Real-Time 3D Posture Tracking for Surgeons in Pediatric Minimally

Invasive Surgery

Pietro Neroni

1 a

, Giuseppe Caggianese

1 b

, Ciro Esposito

2 c

and Luigi Gallo

1,3 d

Institute for High Performance Computing and Networking, National Research Council of Italy,

Via Pietro Castellino 111, 80131, Naples, Italy

Azienda Ospedaliera Universitaria Federico II, Via Sergio Pansini 5, 80131, Naples, Italy

Department of Sciences and Information Technologies, Pegaso University, Piazza Trieste e Trento 48, 80132, Naples, Italy

Keywords:

Posture Monitoring, Depth Camera Tracking, Musculoskeletal Disorders, Ergonomics, Surgical Training,

Pediatric Surgery.

Abstract:

Minimally invasive pediatric surgery presents ergonomic challenges that signiﬁcantly increase the risk of mus-

culoskeletal disorders (MSDs) among surgeons due to prolonged periods of suboptimal posture. This study

introduces a real-time posture monitoring and correction system designed to address this issue. The system uti-

lizes depth camera technology, interactive feedback mechanisms, advanced skeletal tracking, and ergonomic

assessment algorithms to continuously monitor and evaluate surgeons’ posture. Through rapid data processing,

the system provides real-time feedback, enabling immediate posture adjustments during surgical procedures.

It delivers non-intrusive alerts to inform medical staff when incorrect postures are detected, thereby promoting

ergonomic well-being and reducing the incidence of MSDs. Designed for seamless integration into the peri-

operative environment, the system meets strict requirements for privacy, sterility, and operational efﬁciency.

Beyond its application in surgical practice, the system can also enhance surgical education and training by

providing real-time feedback, enabling personalized learning pathways, and gamiﬁed simulation exercises. It

provides detailed analyses of trainee performance, enabling instructors to deliver targeted feedback and de-

velop adaptive training strategies based on detected posture deviations.

1 INTRODUCTION

Minimally invasive pediatric surgery has experienced

exponential growth over the past two decades, of-

fering signiﬁcant beneﬁts in reducing postoperative

pain, shorter hospital stays, and less invasive scar-

ring (Marinho et al., 2021). However, new chal-

lenges have emerged alongside these evident clini-

cal improvements, mainly related to ergonomics for

the surgical team (Rosenblatt et al., 2013; Marinho

et al., 2020). Although robotic platforms such as the

da Vinci system have enhanced ergonomics through

features like tremor ﬁltering and articulated instru-

ments, they do not entirely resolve posture-related

challenges (Haidegger et al., 2022). Operating in con-

ﬁned spaces with laparoscopic or robotic instruments

https://orcid.org/0000-0002-3802-795X

https://orcid.org/0000-0001-6607-6591

https://orcid.org/0000-0002-8858-6848

https://orcid.org/0000-0002-1281-404X

initially designed for adult surgery often forces the

surgeon to maintain static or suboptimal postures for

extended periods. The result is an increased risk of

developing musculoskeletal disorders (MSDs) (Alle-

blas et al., 2017), which can lead to chronic pain and,

in the long term, decreased performance or work ab-

sences.

Recent studies show that 50-85% of surgeons ex-

perience chronic pain and discomfort, mainly in the

neck, shoulders, and back (Anwary et al., 2021;

Bertram et al., 2023). Lower back and neck pain are

among the most common issues surgeons face, af-

fecting their physical well-being and their ability to

concentrate and make decisions during surgical pro-

cedures (Matern, 2009). Numerous studies have high-

lighted how the use of ergonomically unsuitable tools,

the height of operating tables, and the arrangement of

monitors can contribute to incorrect postures (Wong

et al., 2022; Schlussel and Maykel, 2019). Further-

more, the anatomical characteristics of pediatric pa-

tients require surgeons to perform exact maneuvers

Neroni, P., Caggianese, G., Esposito, C. and Gallo, L.

Real-Time 3D Posture Tracking for Surgeons in Pediatric Minimally Invasive Surgery.

DOI: 10.5220/0013504200003932

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

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 1, pages 921-928

ISBN: 978-989-758-746-7; ISSN: 2184-5026

921

with minimal margins for error, thereby exacerbat-

ing issues related to posture and movement (Walsh,

2023).

