Dr. Med., Professor I.I. Novikova¹
Dr. Hab., Professor V.N. Konovalov²
PhD, Associate Professor E.V. Usacheva³
PhD, Associate Professor O.M. Kulikova^{1, 4
1}Novosibirsk Scientific Research Institute of Hygiene of Rospotrebnadzor, Novosibirsk
²Siberian State University of Physical Education and Sports, Omsk
³Omsk State Medical University, Omsk
⁴The Siberian State Automobile and Highway University (SibADI), Omsk

Objective of the study was to develop a set of indicators for the objectification of human walking.
Methods and structure of the study. The research methodology is based on the use of neural networks to reconstruct the coordinates of points of elements of the human body from a video fragment using the Body_25 model. The calculation of movement parameters was carried out using parametric geometry methods.
Results and conclusions. To assess walking by expert means, the following parameters are identified that are subject to registration: 1) amplitude of the pendulum-like movement of the arms during walking; 2) walking speed; 3) step length; 4) height of the foot; 5) step frequency; 6) the angle between the thigh and lower leg; 7) the angle between the shoulder and forearm; 8) the angle between the head and the vertical, determined by a straight line perpendicular to the plane of the walking surface. Calculation of the coordinates of points of body elements is carried out using the OpenPose system. The developed approach was tested using the example of assessing movement parameters using the walking example of an 8-year-old girl. A new approach to identifying movements during walking has been developed and gait parameters have been defined, which makes gait analysis more accessible, especially in areas where there is no experience in gait assessment, including in medical research.

Keywords: walking, walking parameters, artificial intelligence, OpenPose.

References

1. Gui Yu. Sovremennyye vozmozhnosti i vyzovy dlya uchenykh v oblasti biomekhaniki sporta [Modern opportunities and challenges for scientists in the field of sports biomechanics]. Teoriya i praktika fizicheskoy kultury. 2024. No. 1. pp. 12-13.
2. Koroleva S.V. Novaya tekhnologiya otsenki khodby i primer yeye primeneniya v profilaktike travmatizma pri podgotovke sportsmenov [New technology for assessing walking and an example of its application in the prevention of injuries during the training of athletes]. Fizicheskaya kultura i sport v strukture professionalnogo obrazovaniya: retrospektiva, realnost i budushcheye [Physical culture and sports in the structure of professional education: retrospective, reality and future]. Proceedings national scientific-practical conference, Irkutsk, April 30, 2020. Irkutsk: Vostochno-Sibirskiy institut Ministerstva vnutrennikh del Rossiyskoy Federatsii publ., 2020. pp.  233-237.
3. Rybina E.A., Vlasova A.A., Rotanova V.A. et al. Polozhitelnoye vozdeystviye khodby na organizm [Positive effects of walking on the body]. Sovremennyye nauchnyye issledovaniya i innovatsii. 2020. No. 5 (109). p. 28. Available at: http://web.snauka.ru/issues/2020/05/92508 (date of access: 16.09.2023).
4. Sirotkina I.E. Biomekhanika: mezhdu naukoy i iskusstvom [Biomechanics: between science and art]. Voprosy istorii yestestvoznaniya i tekhniki. 2011. Vol. 32. No. 1. pp. 46-70.
5. Baroudi L., Yang H., Newman M.W. et al. Investigating walking speed variability of young adults in the real world. Gait Posture. 2022. Vol.  98. pp. 69-77.
6. Bohannon R.W., Andrews A.V. Normal walking speed: a descriptive meta-analysis. Physiotherapy. 2011. Vol. 97. No. 3. pp. 182-189.
7. Kamnardsiri T., Boripuntakul S., Kayket S. Computer vision-based instantaneous speed tracking system for measuring the subtask speed in the 100-meter sprinter: Development and concurrent validity study. Heliyon. 2024. p. e24086.
8. Li Q., Young M., Naing V., Donelan J.M. Walking speed estimation using a shank-mounted inertial measurement unit. Biomech. 2010. Vol. 43. No. 8. pp. 1640-1643.
9. Pavei G., Cazzola D., La Torre A., Minetti A.E. The biomechanics of race walking: literature overview and new insights. Eur J Sport Sci. 2014. Vol. 14. No. 7. pp. 661-670.
10. Ramesh S.H. Automated Implementation of the Edinburgh Visual Gait Score (EVGS) Using OpenPose and Handheld Smartphone Video. Sensors (Basel). 2023. 101 p.
11. Rodman C.H., Martin A.E. Quantification of spatiotemporal parameter behavior during walking speed transitions. Biomech. 2020. No. 7. Vol. 112. p. 110068.
12. Smith J.A., Stubbert H., Bagwell J.J. et al. Do people with low back pain walk differently? A systematic review and meta-analysis. Sport Health Sci. 2022. Vol. 11. No. 4. pp. 450-465.
13. Wu J., Mohrenbrecher H., Schaer A. et al. Human gait-labeling uncertainty and a hybrid model for gait segmentation. Front Neurosci. 2022. Vol. 16. p. 976594.
14. Yousef R.N., Khalil A.T., Samra A.S., Ata M.M. Model-based and model-free deep features fusion for high performed human gait recognition. Supercomput. 2023. pp. 1-38.
15. Zihajehzadeh S., Park E.J. Regression Model-Based Walking Speed Estimation Using Wrist-Worn Inertial Sensor. PLoS One. 2016. Vol. 11. No. 10. p. e0165211.