Concept of the Intelligent Support of Decision Making for Manufacturing a 3D-Printed Hand Exoskeleton within Industry 4.0 and Industry 5.0 Paradigms

: Supporting the decision-making process for the production of a 3D-printed hand exoskeleton within the Industry 4.0 and Industry 5.0 paradigms brings new concepts of manufacturing procedures for 3D-printed medical devices, including hand exoskeletons for clinical applications. The article focuses on current developments in the design and manufacturing of hand exoskeletons and their future directions from the point of view of implementation within the Industry 4.0 and Industry 5.0 paradigms and applications in practice. Despite numerous publications on the subject of hand exoskeletons, many have not yet entered production and clinical application. The results of research on hand exoskeletons to date indicate that they achieve good therapeutic effects not only in terms of motor control, but also in a broader context: ensuring independence and preventing secondary motor changes. This makes interdisciplinary research on hand exoskeletons a key study influencing the future lives of patients with hand function deficits and the further work of physiotherapists. The main aim of this article is to check in what direction hand exoskeletons can be developed from a modern economic perspective and how decision support systems can accelerate these processes based on a literature review, expert opinions, and a case study.


Introduction
The concept of assisted human movement by an exoskeleton (external skeleton, wearable robot) has, over the last 70 years, gone from science fiction to developed prototypes, some of which have already been commercialized and others of which are on the threshold of commercialization and industrial production, as personalized products based on the latest technologies within the paradigm of Industry 4.0 and Industry 5.0.The Industry 4.0 paradigm places particular emphasis on computerization, automation, robotization, and technical control throughout the manufacturing process, including those based on the Industrial Internet of Things (IIoT) and artificial intelligence (AI).The Industry 5.0 paradigm is more human-centered and, therefore, the personalization issues of mass production are more prominent in it.This is due to the versatility of exoskeletons supporting both healthy people (e.g., in jobs requiring straining and/or maintaining awkward positions or limbs alone), temporarily in convalescence or permanently due to congenital or acquired functional deficits (including older people with neurodegenerative changes).This article addresses issues of relevance from both industrial and social perspectives.In the industrial area, this includes the use of exoskeletons to reduce fatigue and increase the lifting capacity and productivity of mainly manual workers.In the social area, on the other hand, it concerns the improvement in people with mobility deficits (including those with neurological/neurodegenerative causes, which are increasing in ageing societies).Due to the increasing proportion of older people in the population of developed countries, as well as the increasing automation of production and the growing awareness of the importance of protecting workers from physical strain during prolonged heavy work, the use of this group of technical solutions will increase, as will the requirements for their mass production.To begin with, we analyzed the state of the art to identify research gaps and needs for new approaches to the design and production of exoskeletons in line with the Industry 4.0 and Industry 5.0 paradigms.In the IoT layer, the key roles are played by the following:

•
Used materials, construction features, and adaptation to the functional state of the user; • Effectors (individual actuators of the exoskeleton); • Sensors (within the perceptual layer).
These are included in the exoskeleton control cycle aimed at improving the functions of the exoskeleton user.This brings with it the possibility of improving human-machine cooperation, also with the help of AI or three-dimensional (3D) printing.The data from the sensors, after processing and integration, concern the current positions of individual parts of the body, the strength of movement, or the need for its support.Research is currently underway on the criteria for sensors (kinematic, kinetic) used in exoskeletons and their standard and non-standard (dedicated to medical applications, including those using signals from the nervous system or muscles) varieties [1].Three-dimensional printing (additive manufacturing-AM) increases the versatility, the possibility of personalizing the production process, as well as the possibility of monitoring the life cycle of products for more complete recycling [2].Similar advances are being made in the area of exoskeletal drives and their adaptability [3].
Data-driven decision making in production and maintenance involves combining decision making with virtual or augmented reality to provide a seamless interface reflecting the real-world image and superimposed virtual information augmentation for production operators.Added to this are methods and techniques for dealing with data uncertainty, the integration of maintenance decision making with other operations such as scheduling and planning, the use of the cloud continuum for optimal deployment of decision-making services, the ability of decision-making methods to deal with large data sets, and the consideration of using data for decision making.
The synergy of the described technologies increases the economic, social, and environmental impact, and can result in faster production operations, greater efficiency, and savings (energy, material, and labor costs), as well as a smaller carbon footprint.However, greenwashing that sometimes appears leads to the not-always-true belief that faster mass production goes hand in hand with reducing the total consumption of fossil fuels, even with a smaller carbon footprint of the manufactured devices.The abovereasoning is much more complex and requires additional analyses and comparisons.Problems may also include mechanical resistance of manufactured parts, poor surface quality resulting from these processes, and difficulties in assembly.A clear scientific and technological basis for the proposed production processes is needed here.There is no standardization of production and evaluation, which limits their applications not only in the industrial, but also in the medical field.Research and development works and the construction of prototypes alone will not replace complete implementations.It should not be forgotten that exoskeletons are designed to reduce physical effort, reducing the incidence of musculoskeletal disorders; hence, it is important to try to define measures and frameworks for functional validation of exoskeletons that may constitute the definition of industry standards [4].Three-dimensional-printed medical exoskeletons require specific compliance with standards (Medical Devices Regulation, ISO 13485) of material selection (including due to contact with tissue and body fluids), design, personalization (including the type and degree of deficit), optimization (including using AI), production, technical inspection, testing, and recycling.They also require work on maintaining the technological rigor required for medical devices and developing new business models necessary to achieve the profitability of personalized mass production.This often requires a combination of a whole group of technologies, i.e., 3D scanning, 3D printing, and reverse engineering as part of the procedures of the healthcare system.Preparation of medical specialists, strategic changes, and proper implementation based on the needs of patients will determine the success and opportunities for further development of 3D-printed medical exoskeletons [5].There is still a lot to be done, e.g., hand exoskeletons still do not meet the criteria of mass and control, which limits the development of hand rehabilitation robots, and cases of their commercial and clinical success are rare.Currently, research on exoskeletons focuses on their mechanical design (suitable for the human hand) and control systems (with kinematics parameters compatible with simulations and topology optimization) [6].Personalized elements with shape memory or flexion/extension angle limitations, such as chainmails, are increasingly common.Elements of exoskeletons with programmable properties in the area of, e.g., stiffness/flexibility depending on the direction, better reflect the individual needs of users (different types and degrees of deficit) and are easier to mass-produce, because their properties result from a well-thought-out combination of many small elements, and are easy to mass design and produce.Such 3D-printed chainmail elements create a structure with a personalized one-or two-way bending module.Adaptive machine learning (ML) based chainmails are more efficient to produce, especially for more complex shapes and features.Basing the chainmail project on real patient data and accelerating their analysis and production using AI is an innovative approach with a chance of wider implementation [7].Designing an exoskeleton with a fairly low energy consumption (longer autonomy) is also a challenge [8].
Given the increasing number of articles on exoskeletons in general, the limited number of references makes it difficult to provide an accurate description of the state of the art in the area of manufacturing problems and related decision making.Supporting the decisionmaking process for the production of 3D-printed hand exoskeletonswithin the Industry 4.0 and Industry 5.0 paradigms brings new concepts of manufacturing procedures for 3D-printed medical devices, including hand exoskeletons for clinical applications.Despite numerous publications on the subject of hand exoskeletons, many have not yet entered production and clinical applications.The results of research on hand exoskeletons to date indicate that they achieve good therapeutic effects not only in terms of motor control, but also in a broader context: ensuring independence and preventing secondary motor changes.This makes interdisciplinary research on hand exoskeletons a key study influencing the future lives of patients with hand function deficits and the further work of physiotherapists.The topic under discussion is not as popular and frequently reported in the literature as it seemed: a review of six major bibliographic databases (PubMed, WoS, Scopus, DBLP, Cochrane, EBSCO) with keywords successively 'exoskeleton' (Figure 1) and related words, 'hand exoskeleton' and related words (Figure 2), '3D print' + 'hand exoskeleton' and related words (Figure 3) and 'decision making' + 'hand exoskeleton' yielded, respectively:  Only one publication [9].
