3d anatomy models and impact on learning a review of the

Available online at www.sciencedirect.com

Health Professions Education 2 (2016) 80–98

http://dx.doi.org2452-3011/& 20article under the

Peer review unCorrespondin

Arabia. Tel.: +9E-mail addre

www.elsevier.com/locate/hpe

3D Anatomy Models and Impact on Learning: A Reviewof the Quality of the Literature$

Samy A. Azera,b,n, Sarah AzercaKing Saud University, College of Medicine, Department of Medical Education, PO Box 2925, Riyadh 11461, Saudi Arabia

bAustralian Professional Teaching, 3016 VIC, AustraliacBox Hill Hospital, Eastern Health Melbourne, Australia

Received 22 April 2016; accepted 13 May 2016Available online 24 May 2016

Abstract

Background: The aims of this study were to identify studies exploring three-dimensional (3D) anatomy models and their impacton learning, and to assess the quality of research in this area.Methods: PubMed, EMBASE, and the Web of Knowledge databases were searched using the following keywords "3D anatomy","three dimensional anatomy," "3D virtual reality anatomy," "3D VR anatomy," "3D anatomy model, “3D anatomy teaching", and“anatomy learning VR” . Three evaluators independently assessed the quality of research using the Medical Education ResearchStudy Quality Instrument (MERSQI).Results: Of the 94,616 studies identified initially, 30 studies reported data on the impact of using 3D anatomy models on learning.The majority were of moderate quality with a mean MERSQI score¼10.26 (SD 2.14, range 6.0–13.5). The rater intra-classcorrelation coefficient was 0.79 (95% confidence interval 0.75–0.88). Most studies were from North America (53%), and Europe(33%) and the majority were from medical (73%) and Dental (17%) schools.Conclusions: There was no solid evidence that the use of 3D models is superior to traditional teaching. However, the studiesvaried in research quality. More studies are needed to examine the short- and long-term impacts of 3D models on learning usingvalid and appropriate tools.& 2016 King Saud bin Abdulaziz University for Health Sciences. Production and Hosting by Elsevier B.V. This is an open accessarticle under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Three dimensional anatomy; Anatomy teaching; 3-D models; MERSQ instrument; Research quality

1. Introduction

With the introduction of reformed curricula inmedical, dental and other allied health schools, mostschools have reduced the total hours allocated foranatomy teaching and laboratory practical hours. These

/10.1016/j.hpe.2016.05.00216 King Saud bin Abdulaziz University for Health ScienceCC BY-NC-ND license (http://creativecommons.org/licens

nder the responsibility of King Saud bin Abdulaziz Univerg author at: Department of Medical Education, College of M6614699178.ss: [email protected] (S.A. Azer), drsarahazer@g

changes have been a continuous debate and triggeredthe emergence of innovative teaching and learningstrategies in order to maximize students' learning ofanatomy in the new context.1

Anatomy is a discipline where spatial visualization isof importance. Students need to learn not just

s. Production and Hosting by Elsevier B.V. This is an open accesses/by-nc-nd/4.0/).

sity for Health Sciences.edicine, King Saud University, PO Box 2925, Riyadh 11461, Saudi

mail.com (S. Azer).

www.sciencedirect.com/science/journal/24523011

http://dx.doi.org/10.1016/j.hpe.2016.05.002

http://crossmark.crossref.org/dialog/?doi=10.1016/j.hpe.2016.05.002&domain=pdf

www.elsevier.com/locate/hpe




mailto:[email protected]

mailto:[email protected]

S.A. Azer, S. Azer / Health Professions Education 2 (2016) 80–98 81

anatomical structures and functions but also spatialrelationships to surrounding structures.2 While anat-omy textbooks and anatomy atlases provide two-dimensional (2D) static anatomical illustrations, theyare of limited value in exposing three-dimensional (3D)dynamics of anatomical structures.3 Learners may findit difficult to visualize 2D images as 3D and understandcertain dynamic aspects of functional anatomy. Forexample, identifying the structures related to thecaudate lobe when the liver is moved to differentplanes/positions.

In anatomy, students have to rotate and manipulatestructures from various views to identify anatomicalstructures. Visual-spatial ability has been defined as theability to mentally manipulate objectives in three dimen-sions.4 Such ability is important for medical students tounderstand anatomical structures and is also important tosurgical trainees and surgeons. Therefore, the ability tovisualize and mentally manipulate 3D structures andcorrectly identify them and related structures is animportant skill to medical and dental students when theanatomy is presented in various planes and positions.5

Research in this area may not only assess visual–spatialabilities of students during learning but also assess thedevelopment of a pedagogical technology to enhancestudents’ learning skills and the advancement of medicaltraining.6

Considering the importance of 3D learning andteaching anatomy models in medical, dental and otherallied health curricula, understanding the range ofmodels used and the impact of using such 3D strategieson students' learning necessitates a revision of studiescovering 3D models. The focus is to evaluate differentfactors affecting learning by using 3D anatomy modelsand their impact on the learning process. This reviewaimed at (1) assessing the impact of using 3D anatomymodels on student’s learning using a scoping reviewapproach,7 and (2) assessing the quality of research ofstudies identified using Medical Education ResearchStudy Quality Instrument (MERSQI) for quantitativestudies.8,9 The MERSQI is a multifaceted instrumentfor assessing the quality of medical education studiesand has been shown to have reliability and validityevidence with high inter-rater and intra-rater reliabil-ity8–10 and it has been shown that MERSQI scorescorrelate with expert ratings about study quality.11 Instructuring this review, the findings from the studies areintegrated using critical analysis and thematic synth-esis.12 Gaps in this area and future directions ofresearch have been identified and discussed.

2. Methodology

2.1. Study design

This review is based on a scoping review approach, amethod commonly used to explore questions whenlittle knowledge has been identified. Therefore, the fivestages summarized by Arksey and O’Malley werefollowed. Stage 1: identifying the research question,stage 2: identifying relevant studies, stage 3: studyselection, stage 4: charting the data, and stage 5:collating, summarizing and reporting the results.7

2.2. Study stages

The five stages used in the study can be summarizedas follows.

2.2.1. Stage 1: identifying the research questionThe research questions of this review are:

� What 3D methods used in teaching and learninganatomy?

� What is the impact of using 3D anatomy models onstudent’s learning?

� What is the quality of published research?

2.2.2. Stage 2: identifying relevant studies in theliterature

The following electronic databases: PubMed,EMBASE, and the Web of Knowledge were searched.The keywords used in the search were: "3D anatomy","three dimensional anatomy," "3D virtual reality anat-omy," "3D VR anatomy," "3D anatomy model, “3Danatomy teaching" and “anatomy learning VR” . Thesearch was conducted by the author (who is trained as adoctor and is a professor of medical education) and tworesearch assistants (both have medical background) from01 to 15 January 2015. Papers available online ahead ofthe print version were included.

Three-dimensional (3D) anatomy models comprisedigital, and non-digital (physical) models that can bemoved into different positions/planes to enable learnerto learn the relationship between different anatomicalstructures in space and mentally manipulate objectivesin three dimensions. Visual-spatial ability also knownas spatial visualisation ability is defined as the ability tomentally manipulate objectives in two- and three-dimensional figures,4 while virtual reality (VR) is acomputer-simulated environment that can simulate real

S.A. Azer, S. Azer / Health Professions Education 2 (2016) 80–9882

world or imaged worlds. Most current virtual realityenvironments are displayed either as a computer screenor with special stereoscopic displays.4

Second, using the same search words, the webpages ofthe following medical education journals were searched“Academic Medicine”, “Medical Education”, “MedicalTeacher”, “BMC Medical Education”, “Advances inHealth Sciences Education”, “Teaching and Learning inMedicine”, “European Journal of Dental Education”, and“Journal of Dental Education”.