To overcome these difﬁculties, recent non-

invasive monitoring solutions using depth cameras,

wearables, and machine learning offer real-time pos-

ture feedback for reducing musculoskeletal risks and

improving surgical efﬁciency (Romeo et al., 2022;

Nadeem et al., 2024; Vermander et al., 2024). Effec-

tive ergonomic operating room monitoring requires

key features: accurate surgeon tracking, ignoring

other personnel (Hu et al., 2022); gesture-based ac-

tivation for sterile, seamless operation (Gallo et al.,

2011; Bigdelou et al., 2012); and privacy via non-

RGB video streams (Srivastav et al., 2019; Flouty

et al., 2018).

This paper presents the development of an inno-

vative system designed to assist surgeons in pediatric

minimally invasive surgery by proactively addressing

ergonomic challenges. It aims to reduce the risk of

posture-related musculoskeletal issues, enhance the

comfort of medical staff during operations, and sup-

port sustained concentration throughout surgical pro-

cedures. The underlying idea of this research is to

combine the acquisition of 3D data through depth sen-

sors with real-time processing techniques to immedi-

ately alert the surgeon of any deviations from opti-

mal ergonomic parameters (Romeo et al., 2022; Wang

et al., 2024). The feedback interface, designed to be

non-invasive and adaptable to the operative workﬂow,

represents a further advancement in making the oper-

ating room safer and more comfortable for the entire

team (Ayvaz et al., 2023). The anticipated outcome

is not only the improvement of individual surgeons’

posture but also a broader enhancement of ergonomic

standards in minimally invasive pediatric surgery, ul-

timately reducing the prevalence of posture-related

disorders and the associated socio-health costs (Ay-

vaz et al., 2023). Finally, this paper includes case

studies exploring the co-building of posture monitor-

ing in surgical education and training. The study ana-

lyzes these technologies to understand how they affect

short-term performance improvements and long-term

ergonomics awareness among surgical trainees.

2 RELATED WORK

Research on posture monitoring and ergonomic as-

sessment in minimally invasive surgery has steadily

increased, reﬂecting growing concerns about the im-

pact of surgeons’ posture on their well-being (Van Det

et al., 2008). Traditionally, ergonomic evaluations

have relied on post hoc methods, including question-

naires, perceived exertion measures (e.g., the Borg

scale), and manual video analysis (Wong et al., 2022;

Weitbrecht et al., 2023). One of the most widely used

tools for ergonomic assessment is the Rapid Upper

Limb Assessment (RULA) method, which enables the

static analysis of key postures by evaluating trunk,

neck, and upper limb joint angles (McAtamney and

Corlett, 1993). RULA has been applied across var-

ious healthcare professions, including cleaning staff

(Koskas and Vignais, 2024), laparoscopic surgeons,

surgical assistants, nurses (Pazouki et al., 2017),

and otolaryngologists performing endoscopic sinus

surgery (Dabholkar et al., 2020). However, these con-

ventional methods present signiﬁcant limitations, as

their reliability often depends on evaluator subjectiv-

ity, and post hoc assessments do not facilitate real-

time posture correction (Schlussel and Maykel, 2019;

Ayvaz et al., 2023).

In recent years, several studies have explored

more objective and automated techniques for posture

detection. Notably, there has been increasing interest

in the use of wearable inertial sensors, such as Inertial

Measurement Units (IMUs), for tracking joint move-

ments (Stefana et al., 2021; Haidegger et al., 2022;

Zhou et al., 2006). These devices enable continu-

ous and non-invasive data collection; however, they

present challenges related to surgeon comfort and the

need for individual sensor calibration (Romeo et al.,

2022). Moreover, cables or wearable modules may

introduce obstacles in the highly sterile and regu-

lated environment of the operating room (Wang et al.,

2024).

Computer vision systems based on RGB-D cam-

eras have gained increasing attention for their ability

to non-invasively detect posture and body position in

space (Huang et al., 2021; Antico et al., 2021). Orig-

inally developed for the gaming industry and later

adapted for clinical research, depth cameras enable

the extraction of detailed 3D posture information, al-

lowing for precise calculations of joint angles and

postural deviations (Vermander et al., 2024). Many

approaches leverage computer vision and machine

learning algorithms to extract and analyze skeletal

data, facilitating real-time applications with high re-

liability (Romeo et al., 2022; Gallo, 2013; Bertram

et al., 2023). Additionally, some methods incor-

porate dynamic statistical models (Chai and Hod-

gins, 2007) and data reliability integration techniques

(Shum et al., 2013; Zhou et al., 2014) to mitigate

noise and inaccuracies typical of RGB-D sensing de-

vices. Advanced data structures, such as the Filtered

Pose Graph (Plantard et al., 2017), have also been pro-

posed to ensure continuity even in the presence of oc-

clusions. Recent studies have explored the potential

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

922

of Microsoft Kinect sensors for automating the RULA

assessment method. Research suggests that Kinect

sensors can reliably capture joint angle data and com-

pute RULA scores in assembly line operations (Jara

et al., 2022; Jiang et al., 2017). Moreover, the ac-

curacy of Kinect-based RULA assessments has been

validated through expert evaluations and comparisons

with standard motion capture systems.

