Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal

for Comprehensive Recovery

Matthias Maszuhn

, Frerk M

uller-Von Aschwege

, Felix Jansen

, Andreas Hein

, Hester Knol

David Snowdon

, Michael Buscherm

ohle

, Dominik Barth

, Luisa Haag

, Nadine Wohlers

Linda R

ude

and Oliver Pieske

OFFIS e. V., Escherweg 2, Oldenburg, Germany

Herodikos GmbH, August-Hanken-Str. 24, Oldenburg, Germany

Hochschule Osnabr

uck, Albrechtstr. 30, Osnabr

uck, Germany

KIZMO, Clinical Innovation Center for Medical Technology Oldenburg, Marie-Curie-Str. 1, Oldenburg, Germany

Dieter Marquardt Medizintechnik GmbH, Robert-Bosch-Str.1, Spaichingen, Germany

Evangelisches Krankenhaus Oldenburg - Klinik f

ur Unfallchirurgie und Orthop

adie, Sporttraumatologie, Steinweg 13-17,

Oldenburg, Germany

Keywords:

Aftercare, Fracture, Injury, Physiotherapy, Mobile Application, Pose Estimation, Neural Network, Load

Measurement.

Abstract:

This paper explores the prevalent challenges associated with musculoskeletal injuries across various demo-

graphics. It proposes the idea for a comprehensive mobile application designed to improve post-fracture

aftercare by addressing existing gaps in information sharing, personalization, and remote care. Comprising

three core components – recording and assessment of physiotherapy exercises, physical load measurement at

the fracture, and a shared documentation tool for all participants involved in the aftercare process – the system

aims to enhance patient compliance and improve recovery outcomes. The system will then be evaluated tech-

nically with healthy subjects to validate the system components. Subsequent usability evaluations will involve

feedback from both healthy subjects and potential end-users, paving the way for planned clinical investigations

with patients undergoing ankle fracture treatments to assess system efﬁcacy, patient-reported outcomes, and

compliance.

1 INTRODUCTION

Musculoskeletal injuries are prevalent among in-

dividuals in Germany aged 65 and above (Fuchs

et al., 2013), often necessitating surgical interven-

tions. These injuries frequently result from falls

among elderly individuals (Kannus et al., 2002) and

require extended recovery periods. Respecting the de-

mographic shift, the frequency of such injuries is ex-

pected to further increase. Not only elderly people

suffer from musculoskeletal injuries, also younger in-

dividuals experience such injuries with an increasing

frequency (Rupp et al., 2021), particularly after acci-

dents, constituting a signiﬁcant cause of work-related

absences (Bundesanstalt f

ur Arbeitsschutz und Ar-

beitsmedizin, 2022). Beyond the impact on the qual-

ity of life, these injuries result in substantial health-

care and socioeconomic costs, including expenses

for surgical procedures, compensation for disability

leave, early retirement, rehabilitation measures, and

reduced productivity.

The quality of post-injury care is an essential fac-

tor for the success of the healing process. A good

quality can encourage a quick relief, full recovery of

the disease and is able to prevent further complica-

tions. The predominant factor inﬂuencing the over-

all expenses in patients with fractures and non-unions

is indirect costs, such as productivity-related losses

(Hak et al., 2014). Post-injury care encounters sev-

eral challenges in the current healthcare system. This

includes for example the insufﬁciently addressing of

prescriptions. Elderly individuals frequently receive

equivalent exercises and repetitions in physical ther-

apy following injuries, similar to younger patients.

(Teng et al., 2012). For a successful and fast recov-

ery of patients after musculoskeletal trauma surgery,

the patient requires individual treatment and therapy

by multiple healthcare professionals including sur-

geons, general physicians and physiotherapists (Fig-

ure 1). For an optimal multi modal, holistic and in-

646

Maszuhn, M., Aschwege, F., Jansen, F., Hein, A., Knol, H., Snowdon, D., Buschermöhle, M., Barth, D., Haag, L., Wohlers, N., Rüde, L. and Pieske, O.

Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal for Comprehensive Recovery.

