Mar 29, 2026
Scoping Review Protocol: Medical Simulation Manikin Adaptation for Robotic Applications (2004–2026)
- Nabil Zary1,
- Mohamed AlAli1
- 1Institute of Learning, Mohammed Bin Rashid University of Medicine and Health Sciences, Dubai Health
- NeuroInk
Protocol Citation: Nabil Zary, Mohamed AlAli 2026. Scoping Review Protocol: Medical Simulation Manikin Adaptation for Robotic Applications (2004–2026). protocols.io https://dx.doi.org/10.17504/protocols.io.3byl4p8q2lo5/v1
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: March 29, 2026
Last Modified: March 29, 2026
Protocol Integer ID: 314065
Keywords: scoping review, medical simulation, manikins, robotics, convergence research, PRISMA-ScR, patient simulator, medical simulation manikin adaptation for robotic application, medical simulation manikins for robotic application, medical simulation manikin adaptation, medical simulation manikin, medical simulation, robotics engineering, robotics engineering community, robotic application, scr guideline, scoping review protocol, extent of technological adaptation, hybrid technology domain, technological adaptation
Abstract
This protocol describes a scoping review mapping the research landscape at the intersection of Medical
Simulation manikins and robotics engineering from 2004 to 2026.
Using the Arksey and O'Malley (2005) framework, with refinements by Levac et al. (2010)and reported per PRISMA-ScR guidelines, the review systematically identifies and characterises studies that adapt, modify, or repurpose medical simulation manikins for robotic applications.
Three research questions guide the review: (1) the nature and extent of technological adaptations across hardware, software, and hybrid domains; (2) temporal trends and geographic patterns; and (3) the degree of convergence between the medical simulation and robotics engineering communities.
The search is conducted via PubMed, Scopus, Web of Science, IEEE Xplore, ACM Digital Library, and Embase). Three independent reviewers screen records using majority-rule consensus, with inter-rater reliability assessed
via Fleiss' kappa and Prevalence-Adjusted Bias-Adjusted Kappa (PABAK).
Included studies are classified into Hardware, Software, or Hybrid technology domains and charted using a standardised data extraction form. Synthesis includes descriptive numerical summaries, cross-tabulations, convergence indicator analysis (disciplinary asymmetry, cross-disciplinary tool adoption, venue fragmentation, cross-domain clinical coverage, and Hybrid domain trajectory), and narrative thematic analysis organised around the three research questions. Sensitivity analyses address the inclusion threshold and domain classification
confidence.
The review is reported in accordance with all 22 PRISMA-ScR checklist items.
Guidelines
**Step 3: Screen and Select Studies**
**3.1 Reviewer Team**
Three independent reviewers screen all records.
**3.2 Screening Procedure**
For each record, each reviewer independently:
1. Reads the title and abstract
2. Classifies the paper as Include, Exclude, or Maybe
3. Records a brief rationale
4. For Exclude decisions, assigns an exclusion reason code (E1–E8, see Step 1.2)
**3.3 Consensus Rules**
- Inclusion: A paper is included if at least 2 of 3 reviewers agree on inclusion
- Maybe: Papers classified as Maybe by majority are retained as included studies
- Exclusion: A paper is excluded if at least 2 of 3 reviewers agree on exclusion
- Discordant cases: Cases where no majority exists are resolved through discussion among all three reviewers. Document the number of discordant cases and how each was resolved.
**3.4 Inter-Rater Reliability**
Calculate and report the following reliability statistics:
- Fleiss' kappa for three raters
- Pairwise percent agreement for each reviewer pair (R1–R2, R1–R3, R2–R3)
- Three-way agreement (proportion of records where all three reviewers agreed)
- Prevalence-Adjusted Bias-Adjusted Kappa (PABAK): Calculate using PABAK = 2Po – 1, where Po is the observed pairwise agreement proportion. Compute for each pair and report the average. PABAK corrects for statistical deflation of kappa under high inclusion rates expected in scoping reviews (Byrt et al., 1993; Sim 6 Wright, 2005).
