Research on machine learning approaches for upper limb prosthesis control has shown impressive progress. However, translating these results from the lab to patient’s everyday lives remains a challenge, because advanced control schemes tend to break down under everyday disturbances, such as electrode shifts. Recently, it has been suggested to apply adaptive transfer learning to counteract electrode shifts using as little newly recorded training data as possible. In this paper, we present a novel, simple version of transfer learning and provide the first user study demonstrating the effectiveness of transfer learning to counteract electrode shifts. For this purpose, we introduce the novel Box and Beans test to evaluate prosthesis proficiency and compare user performance with an initial simple pattern recognition system, the system under electrode shifts, and the system after transfer learning. Our results show that transfer learning could significantly alleviate the impact of electrode shifts on user performance in the Box and Beans test. INDEX TERMS Electromyography, Electrode Shifts, Transfer Learning.

A multidisciplinary journal on gaming and gamification for health education/promotion, teaching and social change.

Four patients underwent targeted sensory reinnervation (TSR), a surgical technique in which a defined skin area is first selectively denervated and then surgically reinnervated by another sensory nerve. In our case, either the area of the lateral femoral cutaneous nerve or the saphenous nerve was reinnervated by the sural nerve. Patients were then fitted with a special prosthetic device capable of transferring the sense of pressure from the sole of the prosthesis to the newly wired skin area. Pain reduction after TSR was highly significant in all patients. In three patients, permanent pain medication could even be discontinued, in one patient the pain medication has been significantly reduced. Two of the four patients were completely pain-free after the surgical intervention. Surgical rewiring of existing sensory nerves by TSR can provide the brain with new afferent signals seeming to originate from the missing limb. These signals help to reduce phantom limb pain and to restore a more normal body image. In combination with special prosthetic devices, the amputee can be provided with sensory feedback from the prosthesis, thus improving gait and balance.

Background: Restoring hand and finger function after a traumatic hand injury necessitates a regimen of consistent and conscientious exercise. However, motivation frequently wanes due to unchallenging repetitive tasks or discomfort, causing exercises to be performed carelessly or avoided completely. Introducing gamification to these repetitive tasks can make them more appealing to patients, ultimately increasing their motivation to exercise consistently. Objective: This study aims to iteratively develop a serious virtual reality game for hand and finger rehabilitation within an appealing and engaging digital environment, encouraging patient motivation for at least 2 weeks of continuous therapy. Methods: The development process comprised 3 distinct stages, each of which was subject to evaluation. Initially, a prototype was created to encompass the game’s core functionalities, which was assessed by 18 healthy participants and 7 patients with impaired hand function. Subsequently, version 1 of the game was developed and evaluated with 20 patients who were divided into an investigation group and a control group. On the basis of these findings, version 2 was developed and evaluated with 20 patients who were divided into an investigation group and a control group. Motivation was assessed using the Intrinsic Motivation Inventory (IMI), while the application’s quality was rated using the Mobile Application Rating Scale and the System Usability Scale. User feedback was gathered using semistructured interviews. Results: The prototype evaluation confirmed the acceptance and feasibility of the game design. Version 1 significantly increased motivation in 2 IMI subscales, effort (P<.001) and usefulness (P=.02). In version 2, a significant increase in daily performed exercises was achieved (P=.008) compared to version 1, with significantly higher motivation in the IMI subscale effort (P=.02). High Mobile Application Rating Scale scores were obtained for both versions 1 and 2, with version 2 scoring 86.9 on the System Usability Scale, indicating excellent acceptability. User feedback provided by the semistructured interviews was instrumental in the iterative development regarding improvements and the expansion of the playable content. Conclusions: This study presented a virtual reality serious game designed for hand and finger rehabilitation. The game was well received and provided an environment that effectively motivated the users. The iterative development process incorporated user feedback, confirming the game’s ease of use and feasibility even for patients with severely limited hand function.