DOI: 10.5220/0012434000003657

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2024) - Volume 2, pages 646-653

ISBN: 978-989-758-688-0; ISSN: 2184-4305

dividual treatment, information about patients’ recov-

ery should be shared and discussed by the involved

healthcare professionals. However, studies indicate

that there still seems to be a lack of connectivity be-

tween healthcare professionals (Reeves and Zwaren-

stein, 2017). These shortcomings in patient centred

care lead to partially contradictory advice, which can-

not be comprehensively processed by the patients due

to insufﬁcient expertise.

After being discharged from the hospital, surgeons

generally receive limited information about the pa-

tient’s recovery progress unless a subsequent oper-

ation is necessary. Physiotherapists customise ex-

ercise regimes to suit individual patients, however,

they only receive subjective feedback once or twice

a week. To date, there has been no exchange of in-

formation between physiotherapists and other health-

care professionals in Germany to adjust treatment.

If a re-injury takes place or if the process of recov-

ery diverges from expectations, it can prove to be

exceptionally disappointing for all those involved in

the post-injury care process. Insufﬁcient communica-

tion often leads to a failure in documenting the fun-

damental reasons behind such setbacks. Noncompli-

ance with the healthcare professionals’ instructions,

incorrect instructions for speciﬁc patients, or errors

in the initial surgery may all contribute to this uncer-

tainty. Therefore, achieving full recovery from such

injuries hinges on patient compliance with daily exer-

cises, requiring motivation, education, and dedication

of patients and supervision, monitoring and coaching

by health care professionals. Due to the short and

limited therapy sessions, this supervision, coaching

and monitoring is often lacking, fostering insecurity

among both doctors and patients regarding the dura-

bility of the fracture.

An example demonstrating the importance of tai-

lored exercises and supervision during post surgery

treatment of musculoskeletal fractures is the weight-

bearing during recovery. The weight-bearing aspect

within post-care after getting a fracture injury re-

mains a contentious issue, as patients often have only

a rough estimate of the compression they can safely

apply to the affected body region. This is especially

valid for geriatric patients that often struggle to main-

tain the set weight-bearing restrictions (Kammerlan-

der et al., 2018). Commonly used training meth-

ods for weight bearing, for example using house-

hold weight-scales to deﬁne the load, neither show

good short-term nor long-term learning effects of the

prescribed limits (Dabke et al., 2004) (Braun et al.,

2017). Poorly managed weight-bearing during differ-

ent phases of post-injury care can have adverse conse-

quences on the long-term healing process, underscor-

ing the importance of effective communication and

measurement of these parameters.

Another aspect of individual care after muscu-

loskeletal injury is the need for more adaptable op-

tions for participating in physiotherapy. This require-

ment has become less pressing since the outbreak of

the COVID-19 pandemic, as scheduling appointments

proved challenging during times of restricted human

contact. For instance, attending physiotherapy ses-

sions in person often necessitates the assistance of

family members due to the patient’s limited mobility.

In addition, elderly people and patients with jobs re-

quiring traveling or or having an irregular daily work-

ing life beneﬁt from a solution that makes remote

training sessions and exercises possible in an indepen-

dent setting full of ﬂexibility. The reduced availability

of healthcare facilities in rural areas (Kassen

arztliche

Bundesvereinigung Kd

oR, 2022), combined with on-

going shortage of skilled workers (Bundesagentur f

Arbeit, 2023), further fuels the demand for meth-

ods that enable home-based physiotherapy. Nonethe-

less, existing online physiotherapy methods can be

challenging due to the need of therapists to demon-

strate exercises in front of a camera while looking

at the screen at the same time. Moreover, ambula-

tory physiotherapy approaches must be thoughtfully

considered to ensure patient compliance and maintain

the same level of efﬁcacy as in on-site physiother-

apy. ”Its worth noting that correctly executed digital

physiotherapy has the potential to be just as effective

or even more effective than on-site physiotherapy.”

(Lara-Palomo et al., 2022) (Weise et al., 2022).

