sensors

Article
Development of an Exoskeleton Platform of the Finger for
Objective Patient Monitoring in Rehabilitation
Nikolas Jakob Wilhelm 1,2, * , Sami Haddadin 2 , Jan Josef Lang 1 , Carina Micheler 1 , Florian Hinterwimmer 1 ,
Anselm Reiners 3 , Rainer Burgkart 1 and Claudio Glowalla 1,4

1 Department of Orthopedics and Sports Orthopedics, Klinikum rechts der Isar, School of Medicine,
80333 Munich, Germany; [email protected] (J.J.L.); [email protected] (C.M.);
[email protected] (F.H.); [email protected] (R.B.); [email protected] (C.G.)
2 Munich Institute of Robotics and Machine Intelligence, Department of Electrical and Computer Engineering,
Technical University of Munich, 80333 Munich, Germany; [email protected]
3 Klinik für Frührehabilitation und Physikalische Medizin, Zentrum für Orthopädie, Unfallchirurgie und
Sportmedizin, München Klinik Bogenhausen, 81925 Munich, Germany; [email protected]
4 Department of Trauma and Orthopedic Surgery, Berufsgenossenschaftliche Unfallklinik Murnau,
82418 Murnau, Germany
* Correspondence: [email protected]; Tel.: +49-4140-7873

Abstract: This paper presents the application of an adaptive exoskeleton for finger rehabilitation.
The system consists of a force-controlled exoskeleton of the finger and wireless coupling to a mobile
application for the rehabilitation of complex regional pain syndrome (CRPS) patients. The exoskeleton
has sensors for motion detection and force control as well as a wireless communication module. The
proposed mobile application allows to interactively control the exoskeleton, store collected patient-
specific data, and motivate the patient for therapy by means of gamification. The exoskeleton was
applied to three CRPS patients over a period of six weeks. We present the design of the exoskeleton,
Citation: Wilhelm, N.J.; Haddadin, S.;
the mobile application with its game content, and the results of the performed preliminary patient
Lang, J.J.; Micheler, C.;
study. The exoskeleton system showed good applicability; recorded data can be used for objective
Hinterwimmer, F.; Reiners, A.;
therapy evaluation.
Burgkart, R.; Glowalla, C.
Development of an Exoskeleton
Keywords: exoskeleton; hand; CRPS; rehabilitation; mobile application; clinical study
Platform of the Finger for Objective
Patient Monitoring in Rehabilitation.
Sensors 2022, 22, 4804. https://
doi.org/10.3390/s22134804
1. Introduction
Academic Editor: Enrico Meli
Patients with impaired hand functioning may experience severe limitations in daily life [1–6].
Received: 24 May 2022 Limited hand mobility can be caused by various factors, including neuromuscular diseases,
Accepted: 23 June 2022 injuries, restricted motor functions from strokes, or age-related limitations [3]. Further, the
Published: 25 June 2022 complex regional pain syndrome can affect hand functionality, as patients are hypersensi-
Publisher’s Note: MDPI stays neutral tive to stimuli or touch and have limitations of movement, such as fist closure or tweezer
with regard to jurisdictional claims in grip [7].
published maps and institutional affil- The main therapeutic goals in CRPS are to maintain or improve hand mobility and
iations. functioning and to reduce pain. Physiotherapy and occupational therapy are two central
pillars of conservative treatments, which aim to compensate for pathological movement
patterns and prevent long-term damage, such as contractures due to insufficient use as a
result of pain avoidance [8]. In this context, the effectiveness of physiotherapy individually
Copyright: © 2022 by the authors. adapted to the patient using mobilization and traction therapy has been proven. The focus
Licensee MDPI, Basel, Switzerland. is on repetitive pain-free movements and training of fine motor skills with and without
This article is an open access article resistance. The duration of physiotherapy or ergotherapy should be 20–30 min a day with
distributed under the terms and
5 therapy sessions per week if possible [9,10].
conditions of the Creative Commons
The conventional method of manual treatment is time-consuming and often leads to
Attribution (CC BY) license (https://
unsatisfactory results, as therapy is often too short, requires an expert therapist, and has
creativecommons.org/licenses/by/
high financial burdens [11,12]. Further, as patient data are not actively collected, patient
4.0/).

