging of the brain and implements processing of biosignals with a source in the brain. Data were obtained by electroencephalograph measurement when investigating the response to a stimulus by the 3-Oddball paradigm. After processing, the features are extracted by wave transformation, and for the purpose of deep learning, common neural network models were implemented, with a focus on CNN and LSTM network models. Condition assessment is performed based on stimulus response where the models are used as classifiers for the purpose of disease prediction and progression.

• Keywords: EEG; duboko učenje; neurodegenerativne bolesti; CNN; LSTM
• Keywords (english): EEG; deep learning; neurocognitive diseases; CNN; LSTM
• Language: croatian
• URN:NBN: urn:nbn:hr:168:357398
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 08:43:46