In summary, we have identiﬁed three primary

challenges within the current post-fracture aftercare

process we aim to address in this research:

1. How can we enhance the effectiveness and adher-

ence of digital physiotherapy as compared to cur-

rent literature?

2. How can we facilitate patient awareness of the

weight-bearing restrictions associated with their

fracture?

3. How can we enhance the documentation process

to facilitate the sharing of pertinent aftercare in-

formation among all stakeholders involved?

The following chapters are structured as follows:

Firstly, we outline our contribution and discuss cur-

rent related literature in section 2. Then we present

our method for addressing the research questions

mentioned above in section 3. Finally, we present

our ﬁndings. This is followed by a description of the

methodology in section 4. Our goal is to conduct ex-

tensive research to determine the effect of our appli-

cation on post-injury recovery. We compare the af-

Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal for Comprehensive Recovery

647

Figure 1: The diagram depicts a common pathway for patients following an ankle fracture and the stakeholders associated

with each stage. Following a hospital stay for up to three days, the patient is referred to a specialist who conducts periodic

examinations and x-rays on the injured area. The patient then undergoes physiotherapy for roughly twelve weeks with varying

phases of load management.

tercare process to conventional practices in section 5,

providing an objective evaluation. In addition, sec-

tion 6 presents a conclusion and outlines potential fu-

ture work.

2 RELATED WORK

Research into the usability of mobile applications

to assist physiotherapy has gained signiﬁcant impor-

tance, particularly in light of the COVID-19 pan-

demic. Kasnakova et al. conducted a pivotal study

demonstrating that the utilization of mobile health ap-

plications can substantially enhance the rehabilitation

process (Kasnakova et al., 2022). Moreover, there

are already various systems available to assist patients

with at-home exercise programs. For instance, Ko-

maris et al. assessed an IMU-based system designed

to ensure the correct execution of physiotherapy exer-

cises during unsupervised treatments (Komaris et al.,

2022). Others have explored intricate 3D motion

capture systems (Mousavi et al., 2020) or offered

disease-speciﬁc information and structured training

schedules to enhance exercise accuracy (Thiengwit-

tayaporn et al., 2023). Weber et al. introduced a

smartphone-assisted training system that fosters the

repetition and consolidation of learned exercises by

facilitating audio-visual communication between pa-

tients and physiotherapists (Weber et al., 2023). A re-

view of current literature reveals a variety of exercise

evaluation tools based on motion tracking or inertial

measurement units (IMU) to assist patients. However,

we propose an experimental approach by employing

a neural network for the evaluation of exercise cor-

rectness and repetitions. Importantly, this approach

would not require a complex setup and is therefore

more suitable for at-home assessments. Our approach

leverages marker-less motion capture technology us-

ing standard 2D cameras, readily available in patients’

smartphones, tablets, and laptops.

In the domain of load measurement systems in

physiotherapy, Braun et al. conducted a survey in

2018 to explore the usage of measurement instru-

ments by physiotherapists in Germany (Braun et al.,

2018). Their ﬁndings indicated that physiotherapists

generally have a favorable view of using measurement

instruments, provided that there is ﬁnancial compen-

sation for acquiring them. Notably, the prevalent ap-

proach in research involves insole measurement sys-

tems, which are already applied for gait, load, and

activity analysis (Braun et al., 2016). By integrating

the measurement system into the patient’s implant, we

aim to eliminate the need for additional tool purchases

and ensure the continuous availability of load data.

Regarding documentation tools for aftercare,

Griefahn et al. conducted a Delphi study to assess

the potential beneﬁts of digitization in the documen-

tation processes of physiotherapists, outlining speciﬁc

HEALTHINF 2024 - 17th International Conference on Health Informatics

648

requirements for such an application (Griefahn et al.,

2020). Building upon this study, our objective is

to create a more standardized documentation process

that actively involves all stakeholders while maintain-

ing a degree of ﬂexibility.

Finally, we would like to emphasize that the in-

tegration of these three components – mobile appli-

cations to support physiotherapy, load measurement

systems, and documentation tools for aftercare – rep-

resents a promising innovation. To the best of our

knowledge, such a combination has not been exten-

sively explored to date. We believe that this holistic

approach has the potential to signiﬁcantly enhance the

efﬁciency and effectiveness of physiotherapeutic care.