Third, the webpages of the following journals onanatomy and anatomy education were searched: “Anato-mical Science International”, “Journal of anatomy”,“Surgical and Radiologic Anatomy”, “The AnatomicalRecord”, “Anatomical Science Education”, “ClinicalAnatomy” and “Annals of Anatomy”. The same searchwords were used in searching the journals websites. Othersources for eligible studies were the list of references ofrelated systematic reviews and research papers identifiedin this search. Because there were a few papers publishedon the research topic during the years 2000–2004, it wasdecided to focus the search for papers from 01 January2000 up to the end of December 2014.

2.2.3. Stage 3: inclusion and exclusion criteria andstudy selection

Studies were included if they were assessing the impactof using 3D anatomy models on student’s learning. Onlystudies in the English language that addressed 3Danatomy in undergraduate medical, dental and alliedhealth courses were included.

The exclusion criteria were as follows: (1) descriptivestudies on the use of 3D models in teaching withoutassessing the impact on learning, (2) studies descripting thedevelopment of 3D models, (3) studies on the use of 3D inveterinary medicine, (4) the use of 3D in planning surgical,orthopedic or anesthetic procedures, (5) advanced 3Dmodels used in enhancing surgical skills, and advancedtraining, (6) use of 3D in understanding malformation, rareconditions or design of surgical technique, (7) history of3D development, (8) use of 3D models in studyingcomparative anatomy, (9) abstracts, conference proceed-ings, discussion papers on 3 technologies and grossanatomy, (10) reviews, commentaries, debates, letters tothe Editors, editorials on using 3D models in anatomy,(11) papers in languages other than English, and (12)duplicate papers.

The assessors independently reviewed all databasesusing the above stated criteria. Papers identified wereplaced on an Excel sheet (Microsoft Corporation, Red-mond, WA, USA). Each evaluator applied the exclusioncriteria independently. Disagreements were discussed in a

meeting until a final consensus was reached. Copies ofthe full articles were obtained for the studies that appearedto represent the “best fit” with the research questions.

2.2.4. Stage 4: charting the dataThe data were charted on spread sheets to record the

title of the papers, authors, country of the first author/university where work was done, year of publication,journal/source, type of study, aims/objectives, what wascarried out, and other key information about results/conclusions. Data were summarized using numericalsummary. The numerical summary included number ofparticipants, and key information found.12

2.2.5. Stage 5: collating, summarizing and reportingthe results

Systematic examination of the methodological rigor ofthe studies enabled the grouping of the studies. Extracteddata were synthesised descriptively to map differentaspects of the literature as outlined in the study keyquestions. Studies were grouped according to country oforigin of the first author, settings and study participants,study design and research methodology used, outcomesmeasured, and the year of publication.

2.3. MERSQI scoring of studies

The MERSQI has been used in assessing the quality ofpublished medical education research.8 The instrumentcomprises 10 items in six domains: study design, samplingmethod (number of institutes and response rate), type ofdata, validity of the evaluation instrument, data analysis,and outcomes. For each domain the maximum score is3 and the total possible MERSQI score is 18 with a rangeof 5–18.9 Higher scores indicate that the study is welldesigned, has valid instruments and provided measurableoutcomes beyond gain of knowledge and skills. The use ofthe MERSQI in this study follows the study by Reed etal.8 and was carried out at the following stages: First, threeevaluators (the author and two research assistants)reviewed the use of the instrument and practiced its useindependently to assess seven studies other than thoseincluded in the study. Second, the scores obtained werereviewed in a meeting with the aim to clarify the meaningof each item in the scoring scheme, orient the researchersto the use of the instrument and improve consistencyamong all raters. The raters also documented theirdecisions about applying MERSQI coding in a writtenmanual for future references. Third, the 30 studies wererated independently by the three evaluators using theMERSQI scoring manual. The scores obtained from thethree evaluators were used to calculate The rater intra-class


correlation coefficient (IIC) to determine inter-rater relia-bility.8–11

3. Results

3.1. Studies included in this review

The search results of PubMed, EMBASE and theWeb of Knowledge databases yielded 90,393 articlesand 4223 articles were identified from searching eightjournals on medical and dental education and sevenanatomy journals (Fig. 1). The results were as follows:Academic Medicine 21, Medical Education 27, Med-ical Teacher 40, BMC Medical Education 10,Advances in Health Education 19, Teaching andLearning in Medicine 6, European Journal of DentalEducation 22, and Journal of Dental Education 461,

Fig. 1. Results of search strategies and selection procedure for a systemati

Anatomical Science International 266, Journal ofAnatomy 698, Surgical and Radiologic Anatomy 501,The Anatomical Record 914, Anatomical ScienceEducation 249, Clinical Anatomy 651, Annals ofAnatomy 306. Other searches from the list of refer-ences yielded 32 articles—making a total of 4223.After excluding duplicates, a total of 4807 articles wereidentified. On applying the exclusion and inclusioncriteria by two evaluators independently, a total of 83papers were assessed for eligibility. Finally after read-ing the full text of articles, 53 articles were excludedmaking a total of 30 articles meeting the criteria forsynthesis in this review.13–42

Most studies were conducted in North America (16/30; 53%), and Europe (10/30; 33%). The remainingwere from Asia (3/30; 10%), and Australia (1/30; 3%).No studies from Africa or South America were found

c review for studies on 3D anatomy models and impact on learning.


(Table 1). The majorities of the studies (15/30, 50%)were published in 2009–2014 and (9/30; 30%) werepublished in 2009–2011. Only (5/30,17%) were pub-lished in 2006–2008 and (1/30, 3%) were published in2000–2005 (Table 1).

3.2. The 3D anatomy models

The 3D digital models in the studies included in thisreview can be classified into: (i) 3D web-based models(3/30; 10%),14,18,30 (ii) 3D computer- and mobile-basedmodels (21/30; 70%),13,15–17,19,20,22,24–27,31–38,40,41 and(iii) non-digital (physical) 3D anatomy models (6/30;20%). These physical models comprised: clay

(

Table 1Summary of characteristics of the 30 papers included in thesystematic review on 3D anatomy models.

Studies Number (%) References

Place of thestudyAfrica 0 (0) 0Asia 3 (10) 20,21,40Australia 1 (3) 27Europe 10 (33) 15,19,22,25,29,31,34,37,38,42NorthAmerica

16 (53) 13,14,16–18,23,24,26,28,30,32,33,35,36,39,41

SouthAmerica

0 (0) 0

3D models3D Web-basedmodels

3 (10) 14,18,30

3D computer-and mobile-based models

21 (70) 13,15–17,19,20,22,24–27,31–38,40,41

3D non-digital(physical)models

6 (20) 21,23,28,29,39,42

Study typeControlled 7 (23) 20,25,28,32,34,39,41Crossover 3 (10) 19,29,33Quasiexperimental

5 (17) 21,27,36,38,40

Prospectivestudy

1 (3) 30

Randomizedcontrolled

14 (47) 13–18,22–24,26,31,35,37,42

Publication year2012–2014 15 (50) 28–422009–2011 9 (9) 19–272006–2008 5 (17) 14–182003–2005 0 (0) 02000–2002 1 (3) 13

models,21,28 laparoscopic dissection,42 arthroscopicexamination,29 anatomy glove learning system,39 andcolour coded models.23 Interestingly some of the 3Ddigital anatomy models were integrated tools coveringgross anatomy and related radiological knowledge.27,40

3.3. Participants’ schools

The participants were from Medicine (22/30,73%)13–16,18,19,21–26,29,31–37,40,42 and Dentistry (5/30,17%).20,27,30,33,35 The remaining studies were fromKinesiology.28,33 Only one study was from each of thefollowing allied health schools: Health Sciences,33

Massage therapy,39 Occupational therapy,33 Phy-siotherapy,33 Psychology,17 and Social science.41 Somestudies had participants from 2 or more institutes,34 orfrom several schools33,35 (Table 2).

