Leo Kärkkäinen

Professor of Practice
Professor of Practice
T410 Dept. Electrical Engineering and Automation

Deep learning has provided means to model complicated systems using very large amounts of data. I work in collaboration with hospitals to gather image and sensor data, with annotation to build models for classification tasks. The challenge is, after building the models, to find out the essential features that predict the classification. If successful, one can get hold on some part of the “dark knowledge” from decade long experience of professionals, which  they may find hard to quantify and explain.

Also, when data is sparse, simulations of physical systems can be used to produce datasets for training regression models and classifiers. One example is utilizing convolutional neural networks to estimate material parameters in ultrasound tomography.

Finally, when the models are trained and perform well enough, the challenge is to be able to simplify these in a way that makes them run even in embedded systems without significant loss of accuracy.

Full researcher profile

Contact information

Research groups

  • Embedded Systems in Biomedical Technology


Uncertainty-Aware Deep Learning Methods for Robust Diabetic Retinopathy Classification

Joel Jaskari, Jaakko Sahlsten, Theodoros Damoulas, Jeremias Knoblauch, Simo Sarkka, Leo Karkkainen, Kustaa Hietala, Kimmo K. Kaski 2022 IEEE Access

Deep learning-based groundwater storage estimation from seismic data

T. Lähivaara, A. Malehmir, A. Pasanen, L. Kärkkäinen, J. M.J. Huttunen, J. S. Hesthaven 2022 2nd EAGE Conference on Seismic Inversion

On the scattering of a disk source by a rigid sphere for directivity broadening

Tim Mellow, Leo Kärkkäinen 2022 Journal of the Acoustical Society of America

On the sound fields of oblate and prolate hemispheroids in infinite baffles for directivity control

Tim Mellow, Leo Kärkkäinen 2021 Journal of the Acoustical Society of America

Deep learning for prediction of cardiac indices from photoplethysmographic waveform


Pulse transit time estimation of aortic pulse wave velocity and blood pressure using machine learning and simulated training data

Janne M.J. Huttunen, Leo Kärkkäinen, Harri Lindholm 2019 PLoS computational biology

Estimation of groundwater storage from seismic data using deep learning

Timo Lähivaara, Alireza Malehmir, Antti Pasanen, Leo Kärkkäinen, Janne M.J. Huttunen, Jan S. Hesthaven 2019 Geophysical Prospecting

Differential photoplethysmogram sensor with an optical notch filter shows potential for reducing motion artifact signals

K. H. Blomqvist, L. Kärkkäinen 2018 Biomedical Physics and Engineering Express

Deep convolutional neural networks for estimating porous material parameters with ultrasound tomography

Timo Lähivaara, Leo Kärkkäinen, Janne M.J. Huttunen, Jan S. Hesthaven 2018 Journal of the Acoustical Society of America

Bidirectional recurrent neural networks as generative models

Mathias Berglund, Tapani Raiko, Mikko Honkala, Leo Kärkkäinen, Akos Vetek, Juha Karhunen 2015 Advances in Neural Information Processing Systems