Sensors 2022, 22, 4804. https://doi.org/10.3390/s22134804 https://www.mdpi.com/journal/sensors

Sensors 2022, 22, 4804 2 of 12

progress is only measured subjectively. In contrast, exoskeletons can provide accurate

data to track the functional status of the patient’s hand, while not increasing the financial
burden of the treatment over long periods or frequent applications. This enables patients
to follow daily therapy (usually not applicable in manual therapy) and maintain therapy
over a long period of time. However, the use of an exoskeleton system in therapy is
still limited, as only a small subset makes it into clinical testing or practice due to the
complexity and resulting in poor usability in a clinical context [5]. Numerous exoskeletons
have been developed, often requiring a direct fit to the patient’s hand [13–15], or providing
an adaptive actuation [6,16,17]. Often, these concepts do not provide sufficient sensor
applications and lack sufficient force sensing or cannot distinguish between the movements
of MCP, PIP, and DIP joints.
The approach by Dickmann et al. [18] extended the approach by Conti et al. [6] and
provided sufficient sensing for index finger actuation, motion tracking, and force control.
We extended this approach to be suitable for clinical use. A mobile control application
was developed; live tracking of patient data and easy applicability are ensured, making it
possible to simplify the use of the exoskeleton and provide the system as an extension of
conventional therapy, objectifying the patient’s progress through the sensor data collected.
In summary, we make the following contributions:
1. We extended the exoskeleton by Dickmann et al. [18] to incorporate a Bluetooth
remote control and connection to a newly developed mobile application.
2. The application enables adaptive, individual control and is enhanced by two games
to motivate patients for longer therapy sessions and record objective data.
3. We performed a small patient study with three patients and tracked their longitudinal
progress over six weeks; we present the results.
4. We compare the results of the patient to the subject study from Dickmann et al. [18].

2. Materials and Methods

2.1. Mechatronic System
The mechatronic system extends the approach by Dickmann et al. [18] to include a
Bluetooth control unit and improved housing and packaging of all electrical components.
The electrical circuit is shown in Figure 1, including the individual electrical components
and their application for the exoskeleton system.

Figure 1. Overview of the electrical circuit of the extended exoskeleton by Dickmann et al. [18] and
the corresponding use for the exoskeleton system.

The three potentiometers used to measure the angles of the exoskeleton kinematics
are shown in grey. The finger length could be used to calculate the individual joint angles.
Sensors 2022, 22, 4804 3 of 12

The two force-sensing resistor (FSR) units determined the external force applied by the
actuator and are displayed in blue [18]. The motor is powered by an external voltage
source and receives a pulse width modulator (PWM) signal from the microcontroller for
control. The newly integrated DSDTech HM10 Bluetooth module enables external Bluetooth
communication and sends the signals from the three potentiometers and the force signal
with 50 Hz to the mobile application. Conversely, it receives the commands for executing
measurement protocols and actuator trajectories from the application.

2.2. Exoskeleton Framework

The architecture of the exoskeleton system is shown in Figure 2.

Figure 2. Overview of the workflow of the exoskeleton and the app. The exoskeleton from
Dickmann et al. [18] in (a) is connected to the app via Bluetooth and enables games and data acquisi-
tion (b). The received data can be used for therapy or progress assessment (c).

The basis of the developed concept is the exoskeleton by Dickmann et al. [18] for
the index finger. It can be attached to any hand due to the adaptive approach. With the
help of the sensors, all forces and torques in the finger joints and their positions can be
determined [18]. The exoskeleton is shown in Figure 2a. With the help of the therapist,
and in later steps independently, the exoskeleton is applied to the patient. In Figure 2b,
the newly developed application creates an interactive interface between the patient and
the exoskeleton that provides information about the current state of the finger and the
exoskeleton and enables therapy via games. The data collected during this process can then
be reviewed by the therapist or patient and used for therapy objectification and progress
diagnosis (Figure 2c).

2.3. Application Details

Details of the application’s capabilities are shown in Figure 3.
The basis for this is the signal view in Figure 3a, which allows all relevant measurement
signals of the finger to be viewed. These include the angular trajectories of MCP, PIP, and
DIP joint (top view), the corresponding moment curves (middle view), and the measured
force of the force sensor for validation. To validate the measurement signals and to obtain
an intuitive image of the finger with the exoskeleton, a live view was created in Figure 3b.
This allows one to view the finger in its current position based on the solution by Dickmann
Sensors 2022, 22, 4804 4 of 12

et al. An extension is the quasi-static solution of the equations for forces and moments
in the exoskeleton. These are not only solved for the loads within the finger joints, but
extended to the kinematic chain of the exoskeleton, allowing the loads occurring in the
exoskeleton to be represented. Compression rods are dynamically animated and shown
in red and tension rods in green, as shown in Figure 3b. Details of the kinematic solution
and the quasi-static dynamic implementations can be found in the GitHub repository
https://github.com/NikonPic/ExoApp (accessed on 20 May 2022).