3.4. Impact of using 3D anatomy models on learning

To carefully evaluate the impact of using 3D modelson learning, critical evaluation of the studies isaddressed under the following points:

1) Factors affecting learning by using 3D anatomymodels: A number of factors have been shown toaffect learning by using 3D models. These factorsare:
●
●

●

Factors related to the 3D model such as the designof the 3D software, availability of visual andauditory information simultaneously, cognitiveload and complexity of the tasks included in theprogram, and innate difference from traditionalteaching methods,33,39 multiple orientations pro-duced by the software.13

Factors related to learner characteristics such asinnate visual spatial ability,13,17,32 learner’s trans-formation and search skills,41 prior orientation tothe 3D technology,36 student’s personal home-work,38 cognitive load on the learner,35 gender ofthe student,15,16 and degree of learner control.17

Although not all these characteristics can bemodified, identification of learner’s aptitudesand trends may better facilitate teaching andlearning approaches.Factors related to the curriculum and the learningenvironment such as integration of the 3D toolwith other components in the curriculum, timeallocated to the use of the 3D tool, availability offaculty to provide feedback to students whenneeded, anatomical region studied,16 stage oflearning in the curriculum,18 and the use of 3D

Table 2Impact of using anatomy three-dimensional educational tools on student’s learning.

Author, year[Reference]

3D teaching toolused

University/Hospital,country

Type ofstudy

Research question/aim Participants (number) What was carried out/measured Results/limitations

Garg et al.,200213

3D computeranatomy model

McMasterUniversity,Canada.

Randomisedcontrolledstudy

Examining the role of multipleorientations in learninganatomy from 3D computermodels.

Undergraduate medicalstudents (n¼87)

Students were allocated toeither multiple-view (MV)group (any of 36 possibleangles) of rotation or a key-view and wiggle group(KVþW where students cancontrol six views near 0 and180 degree on computerworkstation.

Certain key or canonicalviewpoints of an object arecritically important for spatiallearning. Multiple orientationsprovided by the computer-based anatomy software mayoffer minimal advantage tosome learners.

Nicholsonet al.,200614

3-D anatomical earmodel

McGillUniversity,Montréal,Canada.


Test the educationaleffectiveness of a computergenerated 3-D model on theear and inner ear.

First-year medical students(n¼60)

The intervention groupcompleted a Web-based tutorialon ear anatomy that includedthe interactive model, while acontrol group took the tutorialwithout exposure to the model.At the end, both groupsanswered15 quiz questions toevaluate their knowledge of3-D relationships within theear.

The intervention group’s meanon the quiz was higher than themean score for the controlgroup. The difference wassignificant.

Guillotet al.,200715

Visuo-spatialrepresentationmodel.

UniversitéClaude BernardLyon, France.

Randomizedcontrolledstudy

Investigated the relationshipbetween visuo-spatialrepresentation, mental rotation(MR) and functional anatomyexamination results.


Students completed the GroupEmbedded Figures Test(GEFT), Mental Rotation Test(MRT) and Gordon Test ofVisual Imagery Control. Thetime spent on personalassignment was alsoconsidered.

Men scored better than womenon both GEFT and MRT, butthe gender effect was limited tothe interaction with MRTability in the anatomy learningprocess. Significantcorrelations were foundbetween visuo-spatial, MRabilities, and anatomyexamination results.

Hisley et al.,200816

Direct and indirect3D digital models.

Des MoinesUniversity,Iowa, USA.


Compare physical dissectionusing an embalmed cadaverand digital dissection using 3Dvolume modelling of wholebody.

Undergraduate top first-yearmedical students (n¼16)

The physical dissectorsproceeded with their directmethods, whereas the digitaldissectors generated andmanipulated indirect 3D digitalmodels. After 6 weeks,corresponding studentanatomical assignment teamscompared their results using

All students, regardless ofgender, dissection method, andanatomical region dissectedperformed significantly betteron questions presented asrotating models requiringspatial ordering or viewpointdetermination responses in

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

85

Table 2 (continued )




Type ofstudy


photography and animateddigital visualizations.

contrast to requests for specificlexical feature identifications

Levinsonet al.,200717

VR brain anatomy McMasterUniversity,Ontario,Canada.


Determine the effects oflearner control over thee-learning environment andkey views of the brain versusmultiple views in the learningof brain surface anatomy.

First-year psychology students(n¼120)

Two phases for interventionwere described. Main outcomemeasure was 30-item post-testof brain surface anatomystructure identification.

High degree of learner controlmay reduce effectiveness oflearning. Multiple views mayimpede learning particularly forthose with relatively poorspatial ability.

Marsh et al.,200818

3D models ofembryonicdevelopment.

University ofCincinnatiCollege ofMedicine, Ohio,USA.


To assess the effectiveness ofusing web-based learningmodule that combines 3Dgraphics and 3D models ofembryonic development onstudents’ learning.


Both groups attended lecture(s) on embryonic foldingwhereas only students in thestudy group were allowedaccess to the 3D module viaBlackboard. Both groupscompleted the same 14-question quiz. Results fromboth groups were analysed.

Students who used the moduleperformed better than thosegiven only traditionalresources. The findings suggestthat the 3D computer-assisted-instruction modules in generalare more useful if used towardthe later stages of learning,rather than as an initialresource.

Donnellyet al.,200919

Virtual HumanDissector (VHD)software

University ofDurham, UK.

Cross-overstudy

Investigate the use of VHD infacilitating students’ ability tointerpret cross-sectionalimages and understand therelationships betweenanatomical structures.

First-year medical students(n¼89).

Using a crossover design, theinvestigation was undertakenas two 20-minute self-directedlearning activities using VHDin a computer suite andprosections and models in thedissecting room, interspersedbetween 3 tests identifyinganatomical structures (pre-,mid- and post-session).

There were no significantdifferences between the twogroups at any tested stage.

Hu et al.,200920

Dental 3DMultimediaSystem (D3DM)

WuhanUniversity,Hubei, China.

Controlledstudy.

Investigate the effects ofintroducing a softwareprogram, named the Dental 3DMultimedia System (D3DM),into the education of a groupof junior dental students intheir preclinical practice.

Undergraduate dental students(n¼53)

One group received theirtraining program in thetraditional way, unassisted by3D technology. The secondgroup received their trainingprogram in the traditionalmanner, but also used theD3DM to supplement theireducation.

The D3DM-assisted groupworked faster, and no worsethan the traditional groupduring the training in labs.

Oh et al.,200921

Clay models inlearning anatomy

SungkyunkwanUniversitySchool of

Quasi-experimentalstudy.

Ask students for their viewsabout the use of clay models inlearning anatomy and assess

Undergraduate medicalstudents (n¼70).

Two feedback surveys werecompleted by students. Toassess the impact on learning,

Clay modelling appears to be auseful supplement toconventional anatomy and

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

86

Medicine,Suwon, Korea.

the effectiveness of thismethod.

students who used the claymodels and students fromanother school that did not useclay models were examinedtwice by each school; onceafter finishing the grossanatomy course and six monthslater.

radiologic anatomy education.Students’ average scores on CTexaminations were higher thanthat of a group that did not useclay models.

Abid et al.,201022

3D model ofteaching peritonealembryogenesis.

UniversitiéParis v, Paris,France.


Compare 3D and traditionalchalk teaching efficiency interms of student memorizationconcerning peritonealembryogenesis.

Medical students from twouniversities (n¼165).

Students from two universitieswere taught peritonealembryogenesis either via a 3Dtechnique (interactive DVDROM) or via the traditionalchalk technique. Both groupswere subjected to an evaluationtest including 34 questions

The 3D technique issignificantly more efficient thanthe traditional chalk techniquefor the teaching of peritonealembryogenesis. However, moreassessment is neededparticularly on long-term basis.

Estevezet al.,201023

3D neuroanatomytool

BostonUniversityUSA.


Implement and evaluate a newtool for teaching 3Dneuroanatomy.

First year Undergraduatemedical students (n¼101).