The literature review shows that relatively few publications are devoted to decisionmaking processes in planning, designing, and fitting hand exoskeletons, including 3Dprinted ones.This does not change the fact that interest in this group of technologies has been growing for over ten years, not only in terms of clinical but also industrial applications.Technologically supported functional assessment of the hand and methods and techniques of objective, reliable, and sensitive clinical decision making in this area are being improved.There are early concepts based on gloves, instrumented stations, or sensor systems, sometimes together with vision-based motion capture systems for kinematic and kinetic analysis.But there is a complete lack of standardization in terms of requirements, system metrics, and assessment methods, preventing simple comparisons between concepts and then solutions based on them.For the above reasons, the path to acceptance of the above-mentioned technology in practical applications is still very far away [9].The literature review shows that relatively few publications are devoted to decision-making processes in planning, designing, and fitting hand exoskeletons, including 3D-printed ones.This does not change the fact that interest in this group of technologies has been growing for over ten years, not only in terms of clinical but also industrial applications.Technologically supported functional assessment of the hand and methods and techniques of objective, reliable, and sensitive clinical decision making in this area are being improved.There are early concepts based on gloves, instrumented stations, or sensor systems, sometimes together with vision-based motion capture systems for kinematic and kinetic analysis.But there is a complete lack of standardization in terms of requirements, system metrics, and assessment methods, preventing simple comparisons between concepts and then solutions based on them.For the above reasons, the path to acceptance of the above-mentioned technology in practical applications is still very far away [9].
Bearing in mind the paradigm of Industry 5.0 (human-centered), holistic methodologies for planning, optimizing, and integrating exoskeletons are being developed (Figure 4) [10].The literature review shows that relatively few publications are devoted to decision-making processes in planning, designing, and fitting hand exoskeletons, including 3D-printed ones.This does not change the fact that interest in this group of technologies has been growing for over ten years, not only in terms of clinical but also industrial applications.Technologically supported functional assessment of the hand and methods and techniques of objective, reliable, and sensitive clinical decision making in this area are being improved.There are early concepts based on gloves, instrumented stations, or sensor systems, sometimes together with vision-based motion capture systems for kinematic and kinetic analysis.But there is a complete lack of standardization in terms of requirements, system metrics, and assessment methods, preventing simple comparisons between concepts and then solutions based on them.For the above reasons, the path to acceptance of the above-mentioned technology in practical applications is still very far away [9].
Bearing in mind the paradigm of Industry 5.0 (human-centered), holistic methodologies for planning, optimizing, and integrating exoskeletons are being developed (Figure 4) [10].The literature review shows that relatively few publications are devoted to de sion-making processes in planning, designing, and fitting hand exoskeletons, includi 3D-printed ones.This does not change the fact that interest in this group of technolog has been growing for over ten years, not only in terms of clinical but also industrial a plications.Technologically supported functional assessment of the hand and metho and techniques of objective, reliable, and sensitive clinical decision making in this ar are being improved.There are early concepts based on gloves, instrumented stations, sensor systems, sometimes together with vision-based motion capture systems for k ematic and kinetic analysis.But there is a complete lack of standardization in terms requirements, system metrics, and assessment methods, preventing simple compariso between concepts and then solutions based on them.For the above reasons, the path acceptance of the above-mentioned technology in practical applications is still very away [9].
Bearing in mind the paradigm of Industry 5.0 (human-centered), holistic metho ologies for planning, optimizing, and integrating exoskeletons are being develop (Figure 4) [10].Bearing in mind the paradigm of Industry 5.0 (human-centered), holistic methodologies for planning, optimizing, and integrating exoskeletons are being developed (Figure 4) [10].Challenges, and in some cases research gaps, relate to both greater production flexibility and the standardization and integration of exoskeletons within larger user support environments, as well as operational use (Figures 5and 6) [10].Challenges, and in some cases research gaps, relate to both greater production flexibility and the standardization and integration of exoskeletons within larger user support environments, as well as operational use (Figures 5 and 6) [10].Challenges, and in some cases research gaps, relate to both greater production flexibility and the standardization and integration of exoskeletons within larger user support environments, as well as operational use (Figures 5and 6) [10].Maintaining anthropomorphic shapes causes contradictions between the self-weight and the load capacity of the exoskeleton, especially the upper limb, while maintaining low energy consumption [11].The proposed solution fills a research and clinical gap in  Challenges, and in some cases research gaps, relate to both greater production flexibility and the standardization and integration of exoskeletons within larger user support environments, as well as operational use (Figures 5and 6) [10].Maintaining anthropomorphic shapes causes contradictions between the self-weight and the load capacity of the exoskeleton, especially the upper limb, while maintaining low energy consumption [11].The proposed solution fills a research and clinical gap in Maintaining anthropomorphic shapes causes contradictions between the self-weight and the load capacity of the exoskeleton, especially the upper limb, while maintaining low energy consumption [11].The proposed solution fills a research and clinical gap in the area of 3D-printed arm exoskeletons.Its novelty lies in bringing together and presenting a number of detailed technical solutions from the fields of mechanical engineering and computer science that make up a functional hand exoskeleton that ispractical for production and clinical use.In a broader context, the proposed solutions represent a coherent concept for a novel rehabilitation device within a model of automated, personalized, autonomous, daily home rehabilitation for people with specific hand therapy needs, especially as a result of deficits with neurological or neurodegenerative causes.The methods, tools, and processes presented allow for the optimization of production preparation time, costs, and simplification of handling and will improve the accessibility and ease of daily, often arduous, rehabilitation.
The main goal of the article is to check in what direction hand exoskeletons can be developed from a modern economic perspective and how decision support systems can accelerate these processes based on a literature review, expert opinions, and a case study.The article discusses the current development of the design and production of hand exoskeletons and its future directions from the point of view of implementation within the Industry 4.0 and Industry 5.0 paradigms and applications in practice.
It seems that the 3D-printed hand-powered exoskeleton is one of those devices (mechatronic and therefore at the close interface between electronics, computing, and mechanics) that, when implemented, can cause a technological breakthrough in many areas: from prosthetics and rehabilitation of patients with injuries and neurological deficits (strokes, ageing) through the support of healthy factory workers (to Industry 5.0) and the service sector to military, survival, and travel applications.It is also a solution expected in the market.For the aforementioned reasons, interdisciplinary research in this area is in the interest of electronics and serves its development toward even more advanced technologies, in which the hand exoskeleton itself may only be a tool.It should be noted that the main stakeholders of hand exoskeletons are not only their users: (healthy) workers or patients with hand function deficit and their families, but also:

•
Medical professionals who, in current clinical practice, care about maximizing the therapeutic effect of using a hand exoskeleton and its beneficial effect on the patient's daily activities, living, learning, and work; • Business practitioners and engineers who care about improving production, service, and logistics processes; • Asociety that accelerates innovation and partnership for people and the planet.

Material
We used the Delphi method of evaluation based on surveys of 30 experts (15 from the 3D printing/manufacturing area and five from the area of physiotherapy/rehabilitation using robots/exoskeletons).
We used a case study to better illustrate our concept.

Method
The Delphi method was developed in the 1950s by Olaf Helmer and Norman Dalkey.It is characterized by consensus and aggregation of expert opinions, multistage, anonymity of opinions and experts, independence of opinions, avoidance of dominant personalities, controlled feedback, remote, asynchronous, group communication, and statistical elaboration of results.The problem was defined by the survey organizers and the questionnaire consisted of 15 questions.The following survey methods were used: Repetition of the study reduces the extent of the divergence of opinion and leads to an agreed opinion from the majority of experts.The results presented below refer only to responses in which 70% of the experts reached agreement.

Results
This article focuses on current developments in the field of hand exoskeletons and future directions for their development from the point of view of their production and applications.The thrust of this article is the importance of exoskeletons not only in the economy, but also in their social and health importance in seriously ill people, and in healthy athletes, the elderly, and even children.In view of the high needs and lean resources of the healthcare system, exoskeletons, also as part of preventive medicine, can be effectively supported by technical solutions.Man's most precise and versatile tool is his hands, and any deficits in this area can impair his ability to learn, work, communicate socially, or function in society.Restoration of hand function, even partial (i.e., return to maximum achievable functional capacity, but not full function as in a healthy person), is crucial for recovery of activities of daily living (ADLs) and full return to learning and work.This is particularly important when the deficit affects the dominant hand or even both upper limbs.Consequently, scientists, engineers, and clinicians are constantly looking for new and better technological methods to support the stimulated return of function and, if this is not enough, to replace it with robotic devices.In the case of hand function, these are most commonly exoskeletons-robots worn on the hand in the form of gloves or incomplete gloves, allowing sensors to read the user's movement intentions and assisted by exoskeleton actuators.
The use of the hand exoskeleton as a form of therapy (technical therapy support, robot-assisted rehabilitation) can last longer, be more intensive, and its impact is repetitive and objectified.The therapeutic impact can last around the clock, supporting the patient's performance of various activities (dressing, preparing meals, caring for children, studying, or working), but, moreover, traditional tedious exercises can take on a highly functional character, becoming increasingly demanding tasks or even games in which points or medals are won for the patient's activity.The exoskeleton can be both a carrier for the patient's telemedicine systems and a device that monitors and measures the patient's activity, influencing the adjustment of the therapy plan.
Research to develop new hand exoskeletons, as well as more effective ways of using them, is important not only from a scientific and clinical point of view, but also from an economic and social point of view, as it restores normal functioning to patients more quickly.For these reasons, it is important to take a professional look at the role of the hand exoskeleton as a new tool in the therapeutic palette of interventions.
Interdisciplinary research on the design, construction, and clinical efficacy of the hand exoskeleton is lengthy (usually more than 24 months) and costly (requiring a larger research team and multiple stages of industrial, implementation, and clinical research, and in some cases: certification as a medical device according to the Medical Device Regulation (MDR) and ISO 13485.In addition, the number of specialists combining relevant scientific, engineering, and clinical experience is limited, and they cannot participate in multiple projects simultaneously.
The hand exoskeleton solution is becoming more and more popular, both among the scientists and engineers (mechanical engineers, computer scientists, biomedical engineers) who construct it, as well as clinicians and patients themselves.They can be called an expected solution on the market.Nevertheless, the hand exoskeleton is still a supertechnology in the early stages of readiness for implementation (technology readiness level (TRL) lower than 5-6), and the awaited market products are only just entering production.
The main barriers to the wider implementation of hand exoskeletons in clinical applications are the lack of interdisciplinarity in research; the lack of studies on homogeneous patient groups; their use, especially in post-stroke patients (and less so in other patient groups); the low availability of equipment; and the high cost of purchase and operation.
The result of this situation is a greater emphasis on cooperation within interdisciplinary research teams working on effective hand exoskeletons, supported by state institutions and industry associations along the lines of work on space technologies or electric cars.Indeed, just as a lower limb exoskeleton could prove to be a breakthrough technology resulting in a complete shift away from wheelchairs for people with disabilities, a hand exoskeleton could change the situation for many millions of people suffering from hand dysfunction.A number of technologies are currently being developed that could enable the mass production of personalized hand exoskeletons: 3D scanning of the upper limbs, 3D printing from various materials (recyclable plastics, metals, ceramics), and artificial intelligence for control systems or e-health systems, of which hand exoskeletons could be a part.There are also appropriately trained clinical specialists and rehabilitation supply distribution systems who are already involved in fitting patients and servicing various medical devices, sometimes with a very high degree of technological complexity.
Improving current outcomes requires further research into complementing classic therapies with new approaches, including those that can be used with limited medical staff involvement or even without constant supervision.Self-training devices should also be easy to use and intuitive.This would allow for more varied functional exercises and could potentially increase the overall training dose.
The vast majority of publications on hand exoskeletons currently address their use in post-stroke patients.This results in the design of hand exoskeletons being predominantly tailored to hand deficits in the form typically associated with the consequences of stroke, while potentially affecting various systems of the patient's body (e.g., balance) and, thus, aspects of the user's activity other than those affected by isolated deficits in the function of the hand or both hands.Therefore, the experience of implementing a hand exoskeleton in patients following injury or finger/finger amputation may be different.In addition, most post-stroke patients do not return to full function and therefore suffer long-term or even irreversible disability despite the use of a hand exoskeleton.This differs from the situation where the patient uses the exoskeleton temporarily, in the recovery process, only until the lost functions are restored.
The prototype solution should undergo research and evaluation of its clinical usefulness, and only then be introduced for sale, therapy by physiotherapists, and future upgrades with sensory functions.Interdisciplinary teams allow for a broader and deeper consideration of all factors and faster implementation of the engineers' findings into clinical practice.
Despite the rapid development of research into devices designed to maintain and improve hand function, no clear classification has been developed.These devices range in complexity and functionality from support for movement at a single joint to mechanisms that support movements with multiple degrees of freedom at the wrist and all fingers.This makes it difficult to compare the design and effectiveness of individual hand exoskeletons.This requires the development of a common evaluation platform, e.g., based on similar batteries of clinimetric tests.
The results of research on hand exoskeletons to date indicate that they achieve good therapeutic results not only in terms of motor control, but also in a broader context: ensuring independence and preventing secondary motor changes (contractures and others).This makes interdisciplinary research on hand exoskeletons a key study influencing the future lives of patients with hand function deficits and the further work of physiotherapists.The social and economic reach of this impact is very large, and hand exoskeleton research needs to be stimulated, accelerated, and streamlined for the implementation of new products in the market.
Smart technologies to support clinical practice, including at the preventive level, are becoming increasingly available.Their wider dissemination could not only improve the effectiveness of physiotherapy and rehabilitation (also by combining different interventions), but also strengthen their preventive role related to the prediction of risks, injuries, and harmful secondary changes.This would completely change the face of medicine as preventive rather than interventional, keeping the latter role to the most severe and unavoidable cases.