Figure 3. Application details. The general signals of the exoskeleton are displayed in (a) and the
actual position of the exoskeleton in (b). To motivate the patient, the games “bubble collector” in (c)
and “dodge rectangles” in (d) are displayed.

To motivate the patient to participate in therapy, two different games were developed.
In the first game, “Bubble Collector”, the pure movement of the fingers is intended to
motivate and increase mobility. The live view of the exoskeleton serves as the basis for this
game. The game engine generates random, slowly growing “soap bubbles” around the
patient’s fingertip and displays them on the screen (see Figure 3c). As soon as the patient
touches a soap bubble with the fingertip, it bursts and gives points indirectly proportional
to its size. Thus, bursting smaller and harder-to-reach soap bubbles is rewarded with a
higher score. The second game focuses on the patient’s dexterity, works specifically with the
patient’s force feedback, and is shown in Figure 3d. The goal of the game is for the orange
ball to survive as long as possible without colliding with a rectangle. The rectangles are
randomly generated, move horizontally, and become faster over time. The movement of the
orange ball is directly coupled with the force feedback of the exoskeleton and corresponds
to the acceleration of the ball up or down.
A major advantage of the Festo-based kinematics is the adaptivity of the system,
which can be applied to different hand sizes [6]. Only a few parameters are required from
the patient, such as the respective lengths of the three phalanges of the index finger. The
application provides a separate input mask for this purpose, which must be filled out once
during a new registration. The patient-specific profile is then saved and can be edited at
any time. In addition, the patient receives an overview of all his therapy sessions and
the associated measurements. If the patient’s parameters change, the measurements are
adjusted accordingly.

2.4. Application Games in the Context of Hand Rehabilitation

The goals of these developed games are to implement established concepts of classical
physiotherapy and ergotherapy. The games offer the possibility to train the fine motor
skills of the hand with and without resistance. The visualization and abstraction of the
Sensors 2022, 22, 4804 5 of 12

real patient’s finger onto a virtual model in the video game corresponds to a digital mirror
therapy and is intended to utilize the therapeutic benefit of analog mirror therapy, which
has been demonstrated in acute CRPS in controlled studies [19,20]. A similar approach is
also taken by the “graded motor imagery” therapy concept in the classic CRPS treatment
in which the imagination of movements is combined with mirror therapy and has shown
very good therapeutic success [21,22]. In addition, the game-like and motivating character
and the self-controlled exercise by the patient encourages a degree of movement at which
certain pains are tolerated by the patient. This therapy concept is applied in the CRPS
therapy under the concept of in vivo exposure (EXP), in which the patient intentionally
accepts pain in order to achieve an improvement of the therapy success (no pain, no gain)
and stands in contrast to conventional physiotherapy, in which pain is to be avoided during
exercise (no gain with pain) [23].

2.5. Implementation of Clinical Studies

The study protocol was designed to treat patients with the exoskeleton during out-
patient rehabilitation in addition to conventional physiotherapy. Six study sessions, each
lasting 30 min, were conducted over 6 weeks. Every study session was carried out accord-
ing to the defined protocol. Medical history was initially taken, including demographic
data and the triggering event of CRPS. CRPS was classified using the Budapest diagnostic
criteria [24], and the patient’s functional limitations in daily life, work, and sports were
documented using the validated Quick-DASH score [25]. In addition, associated measure-
ments were performed on the hands to assess other objectifiable parameters related to the
progression of CRPS, which are shown in Figure 4.

Figure 4. Used measurement tools for additional patient monitoring. Displayed are (I) the tempera-
ture by Seek Thermal CompactPRO, (II) pulse oximeter (PULOX PO-200), and (III) skin conductance.

Before and after each session, the (I) temperature of the hands with a thermal imaging
camera (Seek Thermal CompactPRO), the (II) O2 saturation/circulation of the fingers
with the pulse/heart-rate using a pulse oximeter (PULOX PO-200), and the (III) skin
conductance with an impedance device (Mindfield eSense Skin Response) were measured
and documented in a side-by-side comparison. The absolute values of the parameters can
be used as indicators of the patient’s stress level and, as progression parameters, they can
provide indications of the progression of the disease [26,27].
Subsequently, the range of motion, trajectory, and force of the hand parameters were
measured via the applied exoskeleton with and without the motor. Then, the control
games “Bubble Collector” and “Dodge Rectangles”, as well as the Leap Motion games,
were performed over 20 min. Finally, the hand parameters and the accompanying mea-
surements were performed again and documented. At the finalization of the experimental
sessions after 6 weeks, the Budapest diagnostic criteria and the Quick-DASH scores were
additionally collected a second time.

