A.V. Yudin1, E.A. Yeliseichev2, P.S. Vorobyov3, M.V. Pankratov4, I.S. Blinov5

1, 2 Rybinsk State Aviation Technical University (Rybinsk, Russia)
3–5 LLC BioTech (Rybinsk, Russia)
1 judinav@mail.ru, 2 EvgenijEliseichev@mail.ru, 3 vorobps@gmail.com,
4 pankratov_m_v@mail.ru, 5 ilya.blinov.1998@mail.ru

Problem definition. Modern bioprostheses of the upper limbs are complex technical devices, while special attention is paid to the
user's learning process aimed at improving the efficiency of control of such a device. Despite the active development of rehabilitation engineering, in particular, the use of gamification, as well as the introduction of virtual/augmented reality technologies into the lear­ning process, there is currently no unified systematic approach to evaluate its effectiveness. Existing methods of evaluation the effectiveness of training are usually based on functional tests (cylinder test, SHAP, clothespin relocation test, BBT) or user questionnaires. The main disadvantage of these approaches is that they evaluate the final result of mastering a bioprosthesis and do not consider the intermediate stage – the stage of correct training and the use of a classifier for motion recognition, which greatly limits the use of these techniques in the case of training a user to control a bioprosthesis with an artificial neural network-based control system.

Purpose. To evaluate the effectiveness of training methods for using an upper limb prosthesis with a control system based on an artificial neural network. To achieve this purpose, the following tasks were set: to analyze existing teaching methods described in scientific papers on rehabilitation engineering over the past 10 years; to determine quantitative criteria for the effectiveness of teaching methods; to calculate the effectiveness criteria for the author's method based on artificial neural network training.

Results. Three numerical criteria of effectiveness are proposed:  – characterizes the number of mastered movements in relation to the training time;  – characterizes the probability of correct grip realization in relation to the training time;  – characterizes the quality of grip formation. The author's method of teaching the use of an upper limb prosthesis, based on the training of an artificial neural network, was tested on a subject with a healthy forearm. The total training time was
8 hours, and the average probability of correct grip implementation was 81.8%. The calculation of the effectiveness criteria for the author's method gave the following results: CE1 = 1.5 pieces/hour; CE2 = 10.2 %/hour; CE3 = 0.92. Author's method turns out to be more effective than most of the methods proposed in scientific papers over the past 10 years, according to the CE1 criterion, while it is not possible to compare it with existing methods according to the CE2 and CE3 criteria.

Practical significance. The formulated quantitative criteria of effectiveness can become the basis for creating a systematic and universal approach to evaluating methods of teaching the use of an upper limb bioprosthesis with a control system based on an artificial neural network. Using this approach will make it possible to objectively compare different rehabilitation systems and programs, as well as purposefully improve them according to one of the proposed effectiveness criteria. Further investigation of the problem may be related to the creation of a comprehensive criterion for the effectiveness of teaching methods.

Yudin A.V., Yeliseichev E.A., Vorobyov P.S., Pankratov M.V., Blinov I.S. Evaluation of the effectiveness of teaching method for using an upper limb prosthesis with a control system based on an artificial neural network. Biomedicine Radioengineering. 2025. V. 28. № 6.
P. 5–15. DOI: https:// doi.org/10.18127/j15604136-202506-01 (In Russian)