Case Study
As part of the study, a methodology was developed for creating a 3D-printed wrist exoskeleton, and the exoskeleton was built in accordance with the developed methodology (Figures 7 and 8).
The most important issues were the selection of the following: • Criteria for assessing the functionality of the exoskeleton; • Tests: putting on and taking off the exoskeleton, ranges of motion, support forces, opening and closing a healthy or dysfunctional hand with the exoskeleton on, and catching a ball.
It turned out to be necessary to develop our own test stands: • Software compatible with the electronic goniometer; • Test cube with hand dimensioning software, and software for calculating hand parameters; • Software controlling the exoskeleton, including its personalization.
Our own methodology was developed, is presented in the next part of the article, and is compared with other approaches (described in [12,13]), which, however, were not complete from the point of view of making design decisions; i.e., they did not take into account, for example, the mobility of all fingers.
Electronics 2024, 13, x FOR PEER REVIEW 9 of 18 pists.The social and economic reach of this impact is very large, and hand exoskeleton research needs to be stimulated, accelerated, and streamlined for the implementation of new products in the market.Smart technologies to support clinical practice, including at the preventive level, are becoming increasingly available.Their wider dissemination could not only improve the effectiveness of physiotherapy and rehabilitation (also by combining different interventions), but also strengthen their preventive role related to the prediction of risks, injuries, and harmful secondary changes.This would completely change the face of medicine as preventive rather than interventional, keeping the latter role to the most severe and unavoidable cases.

Case Study
As part of the study, a methodology was developed for creating a 3D-printed wrist exoskeleton, and the exoskeleton was built in accordance with the developed methodology (Figures 7 and 8).Electronics 2024, 13, x FOR PEER REVIEW 9 of 18 pists.The social and economic reach of this impact is very large, and hand exoskeleton research needs to be stimulated, accelerated, and streamlined for the implementation of new products in the market.Smart technologies to support clinical practice, including at the preventive level, are becoming increasingly available.Their wider dissemination could not only improve the effectiveness of physiotherapy and rehabilitation (also by combining different interventions), but also strengthen their preventive role related to the prediction of risks, injuries, and harmful secondary changes.This would completely change the face of medicine as preventive rather than interventional, keeping the latter role to the most severe and unavoidable cases.

Case Study
As part of the study, a methodology was developed for creating a 3D-printed wrist exoskeleton, and the exoskeleton was built in accordance with the developed methodology (Figures 7 and 8).

Discussion
Current exoskeleton production analysis showed that in the discussed area of knowledge and experience, data synthesis using a meta-analysis has not been possible so far due to the diversity of research designs, methodologies, and outcome measures.Despite the general confirmation of the effectiveness and desirability of using exoskeletons, more highquality research is still needed to verify the current conclusions [14].There is a consensus among researchers that exoskeletons are designed to strengthen or protect the human body by applying and transmitting force.But an important problem, not only industrial, isthe costly and time-consuming tests of exoskeletons in various device configurations, which causes an increase in interest in the use of AI and digital twins of exoskeletons (devices) and users (people).In this context, the effectiveness of support is only one of the assessed parameters, and problems already occur at the stage of soft tissue modeling or the accuracy of reflecting interactions in relation to reality (e.g., relative movement between the body and the device or force transmission), which is important for personalization [15].Within the Industry 5.0 paradigm, it is also important to examine the impact of exoskeleton rehabilitation and its use in healthy users on quality of life and depression.In general, their positive effect is observed, but these results require further research [16].The effectiveness of solutions is judged by their impact on body function and structure, not on participation or quality of life [17].Increasingly, research on exoskeletons uses physical ergonomics, which monitors potential disturbances related to human-machine cooperation, including based on a combination of wearable sensors and AI.These studies, and then the design, planning, and production of exoskeletons must take into account the limitations of the human body, including helping to define the parameters, characteristics, and thresholds of entire processes [18].In the case of exoskeletons for healthy users used to prevent work-related musculoskeletal disorders, the main considerations are their usability, comfort of use, acceptability, required effort (or reduction), fatigue, improvement of work safety, and impact on work efficiency.Although research results indicate that exoskeletons are not a universal technology, they work better in static activities, and comfort and ease of use remain key [19].Research shows that there is a need to ensure collaboration between disciplines, as the level of stakeholder involvement during exoskeleton development is still low, and there is a lack of methods to quantitatively measure knowledge creation.An evaluation tool for multidisciplinary collaboration using a user-centered design (MCAT) approach has been proposed in this area.It allows a better assessment of the effectiveness, limitations, and barriers associated with multidisciplinary collaboration in exoskeleton development [20].A complex production cycle, complicated construction, selected materials and electronic components, and individual selection result in high production costs.Hence, the proposals for cheap versions of exoskeletons [21].In supporting the movement of the upper limb during everyday activities, soft exoskeletons, thanks to their flexibility, lightness, and comfort, can be an alternative to rigid exoskeletons.Among the identified exoskeletons, more than 80% are dedicated to rehabilitation, assisting, or both.The most commonly used drives include air motors (52%) and cable gear DC motors (29%).Most devices activate 1 (56%) or 2 (28%) degrees of freedom, with the most targeted joints being the elbow and shoulder.Intent detection strategies are implemented in 33% of suits and include the use of switches and buttons, IMUs, stretch and flex sensors, and EMG and EEG measurements.Most devices (75%) reach technology readiness level 4 or 5 [22].In order to minimize the occurrence of adverse events, appropriate selection and preparation of patients is currently mainly used [23].Despite the needs, the methodology for assessing the technical, physical, and psychological aspects of using exoskeletons is not standardized.The differences relate to research goals, analyses, testing procedures, etc.In the case of industrial exoskeletons, men dominate among users.In the case of industrial exoskeletons, this may not only make it difficult to compare systems and their modernization/optimization, but what is worse, it may pose a threat to the health of users [24].It is still difficult to accept the fact that despite the increase in user endurance in the exoskeleton (including lifting and static bending), agility and performance during agility tasks decrease, which translates into moderate user satisfaction.More emphasis should be placed on research in a real environment rather than in a laboratory [25].Currently, it is difficult to verify and compare the readiness levels of different exoskeletons due to the lack of a single recognized methodology for their benchmarking.This hinders the transfer from the research prototype to the actual market product, which results in the unavailability of the product expected on the market in the form of an exoskeleton, despite the growing number of studies.For example, when assessing exoskeletons of the lower limbs, straight gait and kinetic indices are assessed, which is not always reflected in the requirements for gait for everyday activities or human-robot interaction metrics [26].Interestingly, the use of the exoskeleton in stroke patients induces low-or medium-intensity exercise, so the estimated loads on the exoskeleton and its operational wear are rather in the lower range, which may result in longer service intervals [27].
A detailed summary of publications on the use of hand exoskeletons in physiotherapy is provided in Tables 1 and 2.
Table 1.Evidenceof the usefulness of the hand exoskeleton in contemporary physiotherapy.