3 CLINICAL REQUIREMENTS

ANALYSIS

A focus group was conducted with the Pediatric

Surgery Unit at the Federico II University of Naples

to deﬁne the system’s clinical requirements. During

these sessions, surgeons highlighted key ergonomic

issues in pediatric minimally invasive surgery, includ-

ing the prolonged maintenance of static postures and

the difﬁculty of correcting them in real-time without

disrupting surgical workﬂow. The following section

outlines the key requirements that guided its design:

1. Surgeon Recognition and Tracking. The system

must accurately recognize and track the surgeon

among the surgical team members. This function-

ality is essential to ensure that posture monitoring

is applied exclusively to the intended subject, pre-

venting interference or data contamination from

other personnel present in the room. By isolating

relevant parameters, the system enhances the ac-

curacy of postural analysis.

2. Gesture-Based Identiﬁcation. To optimize sur-

geon identiﬁcation and eliminate the need for

manual interactions, the system must incorporate

a gesture-based identiﬁcation mechanism, per-

formed by the surgeon at the beginning of the pro-

cedure. This approach signiﬁcantly reduces auto-

matic identiﬁcation errors and ensures swift and

reliable control. A predeﬁned, system-recognized

gesture allows medical personnel to initiate postu-

ral monitoring automatically, eliminating the need

for touchscreen interfaces or manual devices. This

feature enhances operational efﬁciency while ad-

hering to the strict hygiene and sterility require-

ments of the surgical environment.

3. Privacy Protection. To safeguard the privacy of

both medical personnel and patients, the system

must avoid the use of RGB video streams, which

could inadvertently capture and reveal sensitive

visual information.

4. Non-Interference with Surgical Equipment.

The system must operate without disrupting ex-

isting surgical equipment or restricting the mobil-

ity of medical personnel. Its integration should

be seamless, ensuring unobstructed workﬂow and

adherence to standard surgical protocols.

5. Minimal Interaction and Intuitive Interface.

The system must require minimal interaction from

the surgeon, thereby reducing distractions and op-

erational downtime. Additionally, posture-related

notiﬁcations should be designed to be clear, non-

invasive, and seamlessly integrated into the surgi-

cal workﬂow, ensuring effective communication

without disrupting the procedure.

4 THE PROPOSED SOLUTION

The proposed solution employs a depth camera inte-

grated with advanced skeletal tracking software to au-

tomatically identify and monitor human body joints.

This system is designed to continuously assess the

surgeon’s posture during surgical procedures. Upon

detecting an incorrect posture, the system generates

real-time visual notiﬁcations on the operating room

display. These alerts enable the surgeon to promptly

adjust their position, thereby mitigating the risk of

musculoskeletal injuries and promoting ergonomic

practices in the operating environment.

4.1 Hardware Conﬁguration

Figure 1: Hardware conﬁguration of the proposed solution.

The hardware conﬁguration is illustrated in Fig-

ure 1. To ensure seamless integration without in-

terfering with surgical equipment, the detection de-

vice is strategically positioned at an elevated loca-

tion above the primary operating room display. This

placement minimizes obstruction to surgical proce-

dures and avoids hindering the operators’ movements.

Furthermore, the system is designed to be easily re-

movable or adjustable, facilitating quick modiﬁca-

tions as needed. The selected devices have been op-

timized for miniaturization to reduce physical clutter

and maintain an unobtrusive presence in the operating

environment.

Real-Time 3D Posture Tracking for Surgeons in Pediatric Minimally Invasive Surgery

923

For surgeon tracking, the system utilizes the Mi-

crosoft Azure Kinect DK, a depth camera that em-

ploys time-of-ﬂight technology to generate three-

dimensional skeletal maps. This approach eliminates

the need for processing color images, focusing exclu-

sively on detecting joint positions and tracking the

surgeon’s movements. This design choice not only

safeguards privacy but also reduces computational de-

mands, thereby enhancing system efﬁciency. Figure 2

depicts the joint mapping capabilities of the sensor.