3 CONTRIBUTION

In this work, we propose a comprehensive mobile ap-

plication that tackles the aforementioned problems.

Our primary objective is to enable individualized,

self performed and (at least partly) automatically su-

pervised training and rehabilitation exercises. Our

secondary objective is to equip patients with a tool

to monitor their weight-bearing and warn them if

they exceed the set limits. The third objective of

this project is to facilitate seamless data exchange

between physicians, physiotherapists, and patients,

streamlining and individualizing the aftercare jour-

ney. By addressing the existing gaps in information

sharing, personalization, and remote care, our system

aims to not only enhance the quality of post-injury

care but also foster patient compliance, ultimately re-

sulting in improved recovery outcomes.

Our system is structured into three core compo-

nents shown in Figure 2, each offering a dedicated

interface for different users:

1. Recording and Assessment System for Physio-

therapy Exercises: This component serves as a com-

prehensive solution for exercise recording and eval-

uation. Physiotherapists can easily record exercises

using a standard camera, whether a smartphone or

webcam, or select from a pre-deﬁned library of exer-

cises and customise them with individual patient an-

notations. These video recordings are employed to

train a neural network, which then assesses exercise

execution for correctness. Using marker-less motion

tracking technology, the system generates a 3D avatar

demonstrating the exercise on the patient’s tablet or

smartphone. AI algorithms dynamically adjust the

camera angle to provide the most optimal view based

on the speciﬁc exercise. The effectiveness and adher-

ence of the system is then being evaluated.

2. Load Measurement System: This system is

used for real-time monitoring and assessment of the

physical stress on the affected area, and alerts the pa-

tient in the event of overexertion. Healthcare profes-

sionals can access a graphical representation which

illustrates the linkage between physical strain, exer-

cise regimes, and the pain reported by patients. This

integrated view promotes more informed decision-

making concerning the suitability of exercises for

each patient’s distinctive situation.

3. Shared Documentation Tool: A collabora-

tive platform facilitates all stakeholders involved in

the aftercare process to access and update relevant

treatment-related data. Such information encom-

passes data obtained through exercise evaluations and

load measurements. Moreover, the tool functions as

an appointment prompt for patients, providing them

with an overview of their present condition and out-

lining the subsequent steps in their recovery process,

thereby serving a motivational purpose.

4 METHODOLOGY

4.1 Exercise Recording and Assessment

In our primary stage, we will concentrate on set-

ting up the tool for exercising and assessment. Al-

though our application has the potential to cover var-

ious fractures with similar post-injury care regimens,

we have chosen to use an ankle fracture as a repre-

sentative example for exercises and studies. Ankle

fractures are of special interest in this context as the

recommended exercises often involve delicate move-

ments of the leg and foot, which can prove difﬁcult

to identify using motion capture tools. Consequently,

any approach capable of assessing these exercises

must also have the capacity to identify more complex

movements. The aftercare for ankle fractures typi-

cally comprises three phases, each varying in dura-

tion depending on the patient’s progress: the initial

stage involves no weight-bearing, the second stage

introduces partial weight-bearing, and the third stage

permits full weight-bearing adapted to pain levels. In

collaboration with physiotherapists in the team, we

have selected two exercises for each stage, resulting

in a total of six exercises that patients with ankle frac-

tures typically perform during their post-injury recov-

ery program.

