Health and biosignals

Clinical prediction and physiological sensing -- survival and comorbidity models, wearables, brain-computer interfaces and motility classification -- where the cost of a wrong prediction is measured in patient outcomes.

10 supervised theses. See all supervised theses for the full list, or open topics if you are looking for one to take on.

Level	Author	Title	Date	Industry	Recording
MSc	L. Tuominen	Robust tap gesture recognition on wearable devices with domain adaptation	July 2025
BSc	N. Vo	Electricity companies as healthcare service	Apr. 2024
BSc	S. Hirvonen	Applying supervised machine learning for predicting maximal heart rate during aerobic exercise	May 2024
MSc	R. Tikkanen	Clustering fitness tracker data to correct interdevice differences in energy expenditure estimation	May 2023	Fjuul
BSc	A. Hakola	Machine learning in prediction of blood glucose values for treatment of type 1 diabetes	Dec. 2022
BSc	E. Hiitola	Predicting level of care requirements and mortality of SARS-CoV-2 cases based on patient comorbidities and characteristics	May 2021
BSc	J. Tuomaala	Learned personalized predictors for COVID-19 infections	Dec. 2021
MSc	P. Vijayakrishnan	Semi-supervised machine learning techniques for infant motility classification	Oct. 2021
MSc	J. Laiho	Recognizing thoughts from bioelectric patterns? A brain-computer interface with deep learning	Aug. 2020	Accenture Liquid Studio (NL)	video
MSc	A. Tolvanen	Cross-Domain Embedding Preservation in Biological Foundation Model Distillation for On-Device Classification	in progress