**3.5 Sensitivity Analysis — Inclusion Threshold**
- Identify all studies included by majority (2/3) but not unanimously
- Also identify studies classified as Maybe
- Report how many studies could potentially change status under unanimous agreement
- Confirm via discussion-based review whether these studies meet eligibility criteria
**Step 4: Develop Analytical Framework (Domain Classification)**
**4.1 Domain Definitions**
Classify each included study into exactly one of three mutually exclusive domains based on its primary technological contribution:
| Domain | Definition | Defining Characteristic |
|----------|-----------------------------------------------------------------------------|--------------------------------------------------------------|
| Hardware | Physical modification of manikins: sensors, actuators, musculoskeletal systems, structural or material modifications | Core innovation resides in the physical apparatus |
| Software | Computational contribution: control algorithms, physiological simulation engines, AI/ML, robotics middleware, teleoperation architectures | No significant physical modification to the manikin |
| Hybrid | Deliberate combination of physical modifications with digital/virtual overlays, mixed reality, or tightly integrated systems | Neither hardware nor software alone constitutes the primary contribution |
**4.2 Decision Rules for Ambiguous Cases**
- If a study explicitly emphasises both hardware and software as equally central → classify as Hybrid
- If dominance remains unclear after careful reading → defer to the stated application or clinical impact
- If still ambiguous → flag as low-confidence and resolve via full-text review and discussion
**4.3 Classification Procedure**
1. Two reviewers independently classify all included studies
2. For each study, record: primary contributing domain, classification confidence rating (high, moderate, or low), and brief justification
3. Resolve disagreements through discussion-based consensus
4. Studies classified as low-confidence require full-text review
5. Track and report the number of discordant and low-confidence cases
**4.4 Sensitivity Analysis — Domain Classification**
1. Restrict to high-confidence classifications only
2. Compare domain distribution (% Hardware, % Software, % Hybrid) with full dataset
3. Conduct independent verification of all classifications
4. Report number and direction of any reclassifications
**Step 5: Chart the Data**
**5.1 Data Charting Form**
Develop a standardised data charting form with the following variables:
| Variable | Description |
|---------------------------|-----------------------------------------------------------------------------|
| Author(s) | All authors |
| Publication year | Year of publication |
| Country | Country of first author affiliation |
| Journal/Conference | Publication venue |
| Study design | Prototype/Development, Experimental/Evaluation, Simulation study, RCT, Review, Technical/Methodological, or Not classified |
| Domain | Hardware, Software, or Hybrid (from Step 4) |
| Clinical application domain | Primary clinical area (assigned inductively) |
| Manikin/simulator type | Type of manikin or simulator described |
| Key technologies | Technologies used (sensors, actuators, AI/ML, middleware, etc.) |
| Key findings | Main findings or contributions (1–2 sentences) |
**5.2 Pilot Testing**
Pilot-test the data charting form on 10 studies. Assess whether all variables are extractable, the form is complete, and coding instructions are clear. Revise as needed before applying to the full dataset.
**5.3 Clinical Application Domain Coding**
Clinical application domains are assigned inductively based on the primary clinical area addressed in each study. Studies addressing multiple clinical areas are coded using the primary application as stated by the authors.
**5.4 Abstract Recovery**
If abstracts are not available for all included studies:
1. Stage 1: Retrieve from primary search platform
2. Stage 2: Query PubMed E-utilities API
3. Stage 3: Manual full-text retrieval through institutional library access, direct author contact, and open-access repositories
Target: 100% abstract coverage.
**Step 6: Synthesise Results**
**6.1 Descriptive Numerical Summary**
Tabulate included studies by:
- Domain (Hardware, Software, Hybrid)
- Publication year and temporal period
- Geographic distribution (country of first author)
- Study design
- Clinical application domain
- Key technology type
**6.2 Cross-Tabulations**
- Domain × publication period (Table 2)
- Domain × clinical application area (Figure 3)
**6.3 Convergence Indicators (RQ3)**
To assess the state of convergence between simulation and robotics communities, examine:
1. Disciplinary asymmetry: Hardware-to-software ratio across time periods
2. Cross-disciplinary tool adoption: Count studies explicitly describing AI/ML and robotics middleware (ROS)
3. Venue fragmentation: Number of unique publication venues; presence/absence of a dominant venue
4. Cross-domain clinical coverage: Proportion of clinical areas with studies from 2+ domains; identify areas addressed by all 3 domains
5. Hybrid domain trajectory: Trend in Hybrid study output over time; citation metrics by domain
**6.4 Narrative Thematic Analysis**
Organise narrative synthesis around the three research questions:
- RQ1: Nature and extent of research (domain distribution, study designs, technology landscape, clinical application areas)
- RQ2: Temporal trends and geographic patterns (publication growth, geographic concentration, centres of activity, venue distribution)
- RQ3: Convergence indicators (disciplinary asymmetry, absence of cross-disciplinary tools, venue fragmentation, cross-domain clinical coverage, Hybrid domain as convergence signal)
**6.5 Sensitivity Analyses**
Report results of both sensitivity analyses:
1. Inclusion threshold: Number of studies affected by varying from 2/3 majority to unanimous agreement; whether any findings change materially
2. Domain classification confidence: Domain distribution when restricted to high-confidence classifications only; number and direction of reclassifications
**Step 7: Report According to PRISMA-ScR**
**7.1 Required Elements**
Ensure the manuscript includes all 22 PRISMA-ScR checklist items (Tricco et al., 2018). Key elements:
- Structured abstract (Background, Methods, Results, Conclusions)
- PRISMA-ScR flow diagram (Figure 1)
- Complete search strategy (Additional file 1)
- Completed PRISMA-ScR checklist (Additional file 3)
- Data charting form with all extracted data (Additional file 2)
**7.2 Mandatory Declarations**
Include the following statements:
- Ethics approval and consent to participate (not applicable for scoping reviews)
- Availability of data and materials
- Competing interests
- Funding source
- Author contributions
- Protocol registration status
**7.3 Protocol Registration**
Register this protocol on protocols.io and cite the DOI in the manuscript Methods section. Update the manuscript text to include the retrospective registration DOI.