Source Design Results
Singh et al. 2021 [28] Patients with stroke within 2 years of chronicity undergoing robotic therapy (n = 12) vs. conventional upper-limb rehabilitation (n = 11), intervention of 20 sessions of 45 min each, five days a week for four weeks, Modified Ashworth Scale, Active Range of Motion, Barthel Index, Brunnstrom-stage and Fugl-Meyer scales, neurophysiological measures of cortical-excitability Motor Evoked Potential and Resting Motor threshold were used to assess state of the patients pre-and post-therapy Application of electromechanical roboticexoskeleton for rehabilitation of wrist joint and metacarpophalangeal joint showed improvement in motor outcomes and cortical-excitability in patients with stroke Chen et al. 2016 [29] Clinical trials have been conducted with six patients (study group n = 5, control group n = 1).
Stroke Rehabilitation Assessment of Movement (STREAM) evaluating motor functioning in stroke patients by physiotherapists reveals promising results

Godfrey et al. 2013 [30]
Patients with chronic stroke (n = 9, followed by one excluded from the analysis) completed 18 training sessions and underwent initial assessment, post-assessment, and a 90-day clinical assessment.
The results are promising, but more research is needed on dealing with higher levels of hypertonia and providing greater support to those with higher levels of impairment.
Cisnal et al. 2023 [31] Healthy subjects (n = 18) performed 1min random hand gesture sequences (resting, opening, and closing) in four different conditions resulting from a combination of EMG-based visual feedback and kinesthetic feedback from exoskeleton movement.
EMG-based visual feedback can help hand exoskeleton users learn to use the exoskeleton faster and increase their motivation in therapy.
Zanona et al. 2023 [32] Post-stroke patients (study group n = 23, reference group n = 21) were treated for 1 h, five times a week, for a fortnight.Diagnosis consisted of functional independence measure (FIM), motor activity log (MAL), amount of use (MAL-AOM), quality of movement (MAL-QOM), box and blocks test (BBT), and the Jebsen hand functional test (JHFT).
Brain-computer interface in combination with mental practice and occupational therapy (including exoskeleton activation) promotes sensorimotor recovery of the upper limb and functional independence in post-stroke patients.
Research to date has often been quite narrowly profiled.In the study by Angerhofer et al. [33], brain-computer interface-controlled functional electrical stimulation (FES) and hand exoskeleton therapy were analyzed, as well as their therapeutic and compensatory potential (once the possibility of restoring function with physiotherapy has been exhausted).In a study by Singh et al. [28], the authors focused on neuroplasticity in post-stroke patients and the effect of training with a hand exoskeleton on improving motor performance and cortical excitability as a result of cortical reorganization and plasticity stimulated by use.The above research is much needed, but it can be expected that the physiotherapist in daily clinical practice with the stroke patient will want to focus on using the most effective hand exoskeletons that have been previously developed, clinically proven, and implemented.After all, the sooner a therapeutic intervention is implemented, the better for the patient.Applied research is therefore crucial here.[34] Included in the review is a human hand exoskeleton with 16 actuated joints, four per finger used in rehabilitation training after hand surgery.
The use of EMG signals to control the movement of the hand exoskeleton was found to be useful.
Rojek et al. 2023 [35] The article discusses the state of the art, limitations, and challenges from the point of view of the interdisciplinary hand exoskeleton design team.
The article focuses on the development of a 3D-printed hand exoskeleton solution ready for implementation in the market and clinical practice.
Ourown approach to manufacturing a hand exoskeleton is shown in Figure 9.
The presented methodology for creating personalized hand exoskeletons for healthy people and patients with hand dysfunction works well in practice and allows for the production of personalized hand exoskeletons.Its further development requires research on adapting hand exoskeletons to the needs of individual workplaces (in healthy people) and hand dysfunction (in sick people).Research on the mechanical properties of these exoskeletons may contribute to the development of computer science, mechanical engineering, biomedical engineering, materials engineering, clinical practice, and other sciences (e.g., user's perception of the exoskeleton in psychology and cognitive science).There are many interesting directions in the development of interdisciplinary research: from controlling the hand exoskeleton using a brain-computer interface as part of neuroprosthetics to the concept of a third hand for a person working, e.g., at high altitudes.
Analysis of the literature indicates that existing physiotherapy and exoskeleton-assisted rehabilitation systems are evolving toward more advanced systems based on automation, robotics, and computerization, up to the use of artificial intelligence for control, data analysis, and exercise planning.The results of further research into the application of IoT and intelligent systems in physiotherapy are expected to contribute to more research and further development of robotic physiotherapy devices and, indirectly, to update rehabilitation programs with new opportunities for impact.Added to this is the possibility of remotely monitoring the activity levels of patients (including children and the elderly) in the actual home environment, which may be more conducive to maintaining existing and acquiring new functions.The rise of AI and the areas of application of AI in the manufacture of the hand exoskeleton are shown in Figure 10.The presented methodology for creating personalized hand exoskeletons for healthy people and patients with hand dysfunction works well in practice and allows for the production of personalized hand exoskeletons.Its further development requires research on adapting hand exoskeletons to the needs of individual workplaces (in healthy people) and hand dysfunction (in sick people).Research on the mechanical properties of these exoskeletons may contribute to the development of computer science, mechanical engineering, biomedical engineering, materials engineering, clinical practice, and other sciences (e.g., user's perception of the exoskeleton in psychology and cognitive science).There are many interesting directions in the development of interdisciplinary research: from controlling the hand exoskeleton using a brain-computer interface as part of neuroprosthetics to the concept of a third hand for a person working, e.g., at high altitudes.
Analysis of the literature indicates that existing physiotherapy and exoskeletonassisted rehabilitation systems are evolving toward more advanced systems based on automation, robotics, and computerization, up to the use of artificial intelligence for control, data analysis, and exercise planning.The results of further research into the application of IoT and intelligent systems in physiotherapy are expected to contribute to more research and further development of robotic physiotherapy devices and, indirectly, to update rehabilitation programs with new opportunities for impact.Added to this is the possibility of remotely monitoring the activity levels of patients (including children and the elderly) in the actual home environment, which may be more conducive to maintaining existing and acquiring new functions.The rise of AI and the areas of application of AI in the manufacture of the hand exoskeleton are shown in Figure 10.
ton-assisted rehabilitation systems are evolving toward more advanced systems based on automation, robotics, and computerization, up to the use of artificial intelligence for control, data analysis, and exercise planning.The results of further research into the application of IoT and intelligent systems in physiotherapy are expected to contribute to more research and further development of robotic physiotherapy devices and, indirectly, to update rehabilitation programs with new opportunities for impact.Added to this is the possibility of remotely monitoring the activity levels of patients (including children and the elderly) in the actual home environment, which may be more conducive to maintaining existing and acquiring new functions.The rise of AI and the areas of application of AI in the manufacture of the hand exoskeleton are shown in Figure 10.

