An Ultra-high Deﬁnition and Interactive Simulator for Human

Dissection in Anatomic Learning

Andrei Rafael Brongel

, William John Pereira Brobouski

, Lucas Murbach Pierin

, Carlos Gomes

Manoel de Campos Almeida

and Edson Jos

e Rodrigues Justino

3,4

Medical School, Pontif

ıcia Universidade Cat

olica do Paran

a - PUCPR, Curitiba, Paran

a, Brazil

Polytechnic School, Pontif

ıcia Universidade Cat

olica do Paran

a - PUCPR, Curitiba, Paran

a, Brazil

Centro de Inovac¸

ao em Imagens M

edicas, CIIM - PUCPR, Curitiba, Paran

a, Brazil

Programa de P

os-Graduac¸

ao em Inform

atica, PPGIa - PUCPR, Curitiba, Paran

a, Brazil

Keywords:

Human Anatomy Learning, Dissection, Simulation, Ultra-high Deﬁnition, Real Bodies Models.

Abstract:

This paper presents an ultra-high deﬁnition simulator for the teaching of human anatomy in class. The simu-

lator in question consists of hardware and software specially developed for this purpose. The hardware seeks

to meet the requirements of real-scale representation of models, visual acuity, color, texture, depth perception

e touch-based interactivity. The software, in turn, offers a set of dissecting tools typically used in anatomical

studies, in addition to allowing connectivity to educational environments and the Internet. The characteristic

that stands out most in this simulator is the fact that it is not an anatomical atlas, but a dissecting table that uses

models from real bodies, which differentiates it from most of the simulators and anatomical atlases developed

to this end.

1 INTRODUCTION

In almost all courses related to life sciences, espe-

cially those related to health such as medicine, nurs-

ing, physiotherapy, dentistry, and others, required

the study of human anatomy. The consensus among

anatomy teachers is that there is currently no replica,

book, or 3D computational tool to replace the direct

manipulation of a real human body. Some features,

such as color, depth, texture, anatomical position, and

detail, are essential for comprehensive learning. It

is also unanimous among anatomy teachers that the

corpse is as close as we can have to a living human

being, even though in dead tissues we cannot observe

some properties. As an example, it is possible to

cite the cerebral activities, vascularisation, heartbeats,

among others. According to the literature (Ghosh,

2015) (Shaikh, 2015) (Singh A.K. and D.M., 2011),

the manipulation of corpses opens the discussion of

three main aspects. The ﬁrst one concerns ethical is-

sues since many of unclaimed bodies come from do-

nations and are made by forensic laboratories or sim-

ple purchase when the law of the country allows. The

second one concerns health issues. Toxic chemical,

such as formaldehyde, is the main component used

in the preservation of bodies and parts. It makes the

anatomy laboratories unhealthy environments and in-

appropriate for teaching and learning. The third one is

the quality of learning. The permanent manipulations

of the bodies makes them not suitable for teaching and

learning in a short time.

Over the point of view of clinical and surgical

procedures, we can observe the increased use of the

non-invasive procedure, whose supporting instrument

is the images made from cameras coupled to surgical

instruments or from CT (Computerised Tomography)

and MR (Magnetic Resonance) exams. Procedures

such as laparoscopy, endoscopy, and surgeries guided

by neuronavigation have become the most viable so-

lutions for the reduction of the risks of postoperative

infection and the rapid recovery of the patient.

Given the current scenario, researchers all over

the world have been efforts in ﬁnding solutions to

minimize the use of cadavers in the teaching of

anatomy and surgical practices. Among another kind

of simulation, the digital models have been showing

promise. There are some excellent examples, such

as the Anatomage table

a high-quality artistic at-

las. However, Ackerman (Ackerman, 1998) and Azer

(Azer and Azer, 2016) asserts there are some reasons

Anatomage - https://www.anatomage.com/table6/

284

Brongel, A., Brobouski, W., Pierin, L., Gomes, C., Almeida, M. and Justino, E.

An Ultra-high Deﬁnition and Interactive Simulator for Human Dissection in Anatomic Learning.

DOI: 10.5220/0007707102840291

In Proceedings of the 11th International Conference on Computer Supported Education (CSEDU 2019), pages 284-291

ISBN: 978-989-758-367-4

to restrict the use of this kind of atlas. Among other

things, the diversity of structures found in nature that

is a limiting factor, according to them, because it is

difﬁcult to replicate in computational models still in

the present day. Moreover, it is almost complicated

for a graphic artist to map all the nuances of the hu-

man body.