For notiﬁcations, a 10-inch Beetronics Mini

Touchscreen Display was selected. This display is

equipped with integrated speakers to provide auditory

feedback in addition to visual alerts when the surgeon

adopts an incorrect posture.

Figure 2: Body joints tracked by the Azure Kinect camera.

Finally, a ZOTAC ZBOX Mini PC equipped with

an Intel Core i7 processor and an NVIDIA RTX

A4500 graphics card featuring 16 GB of dedicated

memory was chosen to handle the graphical process-

ing demands. This hardware conﬁguration ensures ef-

ﬁcient three-dimensional skeletal reconstruction and

real-time management of visual notiﬁcations and en-

ables body tracking at approximately 30 FPS.

4.2 Interface

The interface has been designed to minimize interac-

tion requirements, allowing the surgeon to focus en-

tirely on the procedure without disruptions or com-

plications when using the system. The graphical in-

terface consists of a text-based notiﬁcation system

displayed on a mini-display. Because of the limited

screen size, notiﬁcations must be brief enough to be

read easily during surgery, pointing only to relevant

information. The system’s workﬂow is illustrated in

ﬁgure 3, where the top row shows the perspective

from the Azure Kinect sensor, and the bottom row

represents the surgeon’s viewpoint. Initially, before

recognizing the surgeon, the display shows the mes-

sage “Waiting for identiﬁcation gesture”. To iden-

tify themselves, the surgeon must raise their hand for

two seconds, after which the display updates to “Sur-

geon detected. Monitoring on”. If the surgeon main-

tains a correct posture during surgery, the system dis-

plays “Posture OK. Monitoring...”. However, if a de-

viation from the optimal posture is detected, a spe-

ciﬁc alert indicates the nature of the incorrect posture.

The visual notiﬁcation is supplemented by an unob-

trusive auditory cue that allows surgeons to receive

immediate feedback without looking away from the

surgical ﬁeld. Therefore, surgeons may immediately

adjust their positioning, mitigating musculoskeletal

strain and reducing the risk of long-term ergonomic

issues.

4.3 Postural Tracking

Postural tracking is based on the three-dimensional

reconstruction of the surgeon’s skeleton, utilizing

algorithms that process data from the depth cam-

era. These algorithms analyze joint positions in real

time to assess the surgeon’s posture during the op-

eration. Among the various techniques for assess-

ing postural risk, we chose to base our approach on

the RULA (Rapid Upper Limb Assessment) method,

given its extensive documentation in the literature.

This method generates a score based on the biome-

chanical and postural load demands of work activ-

ities, with a particular focus on the upper body, in-

cluding the neck, trunk, and upper limbs. Given our

emphasis on analyzing postural discomfort related to

the surgeon’s upper body, the RULA method was de-

termined to be the most appropriate for this study.

This approach is particularly relevant as tracking oc-

curs from the waist upwards, with the operating table

obscuring the lower body.

For the calculation of most of the angles re-

quired for assessment, we referenced (Manghisi et al.,

2017), which uses the Kinect v2 sensor to detect

risky postures according to the RULA method. A

comparative analysis was then conducted between

the joints tracked by the Kinect v2 sensor and those

tracked by the Azure Kinect sensor. Unlike (Manghisi

et al., 2017), which calculates a comprehensive pos-

ture score for the surgeon, our method provides real-

time notiﬁcations to alert the surgeon whenever an

improper posture is detected.

Speciﬁcally, the system automatically detects and

notiﬁes the surgeon of the following incorrect pos-

tures:

• Trunk Flexion (Forward or Backward) (see ﬁg-

ure 4.a). Occurs when the trunk is excessively in-

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

924

Figure 3: System workﬂow for surgeon posture monitoring. The top row illustrates the Azure Kinect sensor’s perspective,

while the bottom represents the surgeon’s viewpoint. (a) Initially, the system displays ”Waiting for identiﬁcation gesture.”

(b) When the surgeon raises their hand for two seconds, the display updates to ”Surgeon detected. Monitoring on.” (c) If the

surgeon maintains a correct posture, the display shows ”Posture OK. Monitoring. . . ”. (d) When an incorrect posture (e.g.,

forward trunk ﬂexion) is detected, the system provides real-time feedback through a visual notiﬁcation, prompting the surgeon

to adjust their posture.

clined either forward or backward from the neu-

tral, upright position. The trunk inclination an-

gle is calculated using the positions of the head,

pelvis, and spine joints. Positive values corre-

spond to forward ﬂexion, while negative values

indicate extension or backward bending.