Our primary objective in this stage is to display an

avatar demonstrating these exercises on the patient’s

mobile device as seen in Figure 3. Physiotherapists

should have the ability to record their exercises or

select from a library of pre-recorded exercises. To

facilitate exercise recording, we require a motion-

Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal for Comprehensive Recovery

649

Exercise Recording and Assessment

For physiotherapists:

+ Recording videos of exercises

with custom annotations

+ Choosing from pool of

predefined exercises

For patients:

+ Demonstrates exercises by

displaying an avatar on tablet or

smartphone

+ Estimates the patient's pose in

real-time from a video-stream

+ Evaluates repititions and

accuracy of exercises using a

recurrent neural network (RNN)

Physiotherapist

Patient

Surgeon

Clinic doctor /

Medical specialist

Documentation

+ Display of graphs for exercise

assessments and measured physical

load

+ Overview of injury and rehabilitation

process

+ Collection of documents (e.g. privacy

statement, therapy reports)

+ Schedular and appointment reminder

Load Measurement

+ Integrates different sensors to

measure relevant parameters

+ Includes strain gauge monitoring

physical load

+ Transfers data to application via NFC

Figure 2: The graphic gives an overview of the components planned in our application. The physiotherapists get access to an

exercise recording tool which is then used as the basis for the exercise assessments done by the patients. The documentation

tool gathers the data collected by the exercise assessments and the load measurements and offers further functionality such as

a collection of documents and a scheduler and appointment reminder.

Figure 3: The graphic displays the estimated joint coordinates in a 3D space on a scale from -1 to 1 as estimated by MediaPipe.

We used a standard webcam to record a squat here (left side). On the right side, there is an example of the animation for another

recorded exercise that could be enriched with additional information by the physiotherapist and displayed to a patient.

capture tool. We consider two established open-

source libraries, MediaPipe developed by Google and

OpenPose, which is a library developed by a group

of researchers and is free for commercial and non-

commercial use (Cao et al., 2017). Both libraries are

capable of identifying and tracking key body land-

marks in real-time video, such as those obtained from

a 2D camera stream. This technology allows to con-

vert a video stream into a sequence of joint positions

for each frame, with a target frame rate of approxi-

mately 20 images per second to ensure precision in

capturing even subtle movements. First experiments

with MediaPipe have shown that utilizing MediaPipe

for ankle joint detection poses a considerable chal-

lenge as subjects need to be fully captured within the

camera image, introducing potential deviations. This

difﬁculty arises from MediaPipe’s reliance on only

three coordinates in the ankle and feet area, impact-

HEALTHINF 2024 - 17th International Conference on Health Informatics

650

ing the precision of joint detection. Therefore, our

proposed solution involves integrating the pose esti-

mation capabilities of MediaPipe with a straightfor-

ward 2D camera detection of joint angles. This hybrid

approach aims to enhance the reliability of ankle joint

detection and address the limitations associated with

using MediaPipe alone.

By applying inverse kinematics, we can determine

the joint angles required to achieve the desired posi-

tion, creating a ﬂuid animation of the recorded ex-

ercises that can be applied to any character model

with a humanoid skeletal rig. Adobe’s Mixamo pro-

vides a robust platform for character animation and

rigging, offering pre-rigged characters that are free to

use. Our goal is implementing animations for various

3D characters of Mixamo, enabling patients to select

their preferred avatar.

Another important objective is the detection and

assessment of patients’ exercises, primarily focusing

on two metrics: First the number of exercise repe-

titions and second the alignment with the therapist’s

example. To achieve this, we intend to train a neural

network with the joint angles obtained from record-

ings of exercises demonstrated by physiotherapists.

Speciﬁcally, we intend to apply a recurrent neural net-

work (RNN) to identify patterns and repetitions in ex-

ercises. Recognizing that patients and physiothera-

pists may perform exercises with variations or differ-

ent orientations to the camera system, we intend to es-

tablish joint angle limits for each exercise, which will

be included in the training dataset. Rather than seek-

ing exact replication of exercises, we will prioritize

determining whether a movement exceeds joint angle

limits (indicating potential harm) or falls below the

threshold of necessary motion (indicating ineffective-

ness). To gather a substantial amount of data, phys-

iotherapists shall document and classify ten iterations

of each movement.

4.2 Load Measurement

In the subsequent stage, we will introduce a measure-

ment system to assess the physical load on the frac-

ture. Initially, we equip a stabilizing sleeve with dif-

ferent sensors to measure relevant parameters at the

fracture like the physical load or the range of motion.