Based on the last discussion, this paper presents

an ultra-high deﬁnition simulator for the teaching of

human anatomy in class. The simulator consists of

hardware and software specially developed for this

purpose. The device seeks to meet the requirements

of high-quality representation of models, visual acu-

ity, color, texture, depth perception e touch-based in-

teractivity. The application (App), in turn, offers a

set of dissecting tools typically used in anatomical

studies, in addition to allowing connectivity to educa-

tional environments and the Internet. The character-

istic that stands out most in this simulator is the fact

that it is not an anatomical atlas, but a dissecting table

that may perform almost all operations that can be ex-

ecuted in a real necropsy table, which differentiates it

from most of the digital anatomical atlases developed

to this end. To simulate a necropsy table, we adopted

the same format for the interactive table. However,

the hardware supports two additional projectors with

the same hardware features.

Five sections make up this paper. The ﬁrst section

is this introduction. The second section presents the

material used to make the simulator. The third section

contains the method used for building the software of

the simulator. The fourth section presents the results.

Finally, the last section presents a conclusion and fu-

ture works.

2 MATERIAL

We divided the materials used on the simulator into

two groups. The ﬁrst one belongs to the software and

is made up of the database. The database is the most

relevant material for the characterization of the simu-

lator since it must have its origin in real human bod-

ies and not coming from artistic models. The second

group belongs to the hardware. The hardware must

meet the usability requirements in anatomy classes,

be dynamic and have a user-friendly interface.

2.1 Data Base

In the last four decades, several databases of real hu-

man bodies have emerged, whose objective was to

create a digital database of human bodies, both mas-

culine and feminine, for educational and research pur-

poses. The ﬁrst database started in the 1990s by

Michael J. Ackerman of the US National Library of

Medicine became known as The Visible Human Body

Project (VHP)(Ackerman, 2002) (Spitzer and Whit-

lock, 1998). Other databases came later, like Korean

(Park et al., 2005) and Chinese (Chen et al., 2018) ,

all using the same cryosection process. However, only

the VHP database became public and available for the

development of new applications under an agreement

with the US National Library of Medicine. The other

databases are private.

The VHP database consists of a 1mm thick sliced

male body, photographed on 70mm colored ﬁlms and

then scanned in 2k (2048 x 1216 pixels) and 4k (

4096 x 2700 pixels) resolution, totaling 1871 sec-

tions. In addition to the anatomical images, the X-

ray, Computed Tomography and Magnetic Resonance

complete the set. The female body has the same char-

acteristics as the male corpse, with one exception.

The intervals used in the anatomical images were 0.33

mm instead of 1.0 mm. The female set results in more

than 5,000 anatomical images. In Figure 1, we can see

a sample of the male and female database.

(a) (b)

Figure 1: Database samples: (a) Male criosection, (b) Fe-

male criosection.

2.2 Interactive Table

The use of interactive screen has been growing sig-

niﬁcantly in recent years with the advent of the mul-

titouch screens. This kind of screen allows complex

interactions with the user. Two crucial factors were

taken into account to determine the sizing of the ta-

ble and therefore the dimensioning of the screen. The

ﬁrst one was the screen size. The screen should con-

tain the body of a human with approximately 1.70m.

This feature would ensure the representation of the

organs of the human body in the standard scale (1:1).

The second is the screen resolution. The screen in

question should also support resolutions up to 4k. The

high resolution would guarantee the visual acuity nec-

essary for anatomy studies. Another relevant factor is

the application server, which should support the ma-

nipulation of a high volume of data from the image

databases. The interactive table was designed and

built following the usability parameters in the labo-

ratory for a student group and the teacher. Figure 2

An Ultra-high Deﬁnition and Interactive Simulator for Human Dissection in Anatomic Learning

285

shows the large screen interactive table.

Figure 2: The large screen interactive table.

3 METHOD

The method used to create the simulator took into ac-

count two constructive aspects. The ﬁrst one dealt

with the adequacy of 2k images databases of the fe-

male and male corpses, to be used by the simulator.

The choice of resolution in 2k was due to the mainte-

nance of the body’s near from the natural scale, one

of the ﬁrst prerequisites of the project. The second

took care of the interactivity and the user interface, to

make the interface user-friendly.

The choice of VHP occurred because it is a

database of public domain and the existence of both

sexes, the masculine and the feminine. Another im-

portant aspect is in the fact that they are images of

real bodies in ultra-high deﬁnition 2k and 4k. Since

its making in the 1990s, numerous works have been

developed using the VHP database images (Acker-

man, 2002) (Spitzer and Whitlock, 1998), many of

them resulted and simulators for different purposes

such as endoscopy (Pﬂesser et al., 2001). However,

due to the byte size of the 2k and 4k databases, the

generation of synthetic or rendered models was the

most adopted solution. Even with current technology,

it is necessary to use powerful game support compu-

tational resources to manipulate such bases in their

original form, and this has become the great challenge

of this project (the last generation of microprocessor,

high memory capacity, solid state disk, and others).