Limitations of Own Studies
A limitation is the currently discussed ethical, foreseeable, and usable further development of AI solutions (both simple algorithms and large systems) in the design and manufacture of 3D-printed hand exoskeletons.This directly affects areas of commercialization and business models (i.e., analysis of market, costs and revenues, profitability of developing a product and product family) [36].Although the hand exoskeleton is a product expected in the market, the shallowness of the investor market and the need to implement MDR [37] and ISO 13485 [38] regulations delay implementation activities in this area.

Directions for Further Research and Development
Assistive technologies, including exoskeletons, are used to maintain the health and independence of healthy people and those with deficits (disabled, elderly), especially in the place of living, learning, and working.Industrial production must take into account an ageing population with an increasing likelihood of functional limitations that affect independent living, and adapt the offer to the requirements of the above-mentioned exoskeleton user groups.
Future research should focus on improving the number of good quality studies with similar methodology and comprehensive parametric analysis of the tested exoskeletons, including those related to the Industry 5.0 paradigm:

•
Sustainability and ergonomics of environments, tools, and work organization; • Improving procedures, including signal detection and processing; • Risk assessment support systems, including those based on AI.
Understanding current and future approaches to all stages of exoskeleton design and production becomes crucial.The most important here are the following: planned-use scenarios, type of support, number of degrees of freedom, low-and high-level control (with motion intent detection), technological readiness level, and method of testing the device [22].Future research directions for hand exoskeletons also include the following:

•
Development of clinical guidelines, indications, and contraindications for the use of the hand exoskeleton according to the paradigm of evidence-based medicine, including a combined approach [39] and possible secondary modifications [40][41][42]; • Development and marketing of a commercial hand exoskeleton (with support for all five fingers) following randomized clinical trials on a large group of patients; • Lightweight and environmentally friendly material and manufacturing technologies for personalized medical devices in compliance with the Medical Device Regulations (MDR) and ISO 13485 (quality control and life cycle monitoring of medical products from conception to complete withdrawal from the market); • Advanced artificial intelligence solutions for data analysis and control [43][44][45], includ- ing brain-computer interfaces [46][47][48]; • Use of handheld exoskeletons as part of larger systems, including smart e-health environments, also dedicated to the independence of older people [49][50][51][52][53].
The integration of modern technologies, not only exoskeletons, but also virtual reality, augmented reality, brain-computer interfaces, and neuroprostheses or 3D printing in the development of devices for broadly defined physiotherapy and rehabilitation will ensure intensive, repetitive, and task-oriented exercises, improve patient motivation, and lead to better organized use of the knowledge and experience of the medical staff.A breakthrough could be the system's artificially intelligent ability to generate new exercise sets based on monitoring the exercises already performed by users.This would accelerate the search for sources of progress, and therapeutic success, based on the modification of exercise characteristics.In the event of any problems in the therapy process, it would accelerate their identification and elimination by modifying the exercise program accordingly.Extending the range of sensors and effectors used to the full modifiability of the platform would increase its usability and, in some cases, accuracy in diagnosing the current, and predicting the future, health status of patients.
The results thus presented and their implications in the broadest context can lead to a change in the perception of people with disabilities and an improvement in their independence, and ability to learn and work, which will also affect their financial and social situation, having a strong impact on the economy and society.The word 'disabled person' may take on a different meaning.The ageing of the population may be less severe, at least in a medical context, if older people remain independent for longer.
We also need more advanced analyses of the usefulness of the hand exoskeleton in practical life, not just in everyday life laboratories.Further research is needed into ways of objectively describing hand movement, including automatically.
The number and detail of studies, also in light of the available reviews, show that the evidence to support hand improvement with robotic therapy is preliminary.Nevertheless, studies show improvements in the motor skills of the affected hand and functional improvements in the use of the hand with deficits.This makes this area of research very promising and supports the need for systematic and objective studies in the future.The problem of improving the usability of hand exoskeletons requires interdisciplinary collaboration.There is an urgent need to extend the expertise of teams designing and researching hand exoskeletons toward new effectors, brain-computer interfaces (BCIs) and neuroprostheses, integration with larger systems, including those based on artificial intelligence, and eco-design.Breakthroughs are expected in the areas of control systems for handheld exoskeletons (mainly through wider incorporation of machine learning and the Internet of Things), as well as the technology of their actuators (new, smaller, dedicated actuators with lower energy consumption) and materials, including shape memory and environmentally friendly materials.
The implementation of new IIoT technologies (e.g., digital traceability) will allow for the geolocation of products at all stages of production in the internal supply chain and the optimization of production processes in compliance with the requirements of Industry 4.0/5.0.
The key to decision making is the computational analysis of possible positive and negative scenarios of event development in order to identify an appropriate solution (e.g., technological) and an appropriate plan for its implementation in accordance with the Industry 4.0/5.0paradigm at the production floor level.

