Klasifikacija srčanih bolesti na temelju EKG signala uz pomoć strojnog učenja podržanog evolucijskim računarstvom

undergraduate thesis

• Title (english): Classification of heart diseases based on ECG signal using machine learning supported by evolutionary computation
• Author: Danijel Pavlek
• Mentor: Domagoj Jakobović (mentor)
• Committee member: Domagoj Jakobović (predsjednik povjerenstva)
• Committee member: Davor Kukolja (član povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-02-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Cilj ovog završnog rada je upoznati se s najčešćim načinima za analizu signala u medicinskoj dijagnostici u svrhu prepoznavanja mogućih tegoba srca. Korišteni su modeli iz strojnog učenja, a isti su učeni numeričkim i stohastičkim metodama. Za klasifikaciju su odabrani stroj s potpornim vektorima i neuronska mreža, a učenje njihovih parametara je izvedeno algoritmima evolucijskog računanja: genetski algoritam, evolucijska strategija, kao i neuro-evolucija rastućih topologija. Opisani su problemi u korištenju pojedinih algoritama sa loše odabranim parametrima.

Abstract (english)

Goal of this project is to get familiar with methods for biomedical signal analysis for purpouse of detecting heart wave abnormalities. The machine learning techniques were used and models were trained using numerical, as well as stohastic algorithms. The classification is implemented using support vector machine and feed forward neural network. Their parameters were trained using evolutionary computation algorithms: genetic algorithm, evolution strategy, as well as neuro-evolution of augmented topologies (NEAT). Problems with which one can encounter because of badly selected parameters are described.

• Keywords: EKG; klasifikacija; strojno učenje; SVM; jezgrene metode; neuronske mreže; NEAT; evolucijska strategija; genetski algoritam; evolucijsko računarstvo
• Keywords (english): ECG; classification; machine learning; SVM; kernel methods; neural networks; NEAT; genetic algorithm; evolution strategy; evolutionary computation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:339473
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 23:42:00