Ilkka Suuronen receives award for best master's thesis in computer science


Every year, the Academic Engineers and Architects in Finland TEK, Tekniska Föreningen i Finland TFiF and the Academic Association for Mathematics and Natural Sciences MAL present awards for the best doctoral thesis of the year, the best master’s thesis in engineering or architecture and the best master’s thesis in mathematics, physics or computer science. 

The award for the best master’s thesis in mathematics, physics or computer science was presented to Ilkka Suuronen from the University of Turku, Department of Computing. The name of the thesis is “Parkinsonin taudin tunnistaminen elektroenkefalogrammista koneoppimisteknologian avulla” (“The use of machine learning technology for detecting Parkinson’s disease from EEG data”).

Suuronen examined how accurately a machine learning algorithm can distinguish between healthy subjects and subjects with Parkinson's disease based on EEG data when using a limited number of EEG channels in different regions of the brain. EEG is short for electroencephalogram. In an EEG test, electrodes are attached to the head to measure the electrical impulses of the brain. 

“The idea is to simulate a situation where the EEG has been recorded using a relatively small number of electrodes, which could reduce the preparation time of EEG tests. The electrodes were selected using a greedy search algorithm, meaning that at each stage of the process, the algorithm chooses the electrode that most improves classification accuracy, or the quality variables calculated from its signal, and adds it to the selection,” says Suuronen.

Suuronen discovered this interesting topic during their summer job at the University of Turku and learned more about the topic with the help of the future thesis supervisors and other staff.

“I want to thank Antti Airola, Tapio Pahikkala, Henry Railo and Mika Murtojärvi – I couldn’t have written this thesis without them.”

The outcome of the thesis was that the algorithm learned to distinguish between a person with Parkinson’s disease and a healthy subject with an average classification accuracy of 73%. If data collected by only ten electrodes was used, the result was about one per cent weaker.

“The purpose of my thesis is to improve our understanding of Parkinson's disease. In the long run, I naturally hope that my research would also facilitate the development of clinical applications.”

Suuronen has continued to work with the theme of machine learning even after their thesis.

TEK and MAL presented Suuronen with an honorary certificate and an award of 5 000 euros. TEK and MAL have been giving out the award for the best master’s thesis in mathematics, physics or computer science since 2007. The idea is to highlight the importance of these disciplines in Finnish society.

The award for the best doctoral thesis was given to Juha Heinonen from the Aalto University School of Electrical Engineering, Department of Electronics and Nanoengineering. TEK and TFiF presented the award for the best master’s thesis in engineering or architecture in 2023 to Jari Leinonen from the LUT University School of Engineering Science, Industrial Engineering and Management. 

“These theses demonstrate the excellent technological development that is done in Finland. It’s a pleasure to be able to present these awards and to highlight this work and the people behind it,” says Mikko Särelä, Expert for Industrial and Innovation Policy at TEK.

“The versatile skills, dedication and expertise that come across from these theses is wonderful to witness. The results of the theses benefit a wide range of individuals, companies and society as a whole. A warm congratulations to the winners,” says CEO of TFIF Annika Nylander

Created 13.11.2023 | Updated 13.11.2023