Student Story - How to Combine Biomechanics and Machine Learning

#GetToKnowJYU with Neil J. Cronin, an associate professor of Biomechanics at the Faculty of Sport and Health Sciences and Master's student in Cognitive Computing and Collective Intelligence.

Neil moved from England to Finland for the first time in 2005 and then lived in other countries for a few years. But he moved back to Finland in 2011, and has been here for almost 10 years! As an associate professor of Biomechanics, Neil teaches and studies how our muscles and tendons work in different situations. Traditionally, researchers in the Faculty of Sport and Health Sciences do not need machine learning technologies. But in every exciting project, there is always some boring repetitive work. Therefore, Neil started to learn machine learning and applied it on his projects for improving the research process. Let’s see how he combines these two interesting fields!

One of Neil’s favourite work so far is monitoring muscles movement by ultrasound devices. They observe how muscles and tendons from different parts of the human body change in real-time, such as length and thickness. In the picture above, Neil is showing how to observe the end of a muscle with an ultrasound device. Neil said this is a simple imaging method, and sometimes not as good as MRI. But this is real-time and an MRI machine might cost 1 million euros.

For research purposes, each muscle and tendon has to be marked and measured as shown in the picture above. Imagining how difficult and annoying it is! To solve this problem, Neil started to try utilizing artificial intelligence method for analysing ultrasound images since 2011. However, AI is not magic, the results were not that satisfying at the beginning. But as you can see, after all these years, they have managed to make fully automated ultrasound analysis! You can learn more about this from here: Neil Cronin Fully automated ultrasound analysis

Besides ultrasound imaging, they also analyse human body movements with RGB videos, as shown in the picture above. In the past, they needed specialized camera system for marking. But now, with neural network, they can do motion analysis without markers or special cameras. This kind of approach is widely used in the sports field to understand how our joints are configured, which is particularly useful for rehabing.

There are also other interesting equipment in their lab. For example, this machine can help they experiment how our stretching, age and other factors influence our reflexis function. They have a 4-people technic team for making and maintaining all these devices. The machine shown in the above picture is one of their great works! If you want to learn about Neil’s projects, you can visit his homepage: Neil J. Cronin - Home. Or follow him on Twitter:  Neil Cronin @NeilJCronin84

Now, have you found interdisciplinarity charming? To explore deeper in artificial intelligence, Neil started studying at COIN (Cognitive Computing and Collective Intelligence) programme at JYU from 2018. If you are also interested in our International Master's Degree Programmes, stay tuned with us!