Students were taughtneuroanatomy according totraditional 2D methods. Then,during laboratory review, theexperimental group constructed3D color-coded physicalmodels of the periventricularstructures, while the controlgroup re-examined 2D braincross-sections.

3D physical modelling is aneffective method for teachingspatial relationships of brainanatomy and will better preparestudents for visualization of 3Dneuroanatomy. Limitations:Limited to a small area inneuroanatomy. Based on oneschool.

Hu et al.,201024

A three-dimensional (3D)educationalcomputer model ofthe anatomy of thelarynx

University ofWesternOntario, Canada

Randomisedcontrolledtrial.

Evaluate a 3-d method ofteaching laryngeal anatomy.


The primary outcome measurewas the score on a 20-questionlaryngeal anatomy test; thesecondary outcome measurewas a student opinionquestionnaire.

The 3D educational computermodel of the larynx was notshown to be superior to writtenlecture notes in its efficacy inteaching anatomy. Limitations:The study is based on thelarynx only and conducted inone school.

Codd andChoudhury,201125

Three-dimensionalvirtual realitycomputer models.

University ofManchester,UnitedKingdom

Controlledstudy

Evaluate the use of 3D virtualreality when compared withtraditional anatomy teachingmethods.

Second year Undergraduatemedical students (n¼39).

Three groups were identified:(i) a control group (no priorknowledge of forearmanatomy), (ii) a traditionalmethods group (taught usingdissection and textbooks), and(iii) a model group (taughtsolely using e-resource).

Virtual reality anatomylearning can be used tocompliment traditionalteaching methods effectivelyLimitations: the groups wereassessed on anatomy of theforearm only by using tenquestions.

Keedy et al.,201126

Interactive 3Dpresentation ofliver and biliaryanatomy

University ofCalifornia–SanFrancisco,USA.


Determine whether aninteractive 3D presentationdepicting liver and biliaryanatomy is more effective for


Participants were randomizedinto two groups: 3D group:presented with a computer-based interactive learningmodule comprised of

While the interactive 3Dmultimedia module receivedhigher satisfaction ratings fromstudents, it neither enhancednor inhibited learning of

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

87





Type ofstudy


teaching than a traditionaltextbook

animations and still images tohighlight various anatomicalstructures, or 2D group:presented with a computer-based text containing the sameimages and text withoutinteractive features

complex hepatobiliaryanatomy. Limitations: Basedon one school.

Vuchkovaet al.,201127

3D visualisationsoftware in oralradiography

University ofQueensland,Australia


Investigate the effect of 3Dvisualisation software onstudents’ learning of oralradiographic interpretationfrom 2D radiographic images.

Fourth-year dental students(n¼59).

Students were trained in oralradiographic interpretation byusing the software. Theassessment of the interventionincluded a radiologicinterpretation test and astructured Likert-scale survey.

Quantitative assessment ofstudents did not showimprovement in theirradiographic interpretation testafter training using the 3Dvisualisation software.However, students werepositive about the 3D programas per the survey.

Bareitheret al.,201328

Clay modeling University ofIllinois atChicago, USA.

Controlledstudy

Comparing clay modeling towritten modules to determinethe degree of improvement inlearning and retention ofanatomical relationships.

Undergraduate Kinesiologystudents (n¼39).

Clay and module groupsparticipated in weekly one-hour classes using either claymodules or answering writtenquestions, respectively. Controlgroup received no intervention.Assessment included: pre- andpost-assessment and betweenpost-assessment and retentionexaminations.

No significant differences wereseen between interventions orlearning preferences in anygroup

Knobe et al.,201229

Arthroscopyversus ultrasound.

RWTH AachenUniversity,Germany

Cross-overstudy.

Whether musculoskeletalultrasound or arthroscopicmethods can increase theanatomical knowledge uptake.

Second year, undergraduate,medical students (n¼242)

Comparing musculoskeletalultrasound vs arthroscopicmethods. Both groups alsolearnt anatomy via dissection.The control group only haddissection.

Arthroscopy may be attractiveto students. Ultrasound seemsto be inferior to thearthroscopic, and is regardedby students as more difficult tolearn. Limitation: Limited toone school, small number ofstudents, did not measure longterm impact.

Maggioet al.,201230

Interactive mediain dentalmorphology

University ofPennsylvania,USA.

Prospectivestudy

Analyze the introduction ofonline independent learningmodule for dental morphologyinstruction.

First-year dental students(n¼118).

One-third of students weregiven an interactive mediamodule for dental anatomyinstruction. The remainingstudents experienced the

The interactive module wasjust as effective as thetraditional classroom method.However, the online module

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

88

traditional course mainlylectures. At the end of themodule, a written examinationand survey were given to bothgroups.

positively engaged thestudents.

Metzleret al.,201231

3D presentation University ofHeidelberg,Heidelberg,Germany.

Randomizedcontrolledstudy.

Evaluates whether training on3D presentation enhances theunderstanding of 2D images.


A teaching module was usedconsisting of one learning partand two examination parts(EP). Students wererandomized to training witheither 2D or 3D.

The correct interpretation of2D imaging does not differ instudents trained in either 3D or2D.

Nguyenet al.,201232

A 3D computergenerated visualrepresentation of agroup ofanatomicalstructures (theaorta, trachea, andesophagus).

University ofWesternOntario,Ontario,Canada.

Controlledstudy

Assess factors that influencespatial anatomycomprehension.

Medical students, staff andfaculty members (n¼60).

Participants studied a group ofanatomical structures in one ofthree visual conditions (control,static, dynamic) and one of twointeractive conditions(interactive, non-interactive).Before and after the studyphase, participants’comprehension of spatialanatomical information wasassessed using a multiple-choice spatial anatomy task(SAT) involving the mentalrotation of anatomicalstructures.

Visual ability (VA) had apositive influence on SATperformance but instructionwith different computervisualization could modulatethe effect of VZ on taskperformance.

Roach et al.,201233

3-D videography University ofWesternOntario, Canada

Cross-overstudy

Assess the efficacy of 3Dvideo as a medium to supportthe acquisition of complexsurgical skills. The evaluationwas carried out using a globalrating scale.

Undergraduate students fromallied health sciences includingmedicine, dentistry,kinesiology, occupationaltherapy, and physiotherapy(n¼43).

Students were assigned arandom numeric label for theduration of the study.Following the completion ofthe Mental Rotation Test(MRT), participants wererandomly assorted to one offour groups and given 15minutes to view an 8 min longvideo (2D). Following viewingof the video participants wereasked to perform the surgicalskill they had just viewed. Thesame process was repeatedusing another randomisedsurgical video (3D).

The study did not findsignificant differences orenhanced surgical skills.Limitations: More sensitivescales of measurements mayneed to be used in futurestudies.

Ruisotoet al.,201234

3-Dimensionalneuroimagingmodel.

University ofSalamanca,Salamanca,Spain.

Controlledstudy

Whether 3D volumetricvisualization helps learners toidentify and locate subcorticalstructures more precisely than

Participants were volunteersrecruited from differentuniversities and medicalcenters located in CentralSpain, Northwestern Spain,

Eighty participants wereassigned to each experimentalcondition: 2D cross-sectionalvisualisation vs 3D volumetricvisualization. Accuracy in

3D volumetric visualizationhelps to identify brainstructures such as thehippocampus and amygdala,more accurately and rapidly

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

89





Type ofstudy


classical cross-sectionalimages (2D).

and Northeastern Spain. Theparticipants level of educationwas classified under twocategories: novices and experts(n¼80).

identifying brain structures,execution time, and level ofconfidence in the responsewere measured.

than conventional 2Dvisualization

Khot et al.,201335

Virtual reality(VR) computer-based modelversus staticcomputerisedmodels (KV)versus plasticmodels.

University ofWesternOntario,Ontario,Canada.


Examine the effectiveness ofthe three formats of anatomylearning.

