Assessing the Feasibility of using Augmented Reality to Visualize

Interventional Radiology Imagery

Christopher Bartlett

, Noelle LeRoy

, Damian Schofield

, Jonathan Ford

and Summer Decker

Department of Computer Science, State University of New York, Oswego, New York, U.S.A.

Department of Radiology, Morsani College of Medicine, University of South Florida, Tampa, Florida, U.S.A.

Keywords: Medical Visualization, Augmented Reality, Interventional Radiology, Surgery, Human Computer Interaction.

Abstract: Image-guided procedures, such as those in radiology, are frequently reliant on data which is visualized on

traditional monitors. In an operating theatre, these monitors are often placed at poor ergonomic positions,

causing physicians to rotate their heads to the side while their hands are working before them. This study

seeks to investigate whether visualizing data on an augmented reality headset that projects an image in front

of the participant will reduce task-time and increase efficiency. The primary purpose behind this study is to

alleviate neck and back pain in physicians performing data/image guided procedures. A number of augmented

reality headsets were tested in a clinical setting and a number of experiments were undertaken to test the

viability of this technology in an operating theatre. The experiment consisted of comparing the use of an

augmented reality headset against a computer monitor while performing tasks that required similar hand eye

co-ordination to that needed during a surgery. The research hypothesized that the use of an augmented reality

headset would increase accuracy and efficiency; while decreasing eye fatigue and neck/back pain.

1 INTRODUCTION

Interventional Radiology (IR) is characterized by

minimally invasive procedures which are guided by

images such as Computerized Tomography (CT),

fluoroscopy and ultrasound. These visualization

modalities allow interventional radiologists to guide

small instruments through the body in hopes of

avoiding invasive open surgeries.

Vascular Interventional Radiology (VIR) is a

subfield in which endovascular disorders are treated

using catheters, typically guided through

fluoroscopy. Peripheral artery disease, deep vein

thrombosis, and abdominal aortic aneurysms are

examples of some of the conditions treated through

VIR.

VIR is one of the fastest growing fields in

radiology as it combines surgical procedures with real

time medical imaging and visualization. In order to

accomplish this, special operating suites in the

hospital have been designed with live diagnostic

imaging (fluoroscopy) in addition to the standard

surgical equipment. In these operating suites several

large monitors surround the room to display the live

data from the scanners.

Physicians wear heavy protective vests while they

operate. These surgeons watch the images come on

the screens as they thread catheters through the

patient’s vascular system (BSIR, 2017).

Due to the need for multiple large pieces of

equipment to be installed within the operating suite,

space is at a premium. Often, the monitors displaying

VIR data and images are situated away from the

physician’s view of the patient and catheter controls,

causing the physician to consciously keep their hands

in the right position while their view is, in some cases,

at 90 degrees, or even behind their view of the patient.

This issue is unfortunately common, surgeons often

have to coordinate their movements with monitors in

distant locations while trying not to damage the

patient, an example is shown in Figure 1. It is delicate

work that can easily go wrong (Shinohara, 2015).

Due to the heavy vests that must be worn and the

badly positioned monitors, many surgeons using VIR

present neck and spinal pain and injuries from

constantly turning to see the images (Sacks et al,

2003). Not only does this present a health risk for the

surgeon, but it creates a larger risk in the patient as

erroneous movement of the catheter can lead to injury

or potentially death.

Bartlett, C., LeRoy, N., Schoﬁeld, D., Ford, J. and Decker, S.

Assessing the Feasibility of using Augmented Reality to Visualize Interventional Radiology Imagery.

DOI: 10.5220/0008947701690176

In Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2020) - Volume 3: IVAPP, pages

169-176

ISBN: 978-989-758-402-2; ISSN: 2184-4321

169

Figure 1: A surgeon rotates to view data on a monitor.

Common ergonomic problems during IR include

(Shinohara, 2015) :

• Neuromuscular fatigue due to heavy lead

protectors.

• Inappropriate work postures from unfavourable

arrangements of imaging equipment and their

displays.

• Impaired manoeuvrability due to placement of

imaging equipment and their displays.

• Eyestrain due to dim lighting.

• Neuromuscular fatigue during small calibre

catheter placement.

• Lack of intuitive usability and standardization of

IVR devices and their instruction manuals.

Shinohara (2015) specifically stated that the

configuration of the X-ray apparatus, displays and the

procedural table in the operating theatre usually

cannot be freely arranged, forcing the physicians to

manipulate the devices and their posture to

ergonomically poor positions.

This paper describes work undertaken to alleviates

the risk of injury during VIR operations, increase

perceived accuracy and precision, and potentially

decrease the surgical duration through the

introduction of an Augmented Reality (AR) based

visualization system into the operating theatre.

1.1 Medical Applications of AR

The use of VR has been widespread in the medical

field over the past few decades. The key application

area has primarily been in the realm of training

simulators. Clinicians can learn crucial perspectives

on surgical anatomy and repeatedly practice surgical

procedures until they gain the required skills before

performing surgery on ‘live’ patients. A number of

commentators have stated that this educational

technology is potentially as important to surgery as

the flight simulator is to aviation (Satava, 1993;

Ahlberg et al, 2002; Gallagher et al, 2005).