3. Results
3.1. Demographics
The presented exoskeleton was used to treat three CRPS patients over a period of
6 weeks according to the study protocol. These were two women and one man. In two
Sensors 2022, 22, 4804 6 of 12

patients, the CRPS was related to a fracture and a surgical procedure on the affected limb.
In one patient, CRPS occurred as a result of overuse of the hand while writing a graduate
thesis. The demographic data are summarized in Table 1. Each patient received at least
6 sessions of 30 min of treatment. During the first four weeks, one to two sessions were
conducted per week, and the final treatment was given at the six-week follow-up. Patients
were highly motivated and reported very pleasant session experiences. No complications
or adverse events occurred during or following the trial sessions.

Table 1. Demographics of patients.

Patient Age Sex BMI Duration CRPS Trigger CRPS

Hand overload
1 26 female 32.1 kg/m2 5 years
when writing
2 68 female 28.3 kg/m2 7 months Elbow fracture
3 44 male 21.3 kg/m2 8 months Wrist fracture

Patients were wearing the exoskeleton two times a week over a period of six weeks.

3.2. Accompanying Measurements

The accompanying measurements collected before and after each session (temperature
of the hands, O2 saturation of the finger, heart rate, and skin conductance) were documented
and evaluated. It was found that the individual parameters did not show any significant
change in correlation to the individual treatment or the follow-up. A summary of the
average values obtained is shown in Table 2.

Table 2. Overview of accompanying measurements of all patients in the six-week follow-up.

Patient Temperature O2-Saturation Heart Rate Skin Conductivity

1 pre 37 °C 90% 91 min 0.78 µS
1 post 37 °C 99% 82 min 0.34 µS
2 pre 38 °C 96% 58 min 0.38 µS
2 post 38 °C 96% 62 min 0.09 µS
3 pre 39 °C 97% 64 min 0.64 µS
3 post 36 °C 99% 60 min 0.33 µS

3.3. Patient-Reported Outcome Measures (PROMS)

The evaluation of the recorded PROM and the subjective feelings of the patients
showed improvements in the complaints and the functions of all three patients. The
Budapest criterion (as a sign of the severity of CRPS) decreased in all patients. In addition,
the evaluation of the DASH score showed improvement in all areas. The results of the
patient follow-up study are shown in Table 3.

Table 3. Overview of patient follow-up according to QuickDASH and the Budapest score over a
follow-up period of six weeks. In each case, a lower score corresponds to a better patient condition.

Score Patient 1 Patient 2 Patient 3

Budapest Pre [28] 6/11 5/11 7/11
Budapest Post [28] 5/11 1/11 2/11
QuickDASH Pre [29] 50% 43% 40%
QuickDASH Post [29] 45% 38% 38%

3.4. Exoskeleton Measurements

Further, we present the recordings of the joint and torque curves with the exoskeleton
before and after the six weeks of therapy by the physical therapist. The results of this
Sensors 2022, 22, 4804 7 of 12

follow-up, as well as details about each patient wearing the exoskeleton, are shown in
Figure 5.

Figure 5. Comparative study of the exoskeleton of patient 1 (blue, top), patient 2 (green, middle),
and patient 3 (orange, bottom) over a six-week period. On the left are the patient’s hands wearing
the exoskeleton. On the right are the joint curves of the MCP, PIP, and DIP joints (top row) and
the corresponding torque curves (bottom row). The measurements shown correspond to the mean
(line) ± standard deviation (shaded) of five individual measurements.

The hands of the patients with the exoskeleton and the corresponding joints and
moment curves are shown. The joint curves are displayed for the MCP, PIP, and DIP joints
before and after the 6 weeks. The presented motion was generated by the linear motor of
the exoskeleton, and the patient’s finger followed the flexion provided by the exoskeleton
without active muscle actuation.
The angular measurements resulted from solving the exoskeleton’s kinematic system
with the patient’s parameters. The force measurements also allowed the calculations of
all corresponding torque curves of the finger joints. The torque curves are the quasi-static
solutions of the loading equations [18]. They are shown under each of the joint curves.
The curves shown and the corresponding shaded areas around them represent the mean
Sensors 2022, 22, 4804 8 of 12

and standard deviations over five measured trajectories. In patient 2, a complete trajectory
could not be recorded prior to therapy because a predefined force limit of 6 N was exceeded.