Undergraduate medical anddental students (n¼60).

Students had ten minutes tostudy the names of 20 differentpelvic structures. The outcomemeasure was a 25 item shortanswer test consisting of 15nominal and 10 functionalquestions, based on a cadavericpelvis. All subjects also took abrief mental rotations test(MRT) as a measure of spatialability, used as a covariate inthe analysis.

Computer-based learningresources appear to havesignificant disadvantagescompared to traditionalspecimens in learning nominalanatomy.

Tworeket al.,201336

LINDSAY VirtualHuman Project

University ofCalgary,Calgary,Canada.


Identify possible factors thatcan affect expectations andsuccessful implementation of a3D computer-assisted learningsoftware (LINDSAY).

Faculty and second yearmedical students (n¼180).

A validated tool measuringimpact across pedagogy,resources, interactivity, andfactors outside the immediatelearning event was used inconjunction with observation,and focus groups to criticallyexamine the impact of attitudesand perceptions of allstakeholders in the earlyimplementation of LINDSAY.

External, personal mediausage, along with students'awareness of the need to applyanatomy to clinicalprofessional situations droveexpectations of LINDSAY(3D) Presenter.

Müller-Stichet al.,201337

3D presentation ofliver anatomy

University ofHeidelberg,Heidelberg,Germany


Define the impact of theaddition of key views to CTimages (2Dþ) and the use ofreal 3D (3Dr) on theidentification of liver anatomy


Medical students wererandomized to three groups(2Dþ or 3Dr or 3D) and askedto answer 11 anatomicalquestions and 4 evaluativequestions.

Students exposed to 3Dr and3D performed significantlybetter than those exposed to2D. There were no significantdifferences between 3D and

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

90

when compared with regular3D images (3D).

3Dr and no significant genderdifferences.

Hoyek et al.,201438

3-D digitalanatomy animation

UniversitéClaude BernardLyon,Villeurbanne,France

A quasi-experimentalstudy designby comparingtwo groupsfromdifferentclasses.

Assess the effectiveness of3-D animation method.

First year kinesiology students(n¼391).

The teacher used two-dimensional (2D) drawingsembedded into PowerPointslides and 3D digitalanimations for the first group(2D group) and the second (3Dgroup), respectively.

The findings supported that 3Ddigital animations wereeffective instructionalmultimedia material tools inteaching human anatomyespecially in recallingknowledge requiring spatialability.

Lisk et al.,201439

Anatomy GloveLearning System(AGLS)

University ofToronto,Ontario, Canada

Controlledstudy

Evaluate the effectiveness ofAnatomy Glove LearningSystem (AGLS).

Massage therapy students(n¼73).

Students were allocated intotwo groups and drew musclesonto either: (1) the glove usingAGLS instructional videos (3Dgroup); or (2) paper withpalmar/dorsal views of handbones during an instructor-guided activity (2D group). Aself-confidence measure andknowledge test were completedbefore, immediately after, andone-week following thelearning conditions.

AGLS and the traditional 2Dlearning approach are equallyeffective in promotingstudents’ self-confidence andknowledge of hand anatomy.

Murakamiet al.,201440

Anatomy-CT, a3D modelintegrating humananatomicaldissection withcomputedtomography (CT)radiology.

GunmaUniversityGraduateSchool ofMedicine,Maebashi,Japan.


Assess the impact of usingAnatomy-CT model onstudents’ learning.

Medical students and academicstaff (n¼126).

Students’ perspectives aboutthe project were evaluated byusing surveys. Academicperformance was evaluatedfrom the yearly trends of scoresfor anatomy and individualtypes of problems used inclasses. Correlations betweendifferent types of examinationswere calculated.

The method yielded positivestudent perspectives andsignificant improvements inradiology skills in later clinicalcourse.

Nguyenet al.,201441

Spatialvisualizationmodels

WesternUniversity,Ontario,Canada.

Controlledstudy

Test whether there are multiplestrategies used to solve thespatial task (SAT) and whetherthe strategy choice of high-versus low-visualization of

Undergraduate science andsocial science students (n¼42).

Forty-two students completed astandardised measure of spatialvisualization ability, a novelspatial anatomy task, and aquestionnaire involving

Understanding spatialvisualization ability is the mainsource of variation in spatialanatomy task performance,irrespective of strategy.

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

91

Tab

le2(contin

ued)

Autho

r,year

[Reference]

3Dteaching

tool

used

University

/Hospital,

coun

try

Typ

eof

stud

yResearchquestio

n/aim

Participants

(num

ber)

Whatwas

carriedout/m

easured

Results/limitatio

ns

individu

alsinfluences

task

performance.

personal

self-analysisof

the

processesandstrategies

used

while

performingthespatial

anatom

ytask.

Participants

with

high

spatial

visualizationability

weremore

accurate

whensolvingthetask

prob

lem.

tenBrink

eet

al.,

2014

42

3D-Laparoscopic

dissectio

nErasm

usMedical

Center,

Rotterdam

,the

Netherlands

Randomised

controlled

stud

y

Examinetheaddedvalueof

dissectio

n-basedteaching,

usingmod

elsof

theingu

inal

region

inem

balm

edspecim

ens.

Und

ergraduate

medical

stud

ents

(n¼46

).Studentswererandom

lyassigned

tothreegrou

ps.

Group

Iattend

edlectures,

groupIIattend

eddissectio

n-basedtraining

using

laparoscopicdissectio

nmodels,

andgrou

pIIIattend

edlectures

aswellas

dissectio

n-based

laparoscopic

training.Students

wereexam

ined

immediately

attheendof

teaching

and

2weeks

afterinterventio

n.

Three-dim

ensional

anatom

yeducationwith

dissectio

nmod

elsenhances

anatom

ylearning

bymedical

students,

asevidencedfrom

theirtest

scores

intheshortandlong

-term

s.


models as part of self-directed learning activ-ities.19 Interestingly, different types of computervisualization might be effective for differentlearners,32 and therefore, could affect the learningprocess and the construction of knowledge.39

While one or two of these factors were examinedin some studies, there is a need for carefulassessment of each of these factors and under-standing strategies that can maximize learning byusing 3D tools.

A number of studies showed that volumetricvisualization improves the identification and loca-lization of anatomical structures by learners inboth morphological and functional images andimproved student’s performance in anatomicaltasks on short-term basis.14,18,21,22,34,37,38,42 Thestudy by Ruisoto et al.34 not only showedincreased accuracy in students’ performance butalso less response time to complete a task. Otherstudies found that 3D multimedia learning modelsenhance students' learning compared to traditionalteaching methods,26 and showed a relationshipbetween the use of 3D computer models andspatial abilities.13,17,32 Three studies showedenhancement of student’s performance on long-term basis.16,21,42

However, not all studies showed that 3Dmodels are better than 2D images or traditionalteaching. Khot et al.35 even showed thatcomputer-based modalities are not as effectiveas physical models of pelvic anatomy. Theauthors found that 2D pictures are as effectiveas virtual reality (VR) models. They concludedthat computer based learning 3D resources appearto have significant disadvantages compared totraditional specimens in learning normal anatomy.Similar findings were reported.13,17 Also thestudy by Lisk et al.39 showed that AnatomyGlove Learning System (AGLS) and the tradi-tional 2D learning approaches had the same effecton students’ self-perceived confidence and knowl-edge of hand anatomy. A number of studies alsoshowed that there was no differences when 3Doutcomes where compared to 2D or traditionalteaching.19,21,24,26–28,30,32,33 However, students inthese studies reported that learning by using 3Dmodels was more satisfactory when compared totextbooks images and traditional teaching mod-alities.20,25,26,29,34,36–38,40

(2) In two studies the participants were from morethan one school.33,35 Although 11 studies had 100

Table 3Summary of Medical Education Research Study Quality Instrument (MERSQI) domain and item scoresa for 30 studies on 3d anatomy models and their impact on learning.