Another aspect to be highlighted is the resolution of

the images. Rendered models adapt to the display ca-

pabilities, making it easy to create and manipulate 3D

models. Only a few years ago the technology of GPUs

or Video Graphics Cards started to support the reso-

lution in 4k and have memory capabilities and speed

to allow the processing of the VHP base in its orig-

inal form. Based on that features we build the in-

teractive table hardware. Therefore, for this project,

the VHP database was chosen in 2k resolution. The

chosen resolution allows for the maintenance of the

original scale of the slices of cryosection images and

the FULLHD (1920 x 1080 pixels) projection of the

volumetric visualization model. Figure 3 shows an

example.

Figure 3: An Exemple of the 2k (slice) and FULLHD pro-

jection (volumetric projection).

3.1 Database Pre-processing

In the pre-processing of the database, we applied a

classiﬁcation algorithm based on the SVM (Support

Vector Machine) (Wang et al., 2012) (Tsai et al.,

2006) for the removal of the blue background (ice)

that covered the external part of the bodies and some

internal areas. In Figure 1 it is possible to see such

areas. We apply four different classiﬁers for the com-

plete removal the ice over the skin, one for each four

different body regions. We do not use any post-

processing.

3.2 Screen Interface

Figure 4 shows the different simulator manipulation

capabilities. We divided the screen area into two sub-

areas. The ﬁrst contains the anatomical projection

planes chosen for the visualization (coronal in blue,

axial in green and sagittal in red), following the di-

dactic perception in the anatomy study. Each plane

allows navigation between the slices of the body in

a thickness of 0.33 mm (for the female body) and 1

mm (for the male body) (Figure 5, Figure 6, and Fig-

ure 7). This navigation also allows adjusting the other

two planes in a cut position, in order to enable the vi-

sualization of the three cuts in the volumetric model,

in perspective (Figure 4). The second subarea is able

the projection of the volumetric model in perspective

(Figure 4). The model can undergo several geomet-

ric transformations, such as the rotation in the axis

of the current anatomical plane of cut, inversion from

CSEDU 2019 - 11th International Conference on Computer Supported Education

286

the point of view of the body (volumetric model in

perspective, Figure 8), adjustment of the transparency

(Figure 9), stereoscopic visualization for depth per-

ception (Figure 10), among other resources .

3.3 Stereoscopy

The ability to perceive and interact with the structure

of space is one of the fundamental goals of the visual

system. Two retinal images are different because the

retinas are in slightly different places. Stereopsis is

the ability to use binocular disparity as a cue to depth.

The stereoscopy is the technic used to create the stere-

opsis by devices. There are two kinds of stereoscopy

applicable to digital devices. The ﬁrst and oldest tech-

nic is the anaglyphic. In this kind, two images create

the stereo pair, one in red and the other in cyan. To

view the effect anaglyphic glasses must be used. The

second is the polarized stereoscopy. Its use is more re-

cent, having a wide range of applications, such as 3D

TV’s, 3D cinema and 3D projectors. There are two

kinds of technics of polarization, the linear and the

circular. In both cases, we must use polarized glasses

(Daly et al., 2011).

The use of stereoscopy in medical applications has

been growing in recent years (Livatino et al., 2015)

(van Beurden et al., 2009). In the simulator, we used

anaglyphic stereoscopy initially, in order to evaluate

its performance at the table. Although the results are

promising, as seen in the projection of the body of

Figure 10, its use will be more effective in the vi-

sualization of the inner organs, whose segmentation

process is in progress.

3.4 Interface Objects

The simulator has a graphical interface managed by

an Interface Object called Draggable DIAL (virtual

mouse that concentrates the manipulation tools of the

volumetric model of the bodies such as rotation Fig-

ure 11(f), perspective Figure 11(e), and transparency

Figure 11(a). It also has rotating menu functional-

ity for selecting anatomical pieces available in a li-

brary, and also an option for selecting a type of tis-

sue to be viewed and manipulated from the selected

anatomical piece, Figure 12. It has too a push button

(on/off) for enabling stereoscopic visualization fea-

tures (with anaglyph glasses), Figure 11(b). A push

button (on/off) for visualization of the 2D cut in the

bodies volumetric model, Figure 11(e). A push but-

ton (lock/unlock) the touchscreen to avoid unwanted

touch, Figure 11(c). A push button (on/off) for en-

abling the slices viewer in the volumetric model, Fig-

ure 11(d). A button for activating active blackboard

functions (pens, markers, rubber, save screen, and

colors), Figure 13(c) and Figure 14. A button for acti-

vating the browser for Internet search, 13(b). Finally,

a button for activating accessibility to the Online Ed-

ucation Platforms (OEP), 13(a). In addition to the

DIAL, the simulator has a second control area for the

manipulation of the cuts (1mm slices of the male body

or 0.33mm of the female body). This area has slid-

ing bars of the three anatomical planes (axial, coronal

and sagittal), for each cut of the plane in highlighted.