Troubleshooting
Before start
**Background and Rationale**
Medical simulation manikins and robotics engineering address overlapping challenges — anatomical realism, mechanical compliance, safe human interaction — yet no systematic assessment has examined the intersection of these fields. This scoping review maps the research landscape at the intersection of medical simulation manikins and robotics (2004–2026) and characterises the current state of convergence between the medical simulation and robotics engineering communities.
**Research Questions**
**RQ1: What is the nature and extent of research on adapting medical simulation manikins for robotic applications, in terms of technological approaches (hardware, software, and hybrid), study designs, and clinical application areas?
**RQ2: What temporal trends and geographic patterns characterise this field, and what do they reveal about centres of research activity?
**RQ3: To what extent have the medical simulation and robotics engineering communities converged in this domain, and what barriers to deeper integration can be identified?
**Methodological Framework**
- Primary framework: Arksey 6 O'Malley (2005) five-stage scoping review methodology
- Refinements: Levac et al. (2010); Joanna Briggs Institute (Peters et al., 2020)
- Reporting guideline: PRISMA-ScR (Tricco et al., 2018)
- No formal critical appraisal of individual studies (consistent with scoping review methodology)
Before Start
Background and Rationale
Medical simulation manikins and robotics engineering address overlapping challenges, anatomical realism, mechanical compliance, and safe human interaction, yet no systematic assessment has examined the intersection of these fields. This scoping review maps the research landscape at the intersection of medical simulation manikins and robotics (2004–2026) and characterises the current state of convergence between the medical simulation and robotics engineering communities.
Research Questions
RQ1: What is the nature and extent of research on adapting medical simulation manikins for robotic applications, in terms of technological approaches (hardware, software, and hybrid), study designs, and clinical application areas?
RQ2: What temporal trends and geographic patterns characterise this field, and what do they reveal about centres of research activity?
RQ3: To what extent have the medical simulation and robotics engineering communities converged in this domain, and what barriers to deeper integration can be identified?
Methodological Framework
- Primary framework: Arksey and O'Malley (2005) five-stage scoping review methodology
- Refinements: Levac et al. (2010); Joanna Briggs Institute (Peters et al., 2020)
- Reporting guideline: PRISMA-ScR (Tricco et al., 2018)
- No formal critical appraisal of individual studies (consistent with scoping review methodology)
Step 1: Define Eligibility Criteria
Inclusion Criteria
Include studies that describe the adaptation, modification, or repurposing of medical simulation manikins or patient simulators for robotic applications:
- Hardware modifications: Sensors, actuators, musculoskeletal systems, structural or material modifications to physical manikins
- Software integration: Control architectures, physiological simulation engines, AI/machine learning systems, robotics middleware, or teleoperation architectures applied to existing manikin platforms
- Hybrid approaches: Physical manikins combined with digital interfaces, mixed reality, or tightly integrated physical-plus-software systems
All study designs are eligible. No date restriction. English-language publications only.
Exclusion Criteria
Code | Reason
E1 | Duplicates
E2 | No robotic adaptation component
E3 | Off-topic (not related to manikin–robotics intersection)
E4 | Not manikin-based (non-humanoid robotic systems without relevance to manikin platforms)
E5 | Virtual only (exclusively virtual/screen-based simulation without physical manikin)
E6 | Non-English
E7 | Editorials, commentaries, or opinion pieces without original technical content
E8 | Conference abstracts without full-text availability
1.3 Terminology Note
"Manikin" refers to full-body or partial-body physical training devices. "Patient simulator" encompasses manikins equipped with physiological modelling or responsive capabilities. Both terms are used as reported in the original studies.