4. Discussion
4.1. Applicability of the Exoskeleton System
The platform, consisting of the exoskeleton and mobile applications, was successfully
applied to patients. The combination of the wearing comfort of the exoskeleton and the
ease of use of the application enabled patients to use the system easily and intuitively. The
storage of individual user profiles and the input masks of individual patient parameters
allowed the measurements and life views to be optimized for the patient. The control
system successfully responded to existing force limits, as shown in Figure 5 for patient 2
(green). Exceeding the force limit of 6 N stopped the movement and the finger was
immediately unloaded.

4.2. Applicability of Gamification in Rehabilitation

The presented games could be successfully applied to the patient and formed the
temporally largest part of the therapy session supported by the exoskeleton. The game
“Bubble Collector” from Figure 3c has a similar differentiation to the real hand due to its
similarity to the already established analog mirror therapy [19,20] and motivates an active
movement of the hand. The game “Dodge Rectangles” from Figure 3d is controlled by the
patient’s use of force. Thus, the patient can specifically control the maximum load during
therapy within the limits of his/her pain tolerance and the patient is also motivated to
allow higher forces in the interest of the success of the game.

4.3. Evaluation of Exoskeleton Measurements

The longitudinal study, which shows the progress of patients over a period of six
weeks, contains remarkable observations. First, it can be observed that all three patients
had positive therapeutic courses, as both QuickDASH [29] and Budapest [28] scores fell.
This finding is also reflected in the torque curves from Figure 5. For patient 2, the trajectory
could then be applied to the exoskeleton, as the stiffness of the finger was reduced and,
therefore, the external force of the exoskeleton remained below 6 N during the trajectory.
The joint torque curves (of the MCP joints) for patients 1 and 3 had significantly flatter
values at the second 10 (point of maximum flexion) within the movement trajectory; this
also applied (to a somewhat lesser extent) to the joint torque curves of PIP and DIP.
The joint angle trajectories of the three patients differed greatly from each other and
also changed during therapy. The reason for the high variance between the patients was the
adaptivity of the exoskeleton system. The externally applied force of the linear motor seeks
the path of least resistance in the finger. Therefore, a patient with a relatively high joint
moment in a finger joint has a rather low joint angle amplitude. The relationship between
the joint angles and moment curves can be illustrated by patient 3 (orange) in Figure 5.
Here, the joint moment curve of the MCP joint showed a significantly flatter course after
the six weeks. As a result, the joint also exhibited less resistance to motion and the joint
angle of patient 3 moved earlier in the trajectory than at the beginning of therapy. This also
changed the joint angle curve of the PIP joint, as it no longer had to compensate for the
reduced movement of the MCP joint and had a lower joint angle amplitude.

4.4. Comparison to Healthy Subjects

To put the measured patient data in context, the obtained results of the performed
patient study are compared with the measurements from the healthy subject study of
Dickmann et al. [18]. Both studies were conducted under the same conditions (the linear
motor of the exoskeleton defined the fixed flexion motion). The comparative study of
healthy subjects and patient tests are shown in Figure 6.
Sensors 2022, 22, 4804 9 of 12

The top row shows the individual angular trajectories of the MCP, PIP, and DIP joints,
and the bottom row displays the corresponding torque curves for all three subjects on the
left and all three patients on the right.

Figure 6. Comparative study between the three healthy subjects from Dickmann et al. [18] on the
left and the three patients with CRPS on the right before therapy. The angle trajectories of MCP,
PIP, and DIP joints are on the top row and the respective torque trajectories are on the bottom.
The measurements shown correspond to the mean (line) ± standard deviations (shaded) of five
individual measurements.

When comparing the subject study by Dickmann et al. [18] and the presented patient
study in Figure 6, several observations can be made. Because the exoskeleton was only
allowed to actively apply an external force of 6 N to the patient, the trajectory could not
initially be applied to patient 2 because the finger was too stiff at the beginning of the test.
In addition, the torque curves of the healthy subjects and patients differed considerably.
Patients consistently showed higher amplitudes in the joint torque curve. This can primarily
be explained by the higher stiffness of the patient joints, which induced higher torques in
the motion sequence.