The displacement of those bars in the 2D visualiza-

tion allows the same cut in the volumetric model. It

should note that such planes orthogonal to the body

can adjust to any angle by merely shifting the bar of

the plane to the desired angular position.

3.5 Reproductive System

As already mentioned in previous sessions, the simu-

lator has two bodies, one male, and one female. The

sex selection is in an initial screen of the simulator,

or later, through the rotating menu of the DIAL. The

different sexes allow the anatomy teachers to explore,

along with the students, the anatomical differences of

both sexes, as well as to present in detail the reproduc-

tive apparatus. The use of real body models provides

biological data that could not be observed in synthetic

models, such as the inﬂuence of hormones, testos-

terone, and progesterone, on the formation, growth,

and aging of organs. In Figures 15 and 16 we can see

both male and female sex respectively.

3.6 Imaging Exams

As seen in the sessions where we described the VHP

database, it has two kinds of set of the imaging exams.

The ﬁrst set is the computed tomography (CT) images

and the second set is the magnetic resonance (MR)

imaging . The male database consists of axial MR

images of the head and neck taken at 4mm intervals

and longitudinal sections of the rest of the body also

at 4mm intervals. The resolution of the MR images

is 256 pixels by 256 pixels, in grayscale. The MR

imaging database consists of three acquisition modes,

T1, T2, and Proton Density Image (PD). CT data are

taken in the axial plane of computed tomography of

the whole body taken at intervals of 1mm at a reso-

lution of 512 pixels by 512 pixels in grayscale. The

female corpse data set has the same characteristics as

the male corpse, with one exception. Axial anatomi-

cal images were obtained at intervals of 0.33 mm in-

stead of 1.0 mm intervals. As seen, CT and MR im-

ages can provide important educational resources for

studying and practice in imaging exams analysis. The

An Ultra-high Deﬁnition and Interactive Simulator for Human Dissection in Anatomic Learning

287

Figure 4: The Interactive Simulator Interface.

Figure 5: Adjustable coronal plane view mode.

Figure 6: Adjustable axial plane view mode.

simulator offers teachers and students the possibility

of associating each CT and MR image with the cor-

related anatomical cut of the body and the volumetric

model. In this way, the student can observe the ex-

ams, such as in real practice, and relating it to the

Figure 7: Adjustable sagittal plane view mode.

Figure 8: Adjustable perspective view mode.

body. Figures 17, 18 and 19 show an example of a

slice, the CT and MR (T1, T2, and PD) images.

CSEDU 2019 - 11th International Conference on Computer Supported Education

288

Figure 9: Adjustable transparency view mode.

Figure 10: The full body in stereoscopic view (Anaglyph

Stereoscopy).

Figure 11: The draggable dial in volumetric model tools.

Figure 12: The draggable dial in circular menu library.

Figure 13: The draggable dial in eduction tools.

Figure 14: An exemple of the Active Blackboard.

Figure 15: Male.

4 CONCLUSIONS

This paper presented an ultra-high deﬁnition simula-

tor, now called The Visible Human Table (VHT)

, for

the dissection and teaching of the human anatomy in

the classroom. The differential to other anatomical

atlases and simulators are in the use of real body im-

ages in ultra-high resolution and non-synthetic body

The VHT Project: https://www.youtube.com/watch?

v=yZ-rDkC4UXY

An Ultra-high Deﬁnition and Interactive Simulator for Human Dissection in Anatomic Learning

289

Figure 16: Female.

Figure 17: Male: anatomic image.

models which gives it visual acuity, scale compliance

with real human organs, texture, color, and depth per-

ception. Another positive aspect is the diversity of

sex. From teaching and learning, the simulator com-

plies with the precepts used in the study of anatomy,

that is, respecting the anatomical planes of cutting,

volumetric visualization, and dynamics in the dissec-

tion process. Another essential element that charac-

terizes it as a learning tool is the availability of the

imaging exams associated with the anatomical im-

ages in cuts, the connectivity with resources of the

Online education platform and the active blackboard.