Domain MERSQI item Studies no.(%)

Item possiblescore

Maximumdomain

Item mean(SD)

Domain mean(SD)

Study design 1. Study design 3 2.40 (0.67) 2.40 (0.67)Single group cross-sectional or single group posttest only 3 (10) 1Single group pretest and posttest 0 (0) 1.5Nonrandomized, 2 (or more) groups 12 (40) 2Randomized controlled trial 15 (50) 32. No. of institutions studied 3 0.56 (0.21) 0.85 (0.44)1 27 (90) 0.52 2 (6.6) 142 1 (3.3) 1.5

Sampling 3. Response rate,% (highest reported) 1.15 (0.41)Not applicable 0o50 or not reported 7 (23.3) 0.550–74 7 (23.3) 1Z75 16 (53.3) 1.5

Type of data 4. Type of data 3 2.20 (0.99) 2.20 (0.99)Assessment by study participant (e.g., self-reported data) 12(40) 1Objective measurement (e.g., OSCE, written exam) 18 (60) 3

Validity of evaluationinstrumentb

5. Internal structure 3 0.00 (00) 0.06 (0.25)

Not applicable 0Not reported 30 (100) 0Reported 0 (0) 16. Content (purposeful process to instrument development) 0.13 (0.34)Not applicable 0Not reported 26 (86.6) 0Reported 4 (13.3) 17. Relationships to other variable (criterion, predictive or discriminatevalidity)

0.50 (0.51)

Not applicable 0Not reported 15 (50) 0Reported 15 (50) 1

Data analysis 8. Appropriateness of analysis 3 0.66 (0.47) 0.96 (0.55)Data analysis inappropriate for study design or type of data 10 (33.3) 0

S.A.Azer,

S.Azer

/Health

Professions

Education

2(2016)

80–98

93

Tab

le3(contin

ued)

Dom

ain

MERSQIitem

Studies

no.

(%)

Item

possible

score

Maxim

umdo

main

Item

mean

(SD)

Dom

ainmean

(SD)

Dataanalysisappropriateforstud

ydesign

andtype

ofdata

20(66.6)

19.

Com

plexityof

analysis

1.26

(0.44)

Descriptiv

eanalysisonly

(means

andvariances)

22(73.3)

1Beyonddescriptiveanalysis(any

comparisons)

8(26.6)

210

.Outcomes

31.40

(0.24)

1.40

(0.24)

Satisfaction,

attitude,

perceptio

ns,opinion,

generalfacts.

7(23.3)

1Kno

wledg

e,skills(e.g.,OSCEs,SPsas

outcom

emeasure)

22(73.3)

1.5

Behaviours(e.g.,ph

ysicianactual

practice)

1(3.33)

2Patient/health

care

outcom

es0(0)

3Total

score

1810

.26(2.14)

a Allcomments

inparenthesesaremeant

forclarificatio

n.OSCE¼objectivestructured

clinical

exam

ination,

SP¼standardized

patient,SD¼standard

deviation.

bEachitems5,

6,and7canapplyto

differentinstruments

used

inthestudy.


or more participants,15,17,22–24,29,30,36–38,40 theremaining had the number of participants less than100. One study had the number of participants aslow as 16 students.16

(3) Theoretical basis of the studies: Most studies didnot explore the theoretical basis for their findings orcame with logical explanation to justify theirfindings.(4) Validity of methods used and results: Moststudies did not provide evidence for the validity ofmethods used. The authors tried a number ofstrategies to support their research design. Forexample, the use of randomized controlled stu-dies,13–18,22–24,26,31,35,37,42 controlled stu-dies,20,25,28,32,34,39,41 and crossover studies.19,29,33

However, the assessment of impact was based ontesting knowledge learnt by answering quiz ques-tions,14,16,26 theoretical examinations,20,24,26 andpractical examinations.25

A few studies did not justify their results or providedweak justification. For example, the study by Mura-kami et al.40 reported that the 3D model yieldedstrongly positive students' perspectives and signifi-cantly improved students’ skills in radiology in laterclinical years. However, this improvement in radiologyskills could be an associated change rather than acause-effect outcome.

3.5. Assessing study qualities—MERSQI scores

Table 3 summarizes the MERSQI scores for the 30studies by item and domain. The total mean ofMERSQI score was 10.26 (SD¼2.14, range 6.0–13.5). The study with the highest MERSQI score wasa randomized controlled study showing that three-dimensional presentations improve in the identificationof surgical liver anatomy. This study earned the highestscore because it was a randomized controlled study,had a higher number of participants (475), usedobjective measures, that were appropriate for studydesign and the type of data and the data analysis wasbeyond descriptive analysis. However, the outcomes ofthe study were at the level of knowledge and skillsobtained and did not explore higher levels of outcomesoutlined by MERSQ instrument.

The study with the lowest MERSQI score reported theuse of LINDSAY Virtual Human Project. The study didnot compare work done to a control group, was based onone school, and was not a randomized study. Theassessment did not use objective measures and was notvalidated. The data analysis was inappropriate for study


design and the outcomes were about satisfaction of thestudents that participated in the study.

4. Discussion

4.1. General discussion

A total of 30 studies were identified from searchingPubMed, EMBASE, the Web of Knowledge databases,eight journals on medical and dental education and sevenanatomy journals. These models were 3D web-basedmodels, 3D computer- and mobile-based models and 3Dnon-digital (physical) models. Interestingly these physicalmodels comprised a range of innovative ideas and werefrom different countries including USA,23,28 Canada,39 theNetherlands,42 Germany,29 and Korea.21 The use ofphysical model indicates that schools are not only movinginto 3D digital models but also there is a place for physicalmodel in teaching anatomy; although the major trend isdirected to digital models.

It is obvious from these studies that students favouredthe use of 3D anatomy models and found these modelsmore satisfactory when compared to traditional teaching,textbooks and lectures.19,20,24,26,28,30,31,33,39 However, notall studies demonstrated that 3D anatomy models aresuperior to 2D images or enhanced students' performancein anatomy. One study showed that 3D anatomy modelshas several disadvantages compared to traditional teach-ing,35 and a number of studies showed that there was nodifferences between 3D anatomy models and traditionalteaching or 2D images.19,20,24,26–28,30,33,39 The studies thatshowed an impact on learning, mainly demonstratedan improvement in performance on short-termbasis.14,18,21,22,34,37,38,42

Of these 30 studies, the first study that was indexed inPubMed was published in (year 2002).13 However, nearly80% of the studies found were published in the last 6 yearsindicating progressive interest in 3D anatomy models.Most studies were from North America and Europe. Onlyone study was from Australia and no studies were fromAfrica or South America. The majority of the studies werefrom Medicine and Dentistry. Very few studies were fromother allied health schools.

Because of the variability in the outcomes from thesestudies, it was decided to assess the quality of the studiesusing a standardized measure such as MERSQ instrument.MERSQ instrument has been widely used in the literatureto examine studies on training health care professionalsacross the education continuum on chronic disease care,43

coaching to enhance surgeons' operative performance,9 theimpact of physicians' occupational well-being on thequality of patient care,44 and learning outcomes of

participation in student-run clinics.45 The aims of suchassessment were (1) to assess the quality of the literatureon 3D anatomy models and the impact of using suchapproaches on learning, (2) to explore whether theavailable literature on this area has provided quality studiesthat can answer our questions in regard to the place of 3Dmodels in the teaching and learning, and (3) to enable theresearch community to clearly see the overall pictures andthe gaps in the literature that needs further studies andassessment.

The MERSQI scores obtained from this study arecomparable with the mean scores obtained from thesystematic reviews of simulated-based training for laparo-scopic surgery (mean¼11.9),46 undergraduate medicaleducation in substance abuse (mean¼10.42),47 and useof simulation in neurosurgical education (mean¼9.21).48

In the presence of several variables among these systematicreviews such as the topic researched, the nature of theresearch conducted, the journals in which these studieswere published and others, such comparisons on thequality of educational studies may reflect a commondeficiency in the literature particularly in regard to theassessment of validity to support the tools being used ineducational studies; a common deficiency also observed inthe studies included in this systematic review.