• Lateral Trunk Inclination (see ﬁgure 4.b). Oc-

curs when the trunk is excessively tilted to one

side relative to the vertical position. The lateral

inclination angle is measured based on the posi-

tions of the head, pelvis, and spine joints. Higher

values indicate a greater degree of lateral tilt.

• Trunk Rotation Around the Vertical Axis (see

ﬁgure 4.c). Occurs when the trunk undergoes ex-

cessive rotation around the vertical axis, which

is a critical condition leading to asymmetric pos-

tures. The trunk torsion angle is measured by

comparing the rotation of the torso relative to its

initial position, using the relative positions of the

torso and shoulder joints.

• Excessive Arm Extension (see ﬁgure 4.d). Oc-

curs when the elbow is overextended, keeping the

arm rigidly straight without sufﬁcient ﬂexion for

ergonomic movement. The arm ﬂexion angle is

computed by analyzing the positions of the shoul-

der, elbow, and wrist joints. An angle of 0° repre-

sents a fully extended (straight) arm, with insufﬁ-

cient ﬂexion indicating an incorrect posture.

• Excessive Shoulder Abduction (see ﬁgure 4.e).

Occurs when the shoulder is raised laterally be-

yond ergonomic limits. The shoulder abduc-

tion angle is measured by analyzing the three-

dimensional positions of the shoulder, elbow, and

wrist joints. Positive values indicate outward

movement (abduction), while negative values rep-

resent extension or adduction.

5 CASE STUDY: POSTURE

MONITORING IN SURGICAL

EDUCATION AND TRAINING

The real-time posture monitoring system introduced

in this paper holds much promise for changing surgi-

cal education, particularly in pediatric minimally in-

vasive surgery. The system’s introduction into train-

ing programs could help novices and veteran surgeons

become attuned to proper ergonomic practices, possi-

bly reducing the risk of musculoskeletal disorders in

the long term. The primary advantages of utilizing

such a system in surgical education and training are

as follows:

• Real-Time Feedback in Simulated Environ-

ments. An important component of efﬁcient sur-

gical training is the ability to recreate safe and re-

producible real-life scenarios. The posture moni-

toring system gives trainees ongoing feedback on

Real-Time 3D Posture Tracking for Surgeons in Pediatric Minimally Invasive Surgery

925

Figure 4: Set of postures tracked by the system: a) Forward trunk ﬂexion; b) Lateral trunk inclination; c) Trunk rotation; d)

Arm extension; e) Shoulder abduction.

their posture in a simulation setting. For exam-

ple, if the trainee strays away from an optimal er-

gonomic position, immediate visual and auditory

signals notify them to correct their stance. Over

time, this instant feedback could reinforce healthy

posture habits. It adds to the real-time monitoring

of surgeries, just like they experience at the simu-

lation center.

• Personalized Learning through Analytics. Data

processing functions of the system keep detailed

records of every candidate’s performance through

posture tracking during practice sessions. Analyz-

ing these data allows the instructors to perceive re-

curring faults, customize their interventions, and

present individually tailored comments to their

trainees through custom-made reports. Trainees

can self-navigate individual learning paths that

encourage continuous improvement in posture

management. Tools like heatmaps—which high-

light areas where a candidate frequently deviates

from proper posture—offer clear visual feedback

that should allow trainees to ﬁnd and correct their

common mistakes.

• Gamiﬁcation to Engage in Ergonomics Learn-

ing. Serious games tailored for ergonomic-

focused learning could enhance learning with the

posterior monitoring system. These can simu-

late game-like procedures where trainees perform

simulated surgeries while maintaining an accu-

rate posture with points or levels gained due to

their compliance with ergonomic guidelines. This

gamiﬁcation approach increases motivation and

engagement, making learning enjoyable and ef-

fective. Challenge scenarios such as maintaining

posture during complex maneuvers or under time

pressure can closely mimic the demands of actual

surgery. Such gaming can replicate the physical

and cognitive stressors imposing conﬂicts on the

novice in the operating room, giving an all-around

training opportunity. Examples include holding

a posture during critical interventions or working

with multiple tools. Other beneﬁts are real-time

leaderboards that allow friendly competition and

encourage continuous practice.

• Promoting Long-Term Ergonomic Awareness.