As future work, we are planing the inclusion of new

ultra-high-deﬁnition segmentations of tissues of spe-

ciﬁc organs, such as the skeleton, muscles, circulatory

system, among others. We intend to include the X-ray

exams set, from de VHP databases, in to the simula-

(a) (b)

Figure 18: Database samples: (a) TC image, (b) MR image

T1.

(a) (b)

Figure 19: Database samples: (a) MR image T2, (b) MR

image PD.

tion and support for reading DICOM

standard.

ACKNOWLEDGEMENTS

We want to thank the US National Li-

brary of Medicine for providing us the

databases of The Visible Human Body Project

(www.nlm.nih.gov/reseach/visible/), without which

this project would not become possible. The Voxel-

Man for providing the database of segmented inner

organs (Voxel-Man, www.voxel-man.com), which

is assisting us in the process of ultra-high deﬁnition

segmentation of the two bodies. To the Financiadora

de Estudos e Projetos - FINEP for the ﬁnancial

support for the development of this project. Also,

to the Center of Innovation in Medical Imaging of

the PUCPR (CIIM) for the technical and laboratory

support.

The Digital Imaging and Communications in Medicine

(DICOM) is a standard for handling, storing, printing, and

transmitting information in medical imaging.

CSEDU 2019 - 11th International Conference on Computer Supported Education

290

REFERENCES

Ackerman, M. J. (1998). The visible human project. Pro-

ceedings of the IEEE, 86(3):504–511.

Ackerman, M. J. (2002). Visible human project

: From

data to knowledge. Yearb Med Inform, 11(01):115–

117.

Azer, S. A. and Azer, S. (2016). 3d anatomy models and

impact on learning: A review of the quality of the lit-

erature. Health Professions Education, 2(2):80 – 98.

Chen, D., Zhang, Q., Deng, J., Cai, Y., Huang, J., Li,

F., and Xiong, K. (2018). A shortage of cadav-

ers: The predicament of regional anatomy education

in mainland china. Anatomical Sciences Education,

11(4):397–402.

Daly, S. J., Held, R. T., and Hoffman, D. M. (2011). Per-

ceptual issues in stereoscopic signal processing. IEEE

Transactions on Broadcasting, 57:347–361.

Ghosh, S. K. (2015). Human cadaveric dissection: a histor-

ical account from ancient greece to the modern era. In

Anatomy & cell biology.

Livatino, S., Paolis, L. T. D., D’Agostino, M., Zocco, A.,

Agrimi, A., Santis, A. D., Bruno, L. V., and Lapresa,

M. (2015). Stereoscopic visualization and 3-d tech-

nologies in medical endoscopic teleoperation. IEEE

Transactions on Industrial Electronics, 62(1):525–

535.

Park, J. S., Chung, M. S., Hwang, S. B., Lee, Y. S., Har,

D.-H., and Park, H. S. (2005). Visible korean human:

improved serially sectioned images of the entire body.

IEEE Trans Med Imaging, 24(3):352–360.

Pﬂesser, B., Petersik, A., Pommert, A., Riemer, M., Schu-

bert, R., Tiede, U., Hohne, K. H., Schumacher, U.,

and Richter, E. (2001). Exploring the visible human’s

inner organs with the voxel-man 3d navigator. Stud

Health Technol Inform, 81:379–385.

Shaikh, S. T. (2015). Cadaver dissection in anatomy: The

ethical aspect. Anatomy and Physiology: Current Re-

search, 5(1):1–2.

Singh A.K., Sharma R.C., S. R. and D.M., M. (2011). Chal-

lenges in cadaver availability for learning and research

in medical sciences. International Journal of Medical

& Clinical Research, 2(2):67.

Spitzer, V. M. and Whitlock, D. G. (1998). The visible hu-

man dataset: The anatomical platform for human sim-

ulation. The Anatomical Record, 253(2):49–57.

Tsai, C.-F., McGarry, K., and Tait, J. (2006). Claire: A mod-

ular support vector image indexing and classiﬁcation

system. ACM Trans. Inf. Syst., 24(3):353–379.

van Beurden, M., Van Hoey, G., Hatzakis, H., and Ijssel-

steijn, W. (2009). Stereoscopic displays in medical

domains: A review of perception and performance ef-

fects. volume 7240, page 72400.

Wang, X., Wang, S., Zhu, Y., and Meng, X. (2012). Im-

age segmentation based on support vector machine.

In Proceedings of 2012 2nd International Conference

on Computer Science and Network Technology, pages

202–206.

An Ultra-high Deﬁnition and Interactive Simulator for Human Dissection in Anatomic Learning

291