4.2. Implications for anatomy teaching

Although a number of factors affecting learning byusing 3D models have been identified from these studies,we are still in need of research that carefully assesses theimpact of these factors while using 3D anatomy models.These factors can be grouped into three categories:(i) factors related to the 3D model such as the design ofthe model, availability of visual and auditory informationsimultaneously, and orientations produced by the soft-ware,13,33,39 (ii) factors related to the learner characteris-tics,13,15–17,32,35,36,38,41 and (iii) factors related to thecurriculum and the learning environment,16,18,19,32,39 Thesefactors should be considered by designers of new 3Dmodels and course designers, as well as teaching staff.

Given the increasing interest in 3D anatomy modelsas evidenced from the increasing number of publishedresearch in this area, there is a need for multi-institutional studies that examine theories behind learn-ing by using 3D tools and impact of learning by 3Dmodels on the enhancement of knowledge, comprehen-sion, clinical skills, integration, and application. Cur-rently most studies focused on testing knowledge learntby answering quiz questions to evaluate their knowl-edge of 3D relationships,14,16,26 theoretical post-testexaminations,20,24,26 and practical examinations.25


While these methods may provide limited informationabout the usefulness of 3D anatomy models, there is aneed for in-depth research in this new area that canprovide answers to questions about the purpose ofusing 3D anatomy models in the curriculum, and theplace of 3D anatomy teaching in the undergraduatecurriculum and how we can assess the impact of using3D models on student’s learning. Other questions thatneed answers, will 3D anatomy models prepare stu-dents in a better way to clinical examination andunderstanding of clinical subjects such as surgery andmedicine? What are the long-term impacts of learningby using 3D anatomy model? Can 3D models help inlearning surface anatomy not just gross anatomy andrelated subjects?49

This systematic review is not without limitations. Inorder to ensure that most papers on the topic have beenconsidered, it was decided to design a search strategycovering three major databases: PubMed, EMBASE,and the Web of Knowledge. Seven keywords wereused in searching these databases, as well as thewebsites of eight medical and dental education journalsas well as seven anatomy journals. Also, the lists ofreferences in related research papers identified weresearched for any paper related to this review. Thisrigorous approach of study selection and explicitassessment of relevance of papers as per the inclusionand exclusion criteria resulted in the inclusion of 30studies in this systematic review. However, despitethese precautions, the restriction of study retrieval frommedical, and dental education and anatomy journalsmay not be optimal as no search of other allied healthcare journals was conducted. Because journals fromallied health disciplines such as nursing, physiotherapy,and occupational therapy were not searched, there maybe studies from these disciplines that were notincluded. However, a few papers from these disciplineswere found from searching the three databases and it isunlikely that any more data will make significantchanges to the outcomes of this review.

This review focused only on papers in the Englishlanguage. It is possible that there are papers in theliterature addressing the inclusion criteria and the aimsof the study and were not included because they werein languages other than English.

Finally, extraction and coding of data, as it is thecase with other reviews, can be subject to opinion ofthe observers. To minimize this possible confoundingfactor, it was decided to pilot the study and to use asystematic approach, secure consensus between theevaluators at different stages of data charting, collec-tion and critical evaluation.46

5. Conclusions and future research directions

There is evidence of progressive interest in the use of3D anatomy models over the last 6 years as evidencedfrom the number of publications. These studies showedthat 3D anatomy models in digital and non-digital(physical) format are favored by students in medical,dental and other allied health schools and can be usedto support the curriculum and enhance students’ skillsin spatial visualization of anatomical relationships.

First, factors affecting learning by using 3D models:although a few factors have been identified from thesestudies, there are a number of factors that need to bestudied. For example, student’s learning needs, stu-dent’s learning style, educational design of 3D model,digital versus physical models, and effect of trainingprior to using 3D models. More important, is tounderstand the interactions between learners and 3Dtechnologies in order to identify potential advantagesand limitations and ideal methods to be used inassessing the impact of 3D models on learning.

Second, exploring the medium-term and long-termimpacts of learning by using 3D anatomy models. Forexample, is learning by using 3D models preparestudents and trainees to surgical procedures. Apartfrom the enhancement of student’s skills in anatomyperformance, what are other skills developed bylearners when they use 3D anatomy models in theirlearning?

Third, There is a need for new studies of highresearch quality by considering the limitations identi-fied in this review in their design.

With these recommendations for research in mind,this review presents a framework with which research-ers interested in 3D anatomy models will be able todevelop a pedagogical technology to enhance student’slearning skills and undertake comparative studies ofresearch relating to 3D anatomy models and theirimpact in undergraduate medical, dental and alliedhealthcare curricula.

Conflict of interest

The author declare that they have no competinginterest.

Author’s contribution

SAA, started the design of the study and itsmethodology, SAA and SA searched the databases,collected the data, analysed the findings, and createdthe two lists, SAA and SA interpreted the findings,


ranked the articles, creation of tables, created thefigures and drafted the manuscript. SAA and SAcontributed to the revision of the manuscript andapproved the final manuscript for submission.

Funding/support

This work was funded by the College of MedicineResearch Center, Deanship of Scientific Research,King Saud University, Riyadh, Saudi Arabia.

Acknowledgements

The authors would like to thank Diana Azer for herassistance and reviewing the manuscript. Also thanksDr. Lily Scott for her assistance in this work and Ms.Mae Eustaquio for her secretarial help.

References

1. Azer SA, Eizenberg N. Do we need dissection in an integratedproblem-based learning medical course? Perceptions of first- andsecond-year students. Surg Radiol Anat 2007;29(2):173–180.

2. Berney S, Bétrancourt M, Molinari G, Hoyek N. How spatialabilities and dynamic visualizations interplay when learningfunctional anatomy with 3D anatomical models. Anat Sci Educ2015;8(5):452–462.

3. Langlois J, Wells GA, Lecourtois M, Bergeron G, Yetisir E,Martin M. Spatial abilities of medical graduates and choice ofresidency programs. Anat Sci Educ 2015;8(2):111–119.

4. Vandenberg SG, Kuse AR. Mental rotations, a group test ofthree-dimensional spatial visualization. Percept Mot Skills1978;47(2):599–604.

5. Ullman S. Three-dimensional object recognition based on thecombination of views. Cognition 1998;67(1–2):21–44.

6. Yiannakopoulou E, Nikiteas N, Perrea D, Tsigris C. Virtualreality simulators and training in laparoscopic surgery. Int J Surg2015;13:60–64.

7. Arksey H, O’Malley L. Scoping studies: towards a methodolo-gical framework. Int J Soc Res Methodol Theory Pract 2005;8:19–32.

8. Reed DA, Cook DA, Beckman TJ, Levine RB, Kern DE, WrightSM. Association between funding and quality of publishedmedical education research. J Am Med Assoc 2007;298(9):1002–1009.

9. Min H, Morales DR, Orgill D, Smink DS, Yule S. Systematicreview of coaching to enhance surgeons' operative performance.Surgery 2015;158(5):1168–1191.

10. Reed DA, Beckman TJ, Wright SM, Levine RB, Kern DE, CookDA. Predictive validity evidence for medical education researchstudy quality instrument scores: quality of submissions to JGIM'sMedical Education Special Issue. J Gen Intern Med 2008;23(7):903–907.

11. Cook DA, Reed DA. Appraising the quality of medical educationresearch methods: the medical education research study qualityinstrument and the Newcastle–Ottawa scale-education. Acad Med2015;90(8):1067–1076.

12. McCrae N, Purssell E. Eligibility criteria in systematic reviewspublished in prominent medical journals: a methodologicalreview. J Eval Clin Pract 2015;21(6):1052–1058.

