Gait Analysis Project of St. Pölten UAS Honoured with Lower Austrian Innovation Award in Research Category
A research group of the St. Pölten University of Applied Sciences and the Orthopädisches Spital Speising studies how artificial intelligence can offer support in medical decision-making for clinical gait analysis. Yesterday evening, the group was honoured with the Lower Austrian Innovation Award in the category of research at the award ceremony in Grafenegg.
Another prize was awarded to the start-up Circly. One of the start-up team members is a graduate and Junior Researcher of the St. Pölten UAS.
Insights into a Blackbox
Disorders or injuries of the locomotor system and neurological disorders may lead to pathological impairments of the human motor function. To better understand gait impairments resulting from musculoskeletal or neurological disorders, it is important for clinicians and therapists to describe and analyse the patient’s gait pattern accurately. For this purpose, gait analysis has become a crucial assessment tool. A project of the St. Pölten UAS together with the Orthopädisches Spital Speising explored how artificial intelligence can be used to support medical decision-making in a more transparent and explainable manner.
“Artificial intelligence and machine learning allow us to identify patterns in movement data that are characteristic of various gait problems. The patterns can provide new insights and generate findings that are derived directly from the data”, explains Djordje Slijepčević, researcher at the Institute of Creative\Media/Technologies at the St. Pölten UAS. To perform the analysis, researchers of the St. Pölten UAS have developed methods of pattern recognition, data visualisation, and machine learning.
“There is a big downside to today’s methods of machine learning. They are usually a black box, in other words, it is impossible to understand how the algorithm makes a concrete decision. On the one hand, this factor makes it difficult to derive new findings from the data for clinicians and therapists, and on the other hand, it leads to low levels of trust in the decisions of the algorithms by clinical experts. This is why our research examines how to make complex decision-making processes in machine learning comprehensible for experts in gait analysis”, says sports scientist Brian Horsak, Head of the St. Pölten UAS’ Center for Digital Health and Social Innovation.
Expertise in Digital Health Care and Motor Rehabilitation
“Brian Horsak and his colleagues have been working for years to establish the research focus Digital Health Care and Motor Rehabilitation at the St. Pölten UAS. I believe that this recognition with the Innovation Award is not only to honour an individual project but to give credit to years of tireless efforts to establish an entire research area and build up expertise in this field”, says Hannes Raffaseder, Chief Research and Innovation Officer of the St. Pölten UAS.
“I am very proud that our interdisciplinary team has received the Innovation Award. This highlights once more the excellent work and quality of research in the field of gait analysis achieved at the St. Pölten UAS and together with partners such as the Orthopädisches Spital Speising”, says St. Pölten UAS Executive Director Gernot Kohl.
Transparent Support in Diagnostics and Decision-Making
The interest in artificial intelligence (AI) to aid decision-making in medicine has surged in recent years. The implementation of AI-based methods helps to support clinical experts in making even more accurate and reliable decisions or making them faster and more efficient without a loss of accuracy. The “black box” nature of complex processes, however, often makes it impossible to explain how AI arrives at a concrete decision and which part of or patterns in the data have contributed to this decision. This is a big obstacle to the application AI in clinical practices.
Making these decisions explainable would substantially contribute to increasing trust in these methods. It would furthermore enable new findings from historical data volumes — something that may not be possible using traditional statistical methods. “Our research projects are dedicated to this fundamental problem. We use a practical example from clinical 3D gait analysis, especially for patients with cerebral palsy, to demonstrate how we can make AI explainable and how clinicians and therapists may use this information for decision-making in their clinical practice”, says Horsak.
Project members include Brian Horsak, Djordje Slijepčević, Matthias Zeppelzauer, Anna-Maria Raberger and Bernhard Dumphart (all from the St. Pölten UAS), Andreas Kranzl and Fabian Unglaube (Spital Speising), and Christian Breiteneder (TU Wien).
Start-Up for Trading Forecasts
The “Sonderpreis Künstliche Intelligenz” (special award for artificial intelligence) went to the start-up Circly, which supports manufacturing and commercial enterprises with smart demand forecasts. An AI kit offers a safe start into the commercial future. SMEs may benefit from modern AI technology through a self-learning and self-optimising interactive software solution. Start-up team member Armin Kirchnkopf is a graduate and Junior Researcher of the St. Pölten UAS.
The Circly team at the awards ceremony. Credit: David Schreiber
Project IntelliGait 3D — Gait Data Mining
The Orthopädisches Spital Speising is the UAS’ partner in the project. The project is funded by the Gesellschaft für Forschungsförderung (Association for Research Funding) — formerly NFB FTI Call.
The project was supported by the St. Pölten UAS’ ReMoCap Lab. This “Laboratory for Capturing Motion and Augmenting Environment in Motor Rehabilitation” enables innovative projects and cutting-edge research on the topic of motor rehabilitation in Austria. The ReMoCap Lab is funded by the Federal Ministry for Digital and Economic Affairs within the framework of the initiative “COIN Aufbau”.
Lower Austrian Innovation Award
Each year, the Lower Austrian Innovation Award honours excellent innovative projects by Lower Austrian companies and innovative developments by research institutes. The award is organised by the joint innovation service of the Province and the Economic Chamber of Lower Austria.
FH-Prof. Priv.-Doz. Dr. Brian HorsakHead of Center for Digital Health and Social Innovation Senior Researcher Institute of Health Sciences Department of Health Sciences
Dipl.-Ing. Armin Kirchknopf, BA MA BScJunior Researcher
Media Computing Research Group
Institute of Creative\Media/Technologies Department of Media and Digital Technologies