4.5. Limitations
There are several limitations to our study. The first limitation involved the small
number of patients and the short treatment phase with limited follow-up. This was due to
the rare but severe clinical picture, in which the inclusion of patients in a clinical trial was
significantly more difficult. In addition, regarding new medical device regulations (MDRs),
the approval of a new medical product is considerably more difficult, so our study was
only approved as a feasibility study on a small number of patients. However, the results
are encouraging and should lead to an extension of the exoskeleton to the other three long
fingers and enable us to extend our studies to more patients.
The second limitation was the insufficient reliability of the accompanying measure-
ments. The accompanying measurements were performed to objectify the clinical course
of CRPS using additional parameters. In the literature, temperature, O2 saturation, heart
rate, and skin conductance are used to determine the influence of the sympathetic nervous
system [26,27]. Skin conductance is often used to determine the phasic response (0.1–1 µS,
event correlating value in the baseline difference) over the tonic level (2–20 µS, mean of
absolute values over a long period of time) [30]. In our measurements, no correlation
to the clinical course could be observed due to a very wide dispersion of the absolute
values. The other parameters (temperature, O2 saturation, heart rate) were also influenced
by environmental factors rather than by the test; we recommend the use of exoskeleton
measurements to objectify the course of the disease for further studies. One solution for
improved accompanying measurements would be to include more patients in the study and
track data over a longer period of time in order to gather enough data for a more detailed
Sensors 2022, 22, 4804 10 of 12

analysis. Further, the upgrade to a more sophisticated skin conductivity measurement

system is recommended.

5. Conclusions
The development of the mobile application and the integration and application on the
patient could be successfully implemented and, thus, the exoskeleton of Dickmann et al. [18]
could be extended. The operability with Bluetooth made it possible to flexibly and easily use
the exoskeleton. The comparative study between the subject study of Dickmann et al. [18]
and the conducted patient study showed clear differences in the torque curves for the pa-
tients, which could be attributed to the increased stiffness of the finger joints. The follow-up
study showed a positive trend in QuickDASH and Budapest scores. This trend was also
demonstrated with the exoskeleton, although it should be noted as a limitation that the
number of patients was small. Furthermore, the applied system is not limited to CRPS by
default, but can also be applied to different rehabilitation scenarios, such as strokes, and
can be used as an extension of existing rehabilitation devices, such as the SPIDER system
by Glowinski and Blazejewski [31]. The future goals of this project will be to extend the
exoskeleton to the whole hand as well as extend the application.

Author Contributions: Conceptualization, N.J.W., R.B. and C.G.; methodology, N.J.W.; software,
N.J.W.; validation, N.J.W., R.B., C.G., C.M., J.J.L. and F.H.; formal analysis, N.J.W., R.B. and C.G.;
investigation, N.J.W.; resources, R.B. and A.R.; data curation, N.J.W.; writing—original draft prepara-
tion, N.J.W.; writing—review and editing, N.J.W., C.M., J.J.L., F.H., R.B., S.H. and C.G.; visualization,
N.J.W.; supervision, R.B. and S.H.; project administration, R.B.; funding acquisition, R.B. All authors
have read and agreed to the published version of the manuscript.
Funding: This work was supported by the “Deutsche Forschungsgemeinschaft (DFG)” within the
interdisciplinary project “Robotisch unterstützte Rehabilitation der Funktion der menschlichen Hand
unter Verwendung eines dynamischen Modells beim komplexen regionalen Schmerzsyndrom (CRPS)”
(BU 1154/8-1, Projektnummer 276036034).
Institutional Review Board Statement: The study was conducted according to the guidelines of the
Declaration of Helsinki, and approved by the Institutional Review Board of Technische Universität
München (protocol code 501/21 S-KH, 10 November 2021).
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The data presented in this study are openly available and can be found
in the GitHub repository https://github.com/NikonPic/ExoEval (accessed on 20 May 2022).
Acknowledgments: The authors would like to thank Univ. med. Rüdiger von Eisenhart-Rothe for
providing us with the opportunity to work in orthopaedic research.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations
The following abbreviations are used in this manuscript:

CRPS Complex regional pain syndrome

MCP metacarpophalangeal joint
PIP proximal interphalangeal joint
DIP distal interphalangeal joint
MCU microcontroller unit
PWM pulse-width modulation
CPU central processing unit
ADC analog-to-digital converter
FSR force-sensing resistor
PROMS patient-reported outcome measures
MDR medical device regulation
Sensors 2022, 22, 4804 11 of 12