An integrated sensor will be utilized to closely mon-

itor load levels at frequent intervals. The recorded

data will be seamlessly transferred to a documenta-

tion system and displayed on the patient’s smartphone

or tablet. Our ultimate objective is to integrate a

strain gauge technology into an implant and to com-

bine the measurements of multiple sensors into an

holistic measurement of the load affecting the frac-

ture. However, due to regulatory constraints that ne-

cessitate medical approval for implants, we will also

measure the ground reaction forces using insoles in

our initial studies. We will subsequently compare the

recorded forces obtained from the insoles with those

obtained from our experimental system, allowing us

to assess the performance and accuracy of both mea-

surement methods.

4.3 Documentation

The ﬁnal component of our application will combine

stress measurement data and exercise assessments

into a comprehensive documentation system with sep-

arate views and permissions for doctors, physiothera-

pists and patients. Additionally, the system will in-

corporate essential documents, such as therapy re-

ports and the patient’s privacy statement, ensuring

that these documents only need to be completed once.

Patients will be granted access to a graphical repre-

sentation of exercise history, weight-bearing capabili-

ties of their fracture, and the range of motion in the

affected joint. The app will function as a motiva-

tional tool for patients, inspiring them to adhere to

their workout plans and allowing them to document

personal impressions, including pain levels. This data

will provide physiotherapists with valuable insights

into the patients’ at-home exercise regimens. Fur-

thermore, the tool will help sharing important infor-

mation between healthcare professionals and improve

communication between them. Finally, the docu-

mentation system will feature a scheduler for physio-

therapy sessions, future surgeries, and other appoint-

ments, empowering patients to directly arrange ap-

pointments with doctors and physiotherapists while

receiving timely reminders.

5 STUDY DESIGN

The evaluation of our application will include various

studies to be explained later in this section. The aim

is to analyse the systems prototype and carry out ini-

tial trials in a clinical environment. To examine the

subsystems, a technical evaluation will be conducted

in a descriptive research design with healthy subjects.

For the load measurement system, we need to en-

sure the validity and reliability of the data. There-

fore, we will compare the recordings with a measure-

ment system in the laboratory which we consider to

be the gold standard. The same applies to the motion

tracking system, where we will compare the results of

our single camera approach with MediaPipe or Open-

Pose with a more complex multi-camera system in the

Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal for Comprehensive Recovery

651

laboratory. Following the technical evaluation, we

will continue with a system evaluation including all

the single components in a usability research design.

Healthy subjects representing patients and healthcare

professionals will evaluate the system regarding user-

friendliness and patient safety criteria. Once these

results have been implemented, initial investigations

in a clinical setting are planned. Patients undergoing

open reduction and internal ﬁxation of ankle fractures

(Weber B type; no syndesmotic rupture) will be re-

cruited. Exclusion criteria would be impaired mobil-

ity or gait abnormalities prior to the fracture event,

patients with multiple injuries and patients under 18

years of age. Patient reported outcome measures, gait

parameters, frequency of use and compliance will be

evaluated.

6 CONCLUSION

In conclusion, this paper addresses the pressing chal-

lenges within the realm of post-fracture aftercare, rec-

ognizing the substantial impact of musculoskeletal in-

juries on individuals of all age groups and their asso-

ciated healthcare and socioeconomic costs. The qual-

ity of post-injury care is essential in determining re-

covery outcomes and preventing complications.

Our proposed mobile application, consisting of

three core components, presents a promising solu-

tion to these challenges. By enabling individual-

ized, self-performed exercise routines, facilitating

real-time load measurement, and streamlining data

exchange among patients, physicians, and physiother-

apists, we aim to enhance the overall quality of post-

injury care. The comprehensive methodology out-

lined in this paper provides a clear path for the imple-

mentation of these components, including the inno-

vative use of neural networks for exercise evaluation

and the integration of a strain gauge in an implant for

load measurement.