Step 2: Design Search Strategy
Information Source
Conduct a search using:
1. PubMed
2. Scopus
3. Web of Science Core Collection
4. IEEE Xplore
5. ACM Digital Library
6. Embase
Search Concept Blocks
Organise the search around three concept blocks combined with Boolean AND:
Block 1 — Medical simulation platforms: "medical simulation manikin," "patient simulator," "human patient simulator," "training manikin," "healthcare simulation," and related synonyms
Block 2 — Robotics adaptation: robot*, retrofit*, adapt*, modif*, augment*, sensorized, actuat*, mechatronic, humanoid
Block 3 — Technical integration: actuator*, sensor*, "control system," "control architecture," ROS, "physiological model," "modular platform," "open architecture"
Database-Specific Syntax
Adapt Boolean strings for each database using appropriate field tags: PubMed [tiab]; Scopus TITLE-ABS-KEY; Web of Science TS; IEEE Xplore free-text; ACM Digital Library free-text; Embase ti,ab,kw. Record the complete database-specific Boolean strings as a supplementary file (Additional file 1).
Search Execution
- Execute search with no date filters or publication type filters
- Export ALL records without applying a relevance score threshold
- Record search date and number of records retrieved
- Conduct post hoc validation: confirm that landmark papers known to the research team prior to the search are captured in the returned records
Note: Citation chasing (backward and forward) is not conducted in this protocol. This is a recognised limitation.
Step 3: Screen and Select Studies
Reviewer Team
Three independent reviewers screen all records.
Screening Procedure
For each record, each reviewer independently:
1. Reads the title and abstract
2. Classifies the paper as Include, Exclude, or Maybe
3. Records a brief rationale
4. For Exclude decisions, assigns an exclusion reason code (E1–E8, see Step 1.2)
Consensus Rules
- Inclusion: A paper is included if at least 2 of 3 reviewers agree on inclusion
- Maybe: Papers classified as Maybe by majority are retained as included studies
- Exclusion: A paper is excluded if at least 2 of 3 reviewers agree on exclusion
- Discordant cases: Cases where no majority exists are resolved through discussion among all three reviewers. Document the number of discordant cases and how each was resolved.
Inter-Rater Reliability
Calculate and report the following reliability statistics:
- Fleiss' kappa for three raters
- Pairwise percent agreement for each reviewer pair (R1–R2, R1–R3, R2–R3)
- Three-way agreement (proportion of records where all three reviewers agreed)
- Prevalence-Adjusted Bias-Adjusted Kappa (PABAK): Calculate using PABAK = 2Po – 1, where Po is the observed pairwise agreement proportion. Compute for each pair and report the average. PABAK corrects for statistical deflation of kappa under high inclusion rates expected in scoping reviews (Byrt et al., 1993; Sim 6 Wright, 2005).
Sensitivity Analysis — Inclusion Threshold
- Identify all studies included by majority (2/3) but not unanimously
- Also identify studies classified as Maybe
- Report how many studies could potentially change status under unanimous agreement
- Confirm via discussion-based review whether these studies meet eligibility criteria
Step 4: Develop Analytical Framework (Domain Classification)
Domain Definitions
Classify each included study into exactly one of three mutually exclusive domains based on its primary technological contribution:
| Domain | Definition | Defining Characteristic
| Hardware | Physical modification of manikins: sensors, actuators, musculoskeletal systems, structural or material modifications | Core innovation resides in the physical apparatus |
Decision Rules for Ambiguous Cases
- If a study explicitly emphasises both hardware and software as equally central → classify as Hybrid
- If dominance remains unclear after careful reading → defer to the stated application or clinical impact
- If still ambiguous → flag as low-confidence and resolve via full-text review and discussion
Classification Procedure
1. Two reviewers independently classify all included studies
2. For each study, record: primary contributing domain, classification confidence rating (high, moderate, or low), and brief justification
3. Resolve disagreements through discussion-based consensus
4. Studies classified as low-confidence require full-text review
5. Track and report the number of discordant and low-confidence cases
Sensitivity Analysis — Domain Classification
1. Restrict to high-confidence classifications only
2. Compare domain distribution (% Hardware, % Software, % Hybrid) with full dataset
3. Conduct independent verification of all classifications
4. Report number and direction of any reclassifications
Step 5: Chart the Data
Data Charting Form
Develop a standardised data charting form with the following variables:
| Variable | Description |
|---------------------------|-----------------------------------------------------------------------------|
| Author(s) | All authors |
| Publication year | Year of publication |
| Country | Country of first author affiliation |
| Journal/Conference | Publication venue |
| Study design | Prototype/Development, Experimental/Evaluation, Simulation study, RCT, Review, Technical/Methodological, or Not classified |
| Domain | Hardware, Software, or Hybrid (from Step 4) |
| Clinical application domain | Primary clinical area (assigned inductively) |
| Manikin/simulator type | Type of manikin or simulator described |
Pilot Testing
Pilot-test the data charting form on 10 studies. Assess whether all variables are extractable, the form is complete, and coding instructions are clear. Revise as needed before applying to the full dataset.