1. ResearchGate [E`lektronny`j resurs]. URL: https://www.researchgate.net/ (data obrashheniya: 27.06.2025).
2. arXiv [E`lektronny`j resurs]. URL: https://arxiv.org/ (data obrashheniya: 27.06.2025).
3. Lens [E`lektronny`j resurs]. URL: https://www.lens.org/ (data obrashheniya: 27.06.2025).
4. Perry B., Armiger R., Yu K., Alattar A., Moran C., Wolde M., Scippa K., Pasquina P., Tsao J. Virtual Integration Environment as an Advanced Prosthetic Limb Training Platform. Frontiers in neurology. 2018. V. 9. P. 1–8. DOI: 10.3389/fneur.2018.00785
5. Boschmann A., Neuhaus D., Vogt S., Kaltschmidt C., Platzner M., Dosen S. Immersive augmented reality system for the training of pattern classification control with a myoelectric prosthesis. Journal of neuroengineering and rehabilitation. 2021. V. 18. № 25. P. 1–16. DOI: 10.1186/s12984-021-00822-6
6. Kristoffersen M., Franzke A., Bongers R., Wand M., Murgia A., Sluis C. User training for machine learning controlled upper limb prostheses: a serious game approach. Journal of neuroengineering and rehabilitation. 2021. V. 18. № 32. P. 1–16. DOI: 10.1186/s12984-021-00831-5
7. Tchimino J., Maas B., van Dijk B., Murgia A., van der Sluis C.K., Bongers R.M. Effects of game design characteristics of a virtual reality serious game for upper-limb prosthesis control training on motor learning. Frontiers in rehabilitation sciences. 2025. P. 1–15. DOI: 10.3389/fresc.2025.1520184
8. Maas B., van der Sluis C.K., Bongers R.M. Assessing effectiveness of serious game training designed to assist in upper limb prosthesis rehabilitation. Frontiers in rehabilitation sciences. 2024. P. 1–12. DOI:10.3389/fresc.2024.1353077
9. Winslow B., Ruble M., Huber Z. Mobile, game-based training for myoelectric prosthesis control. Frontiers in bioengineering and biotechnology. 2018. V. 6. № 94. P. 1-8. DOI: 10.3389/fbioe.2018.00094
10. Heerschop A., Sluis C., Bongers R. Transfer of mode switching performance: from training to upper-limb prosthesis use. Journal of neuroengineering and rehabilitation. 2021. V. 18. № 85. P. 1–16 p. DOI: 10.1186/s12984-021-00878-4
11. Heerschop A., Sluis C., Bongers R. Training prosthesis users to switch between modes of a multi-articulating prosthetic hand. Disability and Rehabilitation. 2022. V.  46. № 1. P. 187–198. DOI: 10.1080/09638288.2022.2157055
12. Light C.M. An intelligent hand prosthesis and evaluation of pathological and prosthetic hand function. PhD diss., University of Southam­p­ton, UK, 2000. 152 p.
13. Belova A.N., Rukina N.N., Kuzneczov A.N., Borzikov V.V., Belova E.M. Ocenka rezul`tatov protezirovaniya verxnej konechnosti. Medicina. 2017. T.  5. № 3(19). S. 85–107.
14. Mathiowetz V., Volland G., Kashman N. Weber K. Adult norms for the box and block test of manual dexterity. American Journal of Occupational Therapy. 1985. V. 39. № 6. P. 386–391.
15. Bez``yazy`chny`j V.F., Yudin A.V., Pankratov M.V., Eliseichev E.A., Vorob`ev P.S., Blinov I.S. Klassifikaciya dvizhenij iskusstvennoj nejronnoj set`yu dlya upravleniya bionicheskoj kist`yu. Iskusstvenny`j intellekt i prinyatie reshenij. 2025. № 1. S. 33–45. DOI: 10.14357/20718594250103
16. Bez``yazy`chny`j V.F., Yudin A.V., Mixajlov V.V., Eliseichev E.A., Pankratov M.V., Vorob`ev P.S., Blinov I.S. Ocenka tochnosti vy`polneniya xvatov antropomorfnoj bionicheskoj kisti. Vestnik mashinostroeniya. 2025.
17. Burov G.N., Bol`shakov V.A., Drobaxa A.S. Protezirovanie verxnix konechnostej: ocenka uspeshnosti metodom traektornogo analiza. Fizicheskaya i reabilitacionnaya medicina. 2021. T. 3. № 4. S. 32–40. DOI: 10.26211/2658-4522-2021-3-4-32-40
18. Murav`ev S.A. Bionicheskij protez kisti ruki. Patent Rossii № 176303U1. 2017.
19. Ivolga D.V., Xomutov E.E`., Borisov I.I., Molchanov N.A., Maksimov I.A., Kolyubin S.A. Programmno apparatnaya arxitektura sistemy` upravleniya antropomorfnoj kist`yu robota. Izv. vuzov. Priborostroenie. 2022. T. 65. № 3. S. 164–173. DOI: 10.17586/0021-3454-2022-65-3-164-173
20. Krechetov I., Skvortsov A., Poselsky I. Experimental Study of 7 DOF Bionic Forearm Prosthesis. International Journal of Engineering Trends and Technology. 2020. V. 68. P. 40–48. DOI: 10.14445/22315381/IJETT-V68I9P208
21. Borisov I., Borisova O., Krivosheev S., Oleynik R., Reznikov S. Prototyping of EMG-Controlled Prosthetic Hand with Sensory System. IFAC-PapersOnLine. 2017. V. 50. № 1. P. 16027–16031. DOI: 10.1016/j.ifacol.2017.08.1915
22. Controzzi M., DAlonzo M., Peccia C., Oddo C., Carrozza M.C., Cipriani C. Bioinspired Fingertip for Anthropomorphic Robotic Hands. Applied Bionics and Biomechanics. 2014. V. 11. P. 25–38. DOI: 10.3233/ABB-140092
23. Guochao B. Adaptive and Reconfigurable Robotic Gripper Hands with a Meso-Scale Gripping Range. PhD diss., Heriot-Watt University, School of Engineering and Physical Sciences, Scotland, 2018. 185 p.
24. Claret J.R., Suárez R. Efficient and practical determination of grasping configurations for anthropomorphic hands. IFAC Proceedings Volumes. 2011. V. 44. № 1. P. 14660–14666. DOI: 10.3182/20110828-6-IT-1002.00711
25. Manoharan V.S., Sundaram S., Jason I. Factors Affecting Hand Grip Strength and its Evaluation: a Systemic Review. International Journal of Physiotherapy and Research. 2015. V. 3. P. 1288–1283. DOI: 10.16965/ijpr.2015.193