13. Garg AX, Norman GR, Eva KW, Spero L, Sharan S. Is there anyreal virtue of virtual reality?: the minor role of multipleorientations in learning anatomy from computers. Acad Med2002;77(10 Suppl):S97–S99.

14. Nicholson DT, Chalk C, Funnell WR, Daniel SJ. Can virtualreality improve anatomy education? A randomised controlledstudy of a computer-generated three-dimensional anatomical earmodel. Med Educ 2006;40:1081–1087.

15. Guillot A, Champely S, Batier C, Thiriet P, Collet C. Relation-ship between spatial abilities, mental rotation and functionalanatomy learning. Adv Health Sci Educ Theory Pract 2007;12(4):491–507.

16. Hisley KC, Anderson LD, Smith SE, Kavic SM, Tracy JK.Coupled physical and digital cadaver dissection followed by avisual test protocol provides insights into the nature of anatomi-cal knowledge and its evaluation. Anat Sci Educ 2008;1(1):27–40.

17. Levinson AJ, Weaver B, Garside S, McGinn H, Norman GR.Virtual reality and brain anatomy: a randomised trial of e-learning instructional designs. Med Educ 2007;41:495–501.

18. Marsh KR, Giffin BF, Lowrie Jr. DJ. Medical student retentionof embryonic development: impact of the dimensions added bymultimedia tutorials. Anat Sci Educ 2008;1(6):252–257.

19. Donnelly L, Patten D, White P, Finn G. Virtual human dissectoras a learning tool for studying cross-sectional anatomy. MedTeach 2009;31(6):553–555.

20. Hu J, Yu H, Shao J, Li Z, Wang J, Wang Y. Effects of dental 3Dmultimedia system on the performance of junior dental studentsin preclinical practice: a report from China. Adv Health Sci EducTheory Pract 2009;14(1):123–133.

21. Oh CS, Kim JY, Choe YH. Learning of cross-sectional anatomyusing clay models. Anat Sci Educ 2009;2(4):156–159.

22. Abid B, Hentati N, Chevallier JM, Ghorbel A, Delmas V,Douard R. Traditional versus three-dimensional teaching ofperitoneal embryogenesis: a comparative prospective study. SurgRadiol Anat 2010;32(7):647–652.

23. Estevez ME, Lindgren KA, Bergethon PR. A novel three-dimensional tool for teaching human neuroanatomy. Anat SciEduc 2010;3(6):309–317.

24. Hu A, Wilson T, Ladak H, Haase P, Doyle P, Fung K.Evaluation of a three-dimensional educational computer modelof the larynx: voicing a new direction. J Otolaryngol Head NeckSurg 2010;39(3):315–322.

25. Codd AM, Choudhury B. Virtual reality anatomy: is it compar-able with traditional methods in the teaching of human forearmmusculoskeletal anatomy?. Anat Sci Educ 2011;4(3):119–125.

26. Keedy AW, Durack JC, Sandhu P, Chen EM, O'Sullivan PS,Breiman RS. Comparison of traditional methods with 3Dcomputer models in the instruction of hepatobiliary anatomy.Anat Sci Educ 2011;4(2):84–91.

27. Vuchkova J, Maybury TS, Farah CS. Testing the educationalpotential of 3D visualization software in oral radiographicinterpretation. J Dent Educ 2011;75(11):1417–1425.

28. Bareither ML, Arbel V, Growe M, Muszczynski E, Rudd A,Marone JR. Clay modeling versus written modules as effectiveinterventions in understanding human anatomy. Anat Sci Educ2013;6(3):170–176.


29. Knobe M, Carow JB, Ruesseler, M, et al. Arthroscopy orultrasound in undergraduate anatomy education: a randomizedcross-over controlled trial. BMC Med Educ 2012;12:85.

30. Maggio MP, Hariton-Gross K, Gluch J. The use of independent,interactive media for education in dental morphology. J DentEduc 2012;76(11):1497–1511.

31. Metzler R, Stein D, Tetzlaff, R, et al. Teaching on three-dimensional presentation does not improve the understanding ofaccording CT images: a randomized controlled study. TeachLearn Med 2012;24(2):140–148.

32. Nguyen N, Nelson AJ, Wilson TD. Computer visualizations:factors that influence spatial anatomy comprehension. Anat SciEduc. 2012;5(2):98–108.

33. Roach VA, Brandt MG, Moore CC, Wilson TD. Is three-dimensional videography the cutting edge of surgical skillacquisition?. Anat Sci Educ 2012;5(3):138–145.

34. Ruisoto P, Juanes JA, Contador I, Mayoral P, Prats-Galino A.Experimental evidence for improved neuroimaging interpretationusing three-dimensional graphic models. Anat Sci Educ 2012;5(3):132–137.

35. Khot Z, Quinlan K, Norman GR, Wainman B. The relativeeffectiveness of computer-based and traditional resources foreducation in anatomy. Anat Sci Educ 2013;6(4):211–215.

36. Tworek JK, Jamniczky HA, Jacob C, Hallgrímsson B, Wright B.The LINDSAY Virtual Human Project: an immersive approachto anatomy and physiology. Anat Sci Educ 2013;6(1):19–28.

37. Müller-Stich BP, Löb N, Wald, D, et al. Regular three-dimensional presentations improve in the identification ofsurgical liver anatomy – a randomized study. BMC Med Educ2013;13:131.

38. Hoyek N, Collet C, Di Rienzo F, De Almeida M, Guillot A.Effectiveness of three-dimensional digital animation in teachinghuman anatomy in an authentic classroom context. Anat Sci Educ2014;7(6):430–437.

39. Lisk K, McKee P, Baskwill A, Agur AM. Student perceptionsand effectiveness of an innovative learning tool: anatomy glovelearning system. Anat Sci Educ 2015;8(2):140–148.

40. Murakami T, Tajika Y, Ueno, H, et al. An integrated teachingmethod of gross anatomy and computed tomography radiology.Anat Sci Educ 2014;7(6):438–449.

41. Nguyen N, Mulla A, Nelson AJ, Wilson TD. Visuospatialanatomy comprehension: the role of spatial visualization abilityand problem-solving strategies. Anat Sci Educ 2014;7(4):280–288.

42. ten Brinke B, Klitsie PJ, Timman R, Busschbach JJ, Lange JF,Kleinrensink GJ. Anatomy education and classroom versuslaparoscopic dissection-based training: a randomized study atone medical school. Acad Med 2014;89(5):806–810.

43. Bogetz JF, Rassbach CE, Bereknyei S, Mendoza FS, SandersLM, Braddock 3rd CH. Training health care professionals for21st-century practice: a systematic review of educational inter-ventions on chronic care. Acad Med 2015;90(11):1561–1572.

44. Scheepers RA, Boerebach BC, Arah OA, Heineman MJ,Lombarts KM. A systematic review of the impact of physicians'occupational well-being on the quality of patient care. Int JBehav Med 2015;22(6):683–698.

45. Schutte T, Tichelaar J, Dekker RS, van Agtmael MA, de VriesTP, Richir MC. Learning in student-run clinics: a systematicreview. Med Educ 2015;49(3):249–263.

46. Zendejas B, Brydges R, Hamstra SJ, Cook DA. State of theevidence on simulation-based training for laparoscopic surgery: asystematic review. Ann Surg 2013;257(4):586–593.

47. Kothari D, Gourevitch MN, Lee, JD, et al. Undergraduatemedical education in substance abuse: a review of the quality ofthe literature. Acad Med 2011;86(1):98–112.

48. Kirkman MA, Ahmed M, Albert AF, Wilson MH, Nandi D,Sevdalis N. The use of simulation in neurosurgical education andtraining. A systematic review. J Neurosurg 2014;121(2):228–246.

49. Azer SA. The place of surface anatomy in the medical literatureand undergraduate anatomy textbooks. Anat Sci Educ 2013;6(6):415–432.

3d anatomy models and impact on learning a review of the

Documents