In future work, special emphasis should be placed

on the continuous reﬁnement of the user interface

and user experience. Ensuring an intuitive and user-

friendly design will be key to maximising patient and

healthcare professional engagement with the mobile

application. Additionally we will investigate an ap-

proach combining load measurement data with coor-

dinates obtained from pose estimation as inputs for

our neural network. This integration has potential to

substantially improve the accuracy of exercise assess-

ments, offering a more comprehensive insight into the

physical capabilities of patients throughout their reha-

bilitation process.

ACKNOWLEDGMENT

This research paper is a part of the THEBEA

project, which has been generously funded by the

”Bundesministerium f

ur Bildung und Forschung”

(BMBF), the German Federal Ministry of Education

and Research. We extend our gratitude to BMBF for

their support in advancing innovative solutions for im-

proving post-fracture aftercare.

REFERENCES

Braun, B. J., Bushuven, E., Hell, R., Veith, N. T.,

Buschbaum, J., Holstein, J. H., and Pohlemann, T.

(2016). A novel tool for continuous fracture aftercare -

clinical feasibility and ﬁrst results of a new telemetric

gait analysis insole. Injury, 47(2):490–494.

Braun, B. J., Veith, N. T., Rollmann, M., Orth, M., Fritz,

T., Herath, S. C., Holstein, J. H., and Pohlemann, T.

(2017). Weight-bearing recommendations after oper-

ative fracture treatment—fact or ﬁction? gait results

with and feasibility of a dynamic, continuous pedo-

barography insole. Int. Orthop., 41(8):1507–1512.

Braun, T., Rieckmann, A., Weber, F., and Gr

uneberg, C.

(2018). Current use of measurement instruments

by physiotherapists working in germany: a cross-

sectional online survey. BMC health services re-

search, 18(1):810.

Bundesagentur f

ur Arbeit (2023). Blickpunkt arbeitsmarkt

– fachkr

afteengpassanalyse 2022. Technical report,

Statistik/Arbeitsmarktberichterstattung, N

urnberg.

Bundesanstalt f

ur Arbeitsschutz und Arbeitsmedizin

(2022). Volkswirtschaftliche kosten durch arbeit-

sunf

ahigkeit 2021. https://www.baua.de/DE/Themen/

Monitoring-Evaluation/Zahlen-Daten-Fakten/pdf/Ko

sten-2021.pdf.

Cao, Z., Simon, T., Wei, S.-E., and Sheikh, Y. (2017). Real-

time multi-person 2d pose estimation using part afﬁn-

ity ﬁelds.

Dabke, H. V., Gupta, S. K., Holt, C. A., O’Callaghan,

P., and Dent, C. M. (2004). How accurate is partial

weightbearing? Clin. Orthop. Relat. Res., 421:282–

286.

Fuchs, J., Rabenberg, M., and Scheidt-Nave, C.

(2013). Pr

avalenz ausgew

ahlter muskuloskelettaler

erkrankungen. Bundesgesundheitsblatt - Gesundheits-

forschung - Gesundheitsschutz, 56(5):678–686.

Griefahn, A., Wolf, E., and Zalpour, C. (2020). Wie kann

die dokumentation in der physiotherapie durch die

digitalisierung effektiver und efﬁzienter werden? –

eine delphi-studie. physioscience, 16(03):101–110.

Hak, D. J., Fitzpatrick, D., Bishop, J. A., Marsh, J. L., Tilp,

S., Schnettler, R., Simpson, H., and Alt, V. (2014).

Delayed union and nonunions: Epidemiology, clinical

issues, and ﬁnancial aspects. Injury, 45:S3–S7.

Kammerlander, C., Pfeufer, D., Lisitano, L. A., Mehaffey,

S., B

ocker, W., and Neuerburg, C. (2018). Inability of

HEALTHINF 2024 - 17th International Conference on Health Informatics

652

older adult patients with hip fracture to maintain post-

operative weight-bearing restrictions. J. Bone Joint

Surg. Am., 100(11):936–941.

Kannus, P., Palvanen, M., Niemi, S., Parkkari, J., and

arvinen, M. (2002). Increasing number and inci-

dence of low-trauma ankle fractures in elderly peo-

ple: Finnish statistics during 1970-2000 and projec-

tions for the future. Bone, 31(3):430–433.