References
1. Birch, B.; Haslam, E.; Heerah, I.; Dechev, N.; Park, E.J. Design of a Continuous Passive and Active Motion Device for Hand
Rehabilitation. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society,
Vancouver, BC, Canada, 20–24 August 2008; Volume 2008, pp. 4306–4309. [CrossRef]
2. Wang, J.; Li, J.; Zhang, Y.; Wang, S. Design of an exoskeleton for index finger rehabilitation. In Proceedings of the 31st Annual
International Conference of the IEEE Engineering in Medicine and Biology Society, Minneapolis, MN, USA, 3–6 September 2009;
pp. 5957–5960. [CrossRef]
3. Heo, P.; Gu, G.M.; Lee, S.J.; Rhee, K.; Kim, J. Current hand exoskeleton technologies for rehabilitation and assistive engineering.
Int. J. Precis. Eng. Manuf. 2012, 13, 807–824. [CrossRef]
4. Cempini, M.; Cortese, M.; Vitiello, N. A Powered Finger–Thumb Wearable Hand Exoskeleton With Self-Aligning Joint Axes.
IEEE/ASME Trans. Mechatron. 2015, 20, 705–716. [CrossRef]
5. Yue, Z.; Zhang, X.; Wang, J. Hand Rehabilitation Robotics on Poststroke Motor Recovery. Behav. Neurol. 2017, 2017, 3908135.
[CrossRef] [PubMed]
6. Conti, R.; Meli, E.; Ridolfi, A.; Bianchi, M.; Governi, L.; Volpe, Y.; Allotta, B. Kinematic synthesis and testing of a new portable
hand exoskeleton. Meccanica 2017, 52, 2873–2897. [CrossRef]
7. Maihöfner, C.; Seifert, F.; Markovic, K. Complex regional pain syndromes: new pathophysiological concepts and therapies. Eur. J.
Neurol. 2010, 17, 649–660. [CrossRef]
8. Barnhoorn, K.J.; van de Meent, H.; van Dongen, R.T.M.; Klomp, F.P.; Groenewoud, H.; Samwel, H.; Nijhuis-van der Sanden,
M.W.G.; Frölke, J.P.M.; Staal, J.B. Pain exposure physical therapy (PEPT) compared to conventional treatment in complex regional
pain syndrome type 1: A randomised controlled trial. BMJ Open 2015, 5, e008283. [CrossRef]
9. Oerlemans, H.M.; Oostendorp, R.A.; de Boo, T.; Goris, R.J. Pain and reduced mobility in complex regional pain syndrome I:
Outcome of a prospective randomised controlled clinical trial of adjuvant physical therapy versus occupational therapy. Pain
1999, 83, 77–83. [CrossRef]
10. Oerlemans, H.M.; Oostendorp, R.A.; de Boo, T.; van der Laan, L.; Severens, J.L.; Goris, J.A. Adjuvant physical therapy versus
occupational therapy in patients with reflex sympathetic dystrophy/complex regional pain syndrome type I. Arch. Phys. Med.
Rehabil. 2000, 81, 49–56. [CrossRef]
11. Epstein, D.; Mason, A.; Manca, A. The hospital costs of care for stroke in nine European countries. Health Econ. 2008, 17, S21–S31.
[CrossRef]
12. Elsamadicy, A.A.; Yang, S.; Sergesketter, A.R.; Ashraf, B.; Charalambous, L.; Kemeny, H.; Ejikeme, T.; Ren, X.; Pagadala, P.;
Parente, B.; et al. Prevalence and Cost Analysis of Complex Regional Pain Syndrome (CRPS): A Role for Neuromodulation.
Neuromodul. Technol. Neural Interface 2017, 21, 423–430. [CrossRef]
13. Wege, A.; Kondak, K.; Hommel, G. Mechanical design and motion control of a hand exoskeleton for rehabilitation. In Proceedings
of the IEEE International Conference Mechatronics and Automation, Niagara Falls, ON, Canada, 20 July–1 August 2005; Volume 1,
pp. 155–159.
14. Chiri, A.; Giovacchini, F.; Vitiello, N.; Cattin, E.; Roccella, S.; Vecchi, F.; Carrozza, M.C. HANDEXOS: Towards an exoskeleton
device for the rehabilitation of the hand. In Proceedings of the 2009 IEEE/RSJ International Conference on Intelligent Robots and
Systems, St. Louis, MO, USA, 10–15 October 2009; IEEE: Piscataway, NJ, USA, 2009; pp. 1106–1111. [CrossRef]
15. Lee, J.; Lee, M.; Bae, J. Development of a Hand Exoskeleton System for Quantitative Analysis of Hand Functions. J. Bionic. Eng.
2018, 15, 783–794. [CrossRef]
16. Iqbal, J.; Khan, H.; Tsagarakis, N.G.; Caldwell, D.G. A novel exoskeleton robotic system for hand rehabilitation – Conceptualization
to prototyping. Biocybern. Biomed. Eng. 2014, 34, 79–89. [CrossRef]
17. Sarac, M.; Solazzi, M.; Sotgiu, E.; Bergamasco, M.; Frisoli, A. Design and kinematic optimization of a novel underactuated robotic
hand exoskeleton. Meccanica 2016, 52, 749–761. [CrossRef]
18. Dickmann, T.; Wilhelm, N.J.; Glowalla, C.; Haddadin, S.; van der Smagt, P.; Burgkart, R. An Adaptive Mechatronic Exoskeleton
for Force-Controlled Finger Rehabilitation. Front. Robot. 2021, 8, 314. [CrossRef] [PubMed]
19. McCabe, C.S.; Haigh, R.C.; Ring, E.F.J.; Halligan, P.W.; Wall, P.D.; Blake, D.R. A controlled pilot study of the utility of mirror
visual feedback in the treatment of complex regional pain syndrome (type 1). Rheumatology 2003, 42, 97–101. [CrossRef] [PubMed]
20. Cacchio, A.; De Blasis, E.; Necozione, S.; di Orio, F.; Santilli, V. Mirror therapy for chronic complex regional pain syndrome type 1
and stroke. N. Engl. J. Med. 2009, 361, 634–636. [CrossRef] [PubMed]
21. Moseley, G.L. Graded motor imagery is effective for long-standing complex regional pain syndrome: a randomised controlled
trial. Pain 2004, 108, 192–198. [CrossRef]
22. Moseley, G.L. Graded motor imagery for pathologic pain: A randomized controlled trial. Neurology 2006, 67, 2129–2134.
[CrossRef]
23. den Hollander, M.; Goossens, M.; de Jong, J.; Ruijgrok, J.; Oosterhof, J.; Onghena, P.; Smeets, R.; Vlaeyen, J.W.S. Expose or protect?
A randomized controlled trial of exposure in vivo vs pain-contingent treatment as usual in patients with complex regional pain
syndrome type 1. Pain 2016, 157, 2318–2329. [CrossRef]
24. Harden, N.R.; Bruehl, S.; Perez, R.S.G.M.; Birklein, F.; Marinus, J.; Maihofner, C.; Lubenow, T.; Buvanendran, A.; Mackey, S.;
Graciosa, J.; et al. Validation of proposed diagnostic criteria (the “Budapest Criteria”) for Complex Regional Pain Syndrome. Pain
2010, 150, 268–274. [CrossRef]
Sensors 2022, 22, 4804 12 of 12