In recent years, many companies have announced

new AR products, specifically for the medical field

and the use of AR in many other fields of surgery

continues to be reported (ITN, 2017; Fritz et al 2012).

AR surgical applications have been reported in the

fields of maxillofacial surgery (Badiali et al, 2014),

dental surgery (Wang et al, 2014), soft tissue surgery

(Mountney, 2014), and endoscopic surgery (Ishioka

et al, 2014).

The use of AR in surgery is not a new idea, the

use of such systems have been discussed by

academics for many years. The first ‘see-through’

HMDs used for medical purposes were monocular

optical see-through devices which were typically

used to display a patient’s vital information (Keller,

State, and Fuchs, 2017).

Chen et al. (2015) used an optical ‘see-through’

HMD for surgical navigation. Using this AR surgical

navigation system, the surgeon can view a fused

image that virtually displays anatomical structures

such as the soft tissue, blood vessels and nerves

within the intra-operative natural environment. A

preoperative CT scan is performed on the patient, and

the obtained images are segmented so that 3D models

of the hard and soft tissue can be reconstructed. The

AR system is then used to integrate the virtual model

of the patient’s anatomical structures with the real

anatomical structures to aide in surgical navigation.

Diaz et al. (2017) undertook an interesting project

using the Google Glass display for image-guided

brain tumor resection. During brain tumor removal,

neurosurgeons look back and forth between the

surgical field and the navigation display which can

create a delay or, in worse cases, surgical error. The

resulting video feed was transmitted wirelessly to the

Google Glass display. The researchers did note that

cognitive attention could not be simultaneously given

to the operative field and the Google Glass display,

due to the fact that this display technology cannot

superimpose its image onto the surgical field, forcing

users to switch from one view to the other.

Although, at present there is no reported work on

real-time visualization of IR data using AR

techniques in a surgical situation, there is comparable

work being undertaken with Magnetic Resonance

Imaging (MRI) data. Marker et al (2017) report on

work undertaken on MRI-guided paravertebral

sympathetic injections utilizing AR navigation and

1.5 T MRI scanner. The underlying premise for using

imaging guidance is accurate needle placement in

order to provide optimal treatment while avoiding

problems. The study showed that the combination of

IVAPP 2020 - 11th International Conference on Information Visualization Theory and Applications

170

image-overlay navigation and the image quality of

interventional MRI at 1.5 T was able to readily define

accurate needle paths and provide safe needle

guidance in all cases (Marker et al, 2017).

The only similar project to the work described in

this paper was undertaken by the Maryland Blended

Reality Center (MBRC) at the University of

Maryland. In 2018, MBRC performed a live

demonstration where a trauma surgeon at the Cowley

Shock Trauma Center, performed an ultrasound

examination of a patient’s heart using AR displayed

on a Microsoft HoloLens. The success of this

demonstration led researchers to state that they

believe that AR and VR will become a widespread

technology used in surgery, making procedures

simpler and easier (Salopek, 2018).

2 MONITORING

The goal of this project is to improve the current use

of IR with vascular surgical procedures, leading to

fewer mistakes and patient complications. The

research project aims to reach this goal by reducing

the risk of injury during IR procedures by increasing

surgical accuracy and precision. This will be achieved

by decreasing the duration in which the radiologist is

looking away from the patient by introducing an AR

based visualization system into the operating suite.

2.1 Data Collection

Initially, data was collected on the existing conditions

under which the surgeons work. Presently,

radiologists at Tampa General Hospital work while

viewing the IR images on a monitor located on a

swinging arm within the operating theatre. Initial data

collection was undertaken in two discrete phases.

The first phase involved monitoring VIR

procedures and collecting data on how the physician

was positioned when they looked at the monitor with

the live IR feed. The second phase involved timing

how long the physician looked at the fluoroscopy feed

on the monitor with a stop watch.

The purpose of this data collection was to develop

a baseline which could be used to demonstrate

whether the AR headset could significantly reduce the

time the physician spends viewing the monitor feed.

2.1.1 Physician Position

Four lengthy VIR procedures were monitored and the

radiologist’s viewing of the monitor containing the

live visualization was recorded. The radiologist’s

head position, whether their hands were working,

whether they were standing upright or leaning over,

and the monitor’s position in relation to the

radiologist were all logged during each procedure

(Tables 1 and 2).

Table 1: Head rotations during each procedure.

Procedure

Number

80°

Rotation

60°

Rotation

55°

Elevation

Total

Table 2: Time viewing monitor during each procedure.

Procedure

Number

Surgery

Time

Viewing

Time

Working

& Viewing

2h 35m

3m 54s

3m 0s

1h 26m

12m 35s

7m 33s

1h 15m

12m 28s

6m 12s

2h 53m

13m 29s

8m 42s

Total

8h 9m

52m 26s

25m 27s

2.1.2 Physician Timing

Four additional, shorter procedures were viewed

while the physician’s time spent looking at the live

visualization was recorded using a stopwatch.