Conclusions
Already today, production and maintenance decision making benefits from the advanced sensor infrastructure of Industry 4.0 and AI algorithms that analyze data, predict production runs, and recommend actions to ensure smooth production.This is particularly relevant when manufacturing medical devices with high quality, but with personalized parameters, based on 3D printing technologies.
The therapeutic use of 3D-printed hand exoskeletons has a promising future, both in the areas of functional diagnosis and monitoring, local and remote therapy, and care with secondary injury prevention.The wider use of BCIs will allow alternative ways to control body motility using bioelectrical brain signals.As the technical complexity of physiotherapeutic support is higher with the exoskeleton than before, this requires not only interdisciplinary research, efficient manufacturing processes, and intelligent inference and production management systems, but also thoughtful implementation into daily clinical practice, including within revised comprehensive rehabilitation programs, including poststroke rehabilitation.
It is necessary to develop standard characteristics against which each hand exoskeleton will be evaluated, making it easier to compare different solutions.The main criteria proposed are the number and type of grasps and ranges of movement achieved, the strength developed, the speed of the exoskeleton, the types of sensation, the number and location of sensory zones of the exoskeleton, the ability to achieve a weight similar to the natural limb, and aesthetics.

Figure 1 .
Figure 1.Results of review of publications with the keywords 'exoskeleton' and related.

Figure 2 .
Figure 2. Results of review of publications with the keywords 'hand exoskeleton' and related.

Figure 3 .
Figure 3. Results of review of publications with the keywords "3D print" +"hand exoskeleton" and related.

Figure 1 . 18 Figure 1 .
Figure 1.Results of review of publications with the keywords 'exoskeleton' and related.

Figure 2 .
Figure 2. Results of review of publications with the keywords 'hand exoskeleton' and related.

Figure 3 .
Figure 3. Results of review of publications with the keywords "3D print" +"hand exoskeleton" and related.

Figure 2 .Figure 1 .
Figure 2. Results of review of publications with the keywords 'hand exoskeleton' and related.

Figure 2 .
Figure 2. Results of review of publications with the keywords 'hand exoskeleton' and related.

Figure 3 .
Figure 3. Results of review of publications with the keywords "3D print" +"hand exoskeleton" a related.

Figure 3 .
Figure 3. Results of review of publications with the keywords "3D print" + "hand exoskeleton" and related.

Figure 4 .
Figure 4. Problems to solve in exoskeletonproduction and implementation (own version based on [10]).

Figure 4 .
Figure 4. Problems to solve in exoskeletonproduction and implementation (own version based on [10]).

Figure 4 .
Figure 4. Problems to solve in exoskeletonproduction and implementation (own version based on [10]).

Figure 5 .
Figure 5. Impact of novel approaches to exoskeleton production with the degrees of their current development (own version based on [10]).

Figure 6 .
Figure 6.Proposed stages of implementation of exoskeletons with the degrees of their current development (own version based on [10]).

Figure 5 .
Figure 5. Impact of novel approaches to exoskeleton production with the degrees of their current development (own version based on [10]).

Figure 4 .
Figure 4. Problems to solve in exoskeletonproduction and implementation (own version based on [10]).

Figure 5 .
Figure 5. Impact of novel approaches to exoskeleton production with the degrees of their current development (own version based on [10]).

Figure 6 .
Figure 6.Proposed stages of implementation of exoskeletons with the degrees of their current development (own version based on [10]).

Figure 6 .
Figure 6.Proposed stages of implementation of exoskeletons with the degrees of their current development (own version based on [10]).

Figure 7 .
Figure 7. Three-dimensional-printed mechanical part of the exoskeleton.

Figure 8 .
Figure 8.The control and electronic part of the exoskeleton with actuators.The most important issues were the selection of the following:•Criteria for assessing the functionality of the exoskeleton;•The way they are objectively assessed/measured; • Materials;

Figure 7 .
Figure 7. Three-dimensional-printed mechanical part of the exoskeleton.

Figure 7 .
Figure 7. Three-dimensional-printed mechanical part of the exoskeleton.

Figure 8 .
Figure 8.The control and electronic part of the exoskeleton with actuators.The most important issues were the selection of the following:•Criteria for assessing the functionality of the exoskeleton;•The way they are objectively assessed/measured; • Materials;

Figure 8 .
Figure 8.The control and electronic part of the exoskeleton with actuators.

Figure 10 .
Figure10.Proposed AI applications in exoskeletonproduction with the degrees of their current development[36].

Figure 10 .
Figure10.Proposed AI applications in exoskeletonproduction with the degrees of their current development[36].

Table 2 .
Reviews and meta-analyses concerning the application of hand exoskeletons in physiotherapy.