25. Beaton, D.E.; Wright, J.G.; Katz, J.N.; Upper Extremity Collaborative Group. Development of the QuickDASH: comparison of
three item-reduction approaches. J. Bone Jt. Surg. Am. 2005, 87, 1038–1046.
26. Hansen, F.; Sawatzky, J.A.V. Stress in Patients With Lung Cancer: A Human Response to Illness. Oncol. Nurs. Forum 2008,
35, 217–223. [CrossRef] [PubMed]
27. Appelhans, B.M.; Luecken, L.J. Heart Rate Variability as an Index of Regulated Emotional Responding. Rev. Gen. Psychol. 2006,
10, 229–240. [CrossRef]
28. Birklein, F.; Dimova, V. Complex regional pain syndrome–up-to-date. Pain Rep. 2017, 2, e624. [CrossRef] [PubMed]
29. Michalos, A.C. (Ed.) Quick DASH, Questionnaire (13-Item Short Version). In Encyclopedia of Quality of Life and Well-Being Research;
Springer: Dordrecht, The Netherlands, 2014; pp. 5382–5382. [CrossRef]
30. Dawson, M.E.; Schell, A.M.; Filion, D.L.; Berntson, G.G. The Electrodermal System. In Handbook of Psychophysiology; Cacioppo,
J.T., Tassinary, L.G., Berntson, G., Eds.; Cambridge University Press: Cambridge, UK, 2000; pp. 157–181.
31. Glowinski, S.; Blazejewski, A. SPIDER as A Rehabilitation Tool for Patients with Neurological Disabilities: The Preliminary
Research. J. Pers. Med. 2020, 10, 33. [CrossRef] [PubMed]