Integrating ergonomic education into surgical

training is important to prevent long-term health

problems. Integrating a posture monitoring sys-

tem into residency training highlights the impor-

tance of proper ergonomics from the beginning

of a surgeon’s career. Early exposure to these

concepts helps protect physical well-being in the

long run, which, in turn, has a positive effect on

surgical performance. Using theoretical lessons

and practical simulations supported by monitoring

tools like the one proposed, surgical trainees may

better understand ergonomics in practice. Work-

shops and training modules should integrate theo-

retical content with practical tutorial sessions us-

ing the monitoring system, laying the foundation

for fostering an ergonomic culture within surgical

teams.

• Research Opportunities in Educational Set-

tings. Future investigations should concentrate

on how such a monitoring system impacts sur-

gical training in parametric and clinical envi-

ronments. Researchers could consider establish-

ing measurements of musculoskeletal complaints,

RULA scoring improvement, and surgical perfor-

mance metrics. Moreover, this study could as-

sess how automated alerts about posture inﬂu-

enced participant and instructor stress and fo-

cus in a simulated surgical environment. Feed-

back from participants and instructors’ quantita-

tive data related to ICT usage would supplement

other measurement methods to give a more holis-

tic view of usability and educational beneﬁts. Ul-

timately, these studies will contribute to the ex-

panding body of literature supporting the integra-

tion of advanced posture monitoring technologies

into medical education. This, in turn, will en-

able future research to demonstrate clear beneﬁts

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

926

for both health and learning outcomes, reinforc-

ing the role of ergonomic training as a standard

component of surgical education.

6 CONCLUSION AND FUTURE

WORK

Minimally invasive pediatric surgery presents signiﬁ-

cant ergonomic challenges that may increase the risk

of MSDs among surgeons. This study introduced a

real-time posture monitoring system that leverages

depth sensors and 3D skeletal tracking to provide

immediate feedback on postural deviations, with the

goal of enhancing both surgeon well-being and pro-

cedural efﬁciency. The system ensures privacy pro-

tection and offers seamless, unobtrusive integration

into the operating room without disrupting workﬂow.

While the system shows promise, further validation

in routine surgical environments is necessary. Critical

aspects to assess include its stability in tracking the

surgeon despite occlusions from medical staff, accu-

racy under varying lighting conditions, and the poten-

tial for false alarms that could interfere with concen-

tration. Beyond surgical practice, this study also ex-

plored the integration of posture monitoring into sur-

gical education and training. Real-time feedback, per-

sonalized learning pathways, and gamiﬁcation strate-

gies could enhance trainee engagement and encour-

age the early adoption of proper ergonomic habits.

Future research will focus on experimental valida-

tion and evaluating the system’s applicability across

different surgical specialties. Additionally, long-term

studies will examine the impact of integrating posture

monitoring technology into educational programs, as-

sessing its effects on ergonomic awareness, trainee

evaluation, and improvements in occupational health

outcomes.

REFERENCES

Alleblas, C. C., De Man, A. M., Van Den Haak, L., Vier-

hout, M. E., Jansen, F. W., and Nieboer, T. E. (2017).

Prevalence of musculoskeletal disorders among sur-

geons performing minimally invasive surgery: a sys-

tematic review. Annals of surgery, 266(6):905–920.

Antico, M., Balletti, N., Laudato, G., Lazich, A., Notaran-

tonio, M., Oliveto, R., Ricciardi, S., Scalabrino, S.,

and Simeone, J. (2021). Postural control assessment

via microsoft azure kinect dk: An evaluation study.

Computer Methods and Programs in Biomedicine,

209:106324.

Anwary, A. R., Cetinkaya, D., Vassallo, M., Bouchachia,

H., et al. (2021). Smart-cover: A real time sitting

posture monitoring system. Sensors and Actuators A:

Physical, 317:112451.

Ayvaz,

O.,

Ozyıldırım, B. A.,

Is¸sever, H.,

Oztan, G., Atak,

M., and

Ozel, S. (2023). Ergonomic risk assessment

of working postures of nurses working in a medical

faculty hospital with reba and rula methods. Science

Progress, 106(4):00368504231216540.

Bertram, J., Kr

uger, T., R

ohling, H. M., Jelusic, A.,

Mansow-Model, S., Schniepp, R., Wuehr, M., and

Otte, K. (2023). Accuracy and repeatability of the mi-

crosoft azure kinect for clinical measurement of motor

function. PLoS One, 18(1):e0279697.