Clinical Application Domain Coding
Clinical application domains are assigned inductively based on the primary clinical area addressed in each study. Studies addressing multiple clinical areas are coded using the primary application as stated by the authors.
Abstract Recovery
If abstracts are not available for all included studies:
1. Stage 1: Retrieve from primary search platform
2. Stage 2: Query PubMed E-utilities API
3. Stage 3: Manual full-text retrieval through institutional library access, direct author contact, and open-access repositories
Target: 100% abstract coverage.
Step 6: Synthesise Results
Descriptive Numerical Summary
Tabulate included studies by:
- Domain (Hardware, Software, Hybrid)
- Publication year and temporal period
- Geographic distribution (country of first author)
- Study design
- Clinical application domain
- Key technology type
Cross-Tabulations
- Domain × publication period (Create Table 2)
- Domain × clinical application area (Create Figure 3)
Convergence Indicators (RQ3)
To assess the state of convergence between simulation and robotics communities, examine:
1. Disciplinary asymmetry: Hardware-to-software ratio across time periods
2. Cross-disciplinary tool adoption: Count studies explicitly describing AI/ML and robotics middleware (ROS)
3. Venue fragmentation: Number of unique publication venues; presence/absence of a dominant venue
4. Cross-domain clinical coverage: Proportion of clinical areas with studies from 2+ domains; identify areas addressed by all 3 domains
5. Hybrid domain trajectory: Trend in Hybrid study output over time; citation metrics by domain
Narrative Thematic Analysis
Organise narrative synthesis around the three research questions:
- RQ1: Nature and extent of research (domain distribution, study designs, technology landscape, clinical application areas)
- RQ2: Temporal trends and geographic patterns (publication growth, geographic concentration, centres of activity, venue distribution)
- RQ3: Convergence indicators (disciplinary asymmetry, absence of cross-disciplinary tools, venue fragmentation, cross-domain clinical coverage, Hybrid domain as convergence signal)
Sensitivity Analyses
Report results of both sensitivity analyses:
1. Inclusion threshold: Number of studies affected by varying from 2/3 majority to unanimous agreement; whether any findings change materially
2. Domain classification confidence: Domain distribution when restricted to high-confidence classifications only; number and direction of reclassifications
Step 7: Report According to PRISMA-ScR
Required Elements
Ensure the manuscript includes all 22 PRISMA-ScR checklist items (Tricco et al., 2018). Key elements:
- Structured abstract (Background, Methods, Results, Conclusions)
- PRISMA-ScR flow diagram (Figure 1)
- Complete search strategy (Additional file 1)
- Completed PRISMA-ScR checklist (Additional file 3)
- Data charting form with all extracted data (Additional file 2)
Mandatory Declarations
Include the following statements:
- Ethics approval and consent to participate (not applicable for scoping reviews)
- Availability of data and materials
- Competing interests
- Funding source
- Author contributions
- Protocol registration status
Protocol Registration
Register this protocol on protocols.io and cite the DOI in the manuscript Methods section.
Protocol references
1. Arksey H, O'Malley L. Scoping studies: towards a methodological framework. Int J Soc Res Methodol. 2005;8(1):19–32.
2. Levac D, Colquhoun H, O'Brien KK. Scoping studies: advancing the methodology. Implement Sci. 2010;5:69.
3. Peters MDJ, Godfrey C, McInerney P, Munn Z, Tricco AC, Khalil H. Chapter 11: Scoping reviews. In: Aromataris E, Munn Z, editors. JBI Manual for Evidence Synthesis. Adelaide: JBI; 2020.
4. Tricco AC, Lillie E, Zarin W, et al. PRISMA Extension for Scoping Reviews (PRISMA-ScR): checklist and explanation. Ann Intern Med. 2018;169(7):467–473.
5. Byrt T, Bishop J, Carlin JB. Bias, prevalence and kappa. J Clin Epidemiol. 1993;46(5):423–429.
6. Sim J, Wright CC. The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Phys Ther. 2005;85(3):257–268.
7. National Research Council. Convergence: Facilitating Transdisciplinary Integration of Life Sciences, Physical Sciences, Engineering, and Beyond. Washington, DC: National Academies Press; 2014.