Kasnakova, P., Mihaylova, A., Djurdjev, B., and Tornyova,

B. (2022). Randomized controlled trial of multidis-

ciplinary rehabilitation therapy using mobile applica-

tions in cases of ankle fractures. European journal of

translational myology, 32(2).

Kassen

arztliche Bundesvereinigung Kd

oR (2022).

Regionale verteilung der

Arztinnen und

Arzte in der vertrags

arztlichen versorgung.

https://gesundheitsdaten.kbv.de/cms/html/16402.php.

Komaris, D.-S., Tarfali, G., O’Flynn, B., and Tedesco, S.

(2022). Unsupervised imu-based evaluation of at-

home exercise programmes: a feasibility study. BMC

sports science, medicine & rehabilitation, 14(1):28.

Lara-Palomo, I. C., Gil-Mart

ınez, E., Ram

ırez-Garc

ıa, J. D.,

Capel-Alcaraz, A. M., Garc

ıa-L

opez, H., Castro-

anchez, A. M., and Antequera-Soler, E. (2022).

Efﬁcacy of e-health interventions in patients with

chronic low-back pain: A systematic review with

meta-analysis. Telemed. J. E. Health., 28(12):1734–

1752.

Mousavi, S. H., Hijmans, J. M., Moeini, F., Rajabi, R., Fer-

ber, R., van der Worp, H., and Zwerver, J. (2020). Va-

lidity and reliability of a smartphone motion analysis

app for lower limb kinematics during treadmill run-

ning. Physical therapy in sport : ofﬁcial journal of the

Association of Chartered Physiotherapists in Sports

Medicine, 43:27–35.

Reeves, S, P. F. H. R. G. J. and Zwarenstein, M. (2017).

Interprofessional collaboration to improve profes-

sional practice and healthcare outcomes. Cochrane

Database of Systematic Reviews, (6).

Rupp, M., Walter, N., Pfeifer, C., Lang, S., Kerschbaum,

M., Krutsch, W., Baumann, F., and Alt, V. (2021). The

incidence of fractures among the adult population of

germany–an analysis from 2009 through 2019. Dtsch.

Arztebl. Int., 118(40):665–669.

Teng, K., Eng, C., Hess, C. A., Holt, M. A., Moran, R. T.,

Sharp, R. R., and Traboulsi, E. I. (2012). Building an

innovative model for personalized healthcare. Cleve

Clin J Med, 79(Suppl 1):S1–9.

Thiengwittayaporn, S., Wattanapreechanon, P., Sakon, P.,

Peethong, A., Ratisoontorn, N., Charoenphandhu, N.,

and Charoensiriwath, S. (2023). Development of a

mobile application to improve exercise accuracy and

quality of life in knee osteoarthritis patients: a ran-

domized controlled trial. Archives of Orthopaedic and

Trauma Surgery, 143(2):729–738.

Weber, F., M

uller, C., Bahns, C., Kopkow, C., F

arber, F.,

Gellert, P., Otte, I., Vollmar, H. C., Brannath, W.,

Diederich, F., Kloep, S., Rothgang, H., Dieter, V.,

Krauß, I., Kloek, C., Veenhof, C., Collisi, S., Rep-

schl

ager, U., B

obinger, H., Gr

uneberg, C., Thiel, C.,

and Peschke, D. (2023). Smartphone-assisted training

with education for patients with hip and/or knee os-

teoarthritis (smart-e): study protocol for a multicentre

pragmatic randomized controlled trial. BMC Muscu-

loskeletal Disorders, 24(1):221.

Weise, H., Zenner, B., Schmiedchen, B., Benning, L., Bu-

litta, M., Schmitz, D., and Weise, K. (2022). The ef-

fect of an app-based home exercise program on self-

reported pain intensity in unspeciﬁc and degenera-

tive back pain: Pragmatic open-label randomized con-

trolled trial. J. Med. Internet Res., 24(10):e41899.

Bridging Gaps in Fracture Rehabilitation: A Mobile Solution Proposal for Comprehensive Recovery

653