Bigdelou, A., Schwarz, L., and Navab, N. (2012). An adap-

tive solution for intra-operative gesture-based human-

machine interaction. In Proceedings of the 2012

ACM international conference on Intelligent User In-

terfaces, pages 75–84.

Chai, J. and Hodgins, J. K. (2007). Constraint-based mo-

tion optimization using a statistical dynamic model.

In ACM SIGGRAPH 2007 papers, pages 8–es. ACM.

Dabholkar, T., Dabholkar, Y. G., Yardi, S., and Sethi, J.

(2020). An objective ergonomic risk assessment of

surgeons in real time while performing endoscopic si-

nus surgery. Indian Journal of Otolaryngology and

Head & Neck Surgery, 72(3):342–349.

Flouty, E., Zisimopoulos, O., and Stoyanov, D. (2018).

Faceoff: Anonymizing videos in the operating rooms.

In Stoyanov, D., Taylor, Z., Sarikaya, D., McLeod,

J., Gonz

alez Ballester, M. A., Codella, N. C., Mar-

tel, A., Maier-Hein, L., Malpani, A., Zenati, M. A.,

De Ribaupierre, S., Xiongbiao, L., Collins, T., Re-

ichl, T., Drechsler, K., Erdt, M., Linguraru, M. G.,

Oyarzun Laura, C., Shekhar, R., Wesarg, S., Celebi,

M. E., Dana, K., and Halpern, A., editors, OR 2.0

Context-Aware Operating Theaters, Computer As-

sisted Robotic Endoscopy, Clinical Image-Based Pro-

cedures, and Skin Image Analysis, pages 30–38,

Cham. Springer International Publishing.

Gallo, L. (2013). A study on the degrees of freedom in

touchless interaction. In SIGGRAPH Asia 2013 Tech-

nical Briefs, pages 1–4. ACM.

Gallo, L., Placitelli, A. P., and Ciampi, M. (2011).

Controller-free exploration of medical image data:

experiencing the kinect. In 2011 24th Interna-

tional Symposium on Computer-Based Medical Sys-

tems (CBMS), page 1–6. IEEE.

Haidegger, T., Speidel, S., Stoyanov, D., and Satava,

R. M. (2022). Robot-assisted minimally invasive

surgery—surgical robotics in the data age. Proceed-

ings of the IEEE, 110(7):835–846.

Hu, H., Hachiuma, R., Saito, H., Takatsume, Y., and Ka-

jita, H. (2022). Multi-camera multi-person tracking

and re-identiﬁcation in an operating room. Journal of

Imaging, 8(8):219.

Huang, W., Chen, J., Zhao, X., and Liu, Q. (2021). Per-

formance evaluation of azure kinect and kinect 2.0

and their applications in 3d key-points detection of

students in classroom environment. In International

Conference on Artiﬁcial Intelligence in Education

Technology, pages 177–193. Springer.

Jara, H. V., Orejuela, I. Z., and Baydal-Bertomeu, J. (2022).

Study of the ergonomic risk in operators of an assem-

Real-Time 3D Posture Tracking for Surgeons in Pediatric Minimally Invasive Surgery

927

bly line using the rula method in real working condi-

tions through the application of a commercial sensor.

Materials Today: Proceedings, 49:122–128.

Jiang, S., Liu, P., Fu, D., Xue, Y., Luo, W., and Wang,

M. (2017). A low-cost rapid upper limb assessment

method in manual assembly line based on somatosen-

sory interaction technology. In AIP Conference Pro-

ceedings, volume 1834. AIP Publishing.

Koskas, D. and Vignais, N. (2024). Physical ergonomic as-

sessment in cleaning hospital operating rooms based

on inertial measurement units. Bioengineering,

11(2):154.

Manghisi, V. M., Uva, A. E., Fiorentino, M., Bevilacqua, V.,

Trotta, G. F., and Monno, G. (2017). Real time rula as-

sessment using kinect v2 sensor. Applied ergonomics,

65:481–491.

Marinho, M. M., Harada, K., Deie, K., Ishimaru, T., and

Mitsuishi, M. (2021). Smartarm: Suturing feasibility

of a surgical robotic system on a neonatal chest model.

IEEE Transactions on Medical Robotics and Bionics,

3(1):253–256.

Marinho, M. M., Ishida, H., Harada, K., Deie, K., and Mit-

suishi, M. (2020). Virtual ﬁxture assistance for sutur-

ing in robot-aided pediatric endoscopic surgery. IEEE

Robotics and Automation Letters, 5(2):524–531.

Matern, U. (2009). Ergonomic deﬁciencies in the operat-

ing room: examples from minimally invasive surgery.

Work, 33(2):165–168.

McAtamney, L. and Corlett, E. N. (1993). Rula: a sur-

vey method for the investigation of work-related upper

limb disorders. Applied ergonomics, 24(2):91–99.

Nadeem, M., Elbasi, E., Zreikat, A. I., and Sharsheer, M.

(2024). Sitting posture recognition systems: Compre-

hensive literature review and analysis. Applied Sci-

ences, 14(18):8557.

Pazouki, A., Sadati, L., Zarei, F., Golchini, E., Fruzesh,

R., and Bakhtiary, J. (2017). Ergonomic chal-

lenges encountered by laparoscopic surgeons, sur-

gical ﬁrst assistants, and operating room nurses in-

volved in minimally invasive surgeries by using rula

method. Journal of Minimally Invasive Surgical Sci-

ences, 6(4):344–6.

Plantard, P., H. Shum, H. P., and Multon, F. (2017). Fil-

tered pose graph for efﬁcient kinect pose reconstruc-

tion. Multimedia Tools and Applications, 76:4291–

4312.

Romeo, L., Marani, R., Perri, A. G., and D’Orazio, T.

(2022). Microsoft azure kinect calibration for three-

dimensional dense point clouds and reliable skeletons.

Sensors, 22(13):4986.

Rosenblatt, P. L., McKinney, J., and Adams, S. R. (2013).

Ergonomics in the operating room: protecting the

surgeon. Journal of minimally invasive gynecology,

20(6):744.

Schlussel, A. T. and Maykel, J. A. (2019). Ergonomics

and musculoskeletal health of the surgeon. Clinics in

colon and rectal surgery, 32(06):424–434.

Shum, H. P., Ho, E. S., Jiang, Y., and Takagi, S. (2013).

Real-time posture reconstruction for microsoft kinect.

IEEE transactions on cybernetics, 43(5):1357–1369.

Srivastav, V., Gangi, A., and Padoy, N. (2019). Human pose

estimation on privacy-preserving low-resolution depth

images. In International conference on medical image

computing and computer-assisted intervention, pages

583–591. Springer.

Stefana, E., Marciano, F., Rossi, D., Cocca, P., and Toma-

soni, G. (2021). Wearable devices for ergonomics: A

systematic literature review. Sensors, 21(3):777.

Van Det, M., Meijerink, W., Hoff, C., Van Veelen, M., and

Pierie, J. (2008). Ergonomic assessment of neck pos-

ture in the minimally invasive surgery suite during

laparoscopic cholecystectomy. Surgical endoscopy,

22:2421–2427.

Vermander, P., Mancisidor, A., Cabanes, I., and Perez, N.

(2024). Intelligent systems for sitting posture moni-

toring and anomaly detection: an overview. Journal

of NeuroEngineering and Rehabilitation, 21(1):28.

Walsh, C. M. (2023). Enhancing ergonomics in pediatric

endoscopy training and practice. Gastrointestinal En-

doscopy Clinics, 33(2):235–251.

Wang, S.-A., Lu, M.-H., Lee, A.-T., Chen, C.-Y., and

Lee, L.-W. (2024). Performance of microsoft azure

kinect dk as a tool for estimating human body segment

lengths. Scientiﬁc Reports, 14(1):15811.

Weitbrecht, M., Holzgreve, F., Fraeulin, L., Haenel, J.,

Betz, W., Erbe, C., Maurer-Grubinger, C., Wanke,

E. M., Brueggmann, D., Nienhaus, A., et al. (2023).

Ergonomic risk assessment of oral and maxillofa-

cial surgeons–rula applied to objective kinematic data.

Human Factors, 65(8):1655–1673.

Wong, S. W., Ang, Z. H., Yang, P. F., and Crowe, P. (2022).

Robotic colorectal surgery and ergonomics. Journal

of Robotic Surgery, pages 1–6.

Zhou, H., Hu, H., Harris, N. D., and Hammerton, J. (2006).

Applications of wearable inertial sensors in estimation

of upper limb movements. Biomedical Signal Pro-

cessing and Control, 1(1):22–32.

Zhou, L., Liu, Z., Leung, H., and Shum, H. P. (2014). Pos-

ture reconstruction using kinect with a probabilistic

model. In Proceedings of the 20th ACM Symposium

on Virtual Reality Software and Technology, pages

117–125.

ERSeGEL 2025 - Workshop on Extended Reality and Serious Games for Education and Learning

928