Biomedical students won the competition on Parkinson’s disease detection by artificial intelligence

Filip Mívalt, Kamila Lepková and Andrea Beháňová, students of Biomedical Engineering at Brno University of Technology (BUT), used artificial intelligence to detect Parkinson's disease (PD) in a speech record. Their algorithm won the Biosignal Challenge 2018 competition in Prague, the Czech Republic.

The three biomedical students created a team and signed up for the competition. They successfully resolved the main goal of competition being the detection of Parkinson’s disease symptoms in patient’s speech early on, prior to the disease progression and the appearance of the cardinal motor symptoms. By using this method the medical experts would be able to ensure early diagnosis of PD. The societal urgency of this topic is even more emphasized by the representatives of the competition’s organizing committee; the Faculty of Electrical Engineering of the Czech Technical University in Prague, Department of Neurology of the Charles University and the MathWorks, Inc. known as the developer of mathematical computing software MATLAB for engineers and scientists.

The competition set a challenge for all the contestants to programme an algorithm which would be able to distinguish speech characteristics typically found in patients with Parkinson’s disease. The contestants were provided with a set of speech recordings of both healthy and diagnosed individuals. From that point on, it was up to the contestants to decide which method to use in order to detect the speech of those diagnosed with PD.

Filip Mívalt explains Parkinson's disease detection algorithm
Filip Mívalt from the Department of Biomedical Engineering, the Faculty of Electrical Engineering and Communication, BUT explains the principle of algorithm for detection of speech characteristics typically found in patients with Parkinson’s disease. | Photo source: Czech Technical University in Prague archive.

BUT Biomedical Engineers showed a great form in the competition, and managed to beat research results of 26 other contestants, pushing BUT to the very top of the competition. Their accomplishment is also to be admired as they had not dealt with the research of PD patients’ speech prior to the competition. They have managed to utilise knowledge gained during their studies at the Biomedical and Ecological Engineering and Biomedical Engineering and Bioinformatics study programs and further, transfer such knowledge into an innovative idea.

Filip Mívalt and Kamila Lepková students of Biomedical Engineering at BUT receive the Diploma of Biosignal Challenge 2018 Winners. | Photo source: Czech Technical University in Prague archive.

Filip Mívalt revealed the proposed method principle: “The others focused on pronunciation periodicity of the syllable . It is just the periodicity of speech that is difficult for PD patients. However, we have decided to focus on a more innovative approach the analysis of various speech frequency characteristics. We particularly focused on the high and low voice frequency bands.” Despite achieving accuracy of only 56 % the winning team has managed to prove, on a set of real data, that this approach leads to the best differentiation between healthy and diagnosed individuals.

Kamila Lepková at Biosignal Challenge
Kamila Lepková among the Biosignal Challenge competition contestants. | Photo source: Czech Technical University in Prague archive.

The student’s achievements have not gone without notice from the editor Martin Srp of Leonardo Magazine (Czech Radio). The editor was mainly interested in immediate reactions of the winners. “For us, it was a true challenge to try something new,” described Kamila Lepková during the interview. “We deal with artificial intelligence but this was the first application where we could actually implement and test our proposed methods,” added Filip Mívalt.

For the next year’s challenge the competition contestants will deal with the topic of epilepsy detection. You can also take part and boost the passion of Biomedical Engineering students through our Talent Booster or follow their footsteps by enrolling in one of our study programmes.

Follow Us

Get the latest news about our teaching and learning activities or get excited about scientific discoveries in biomedical engineering.

Partner With Us

We are committed to promote collaboration across biomedical disciplines creating mutual benefits with companies and research organizations. Explore opportunities to engage with our research teams, or find out more about our current research partners.

Giving

Donate to support research and education that translates into improved healthcare and people’s wellbeing. With a number of giving options, you can support the area you care about most. Gifts of any size help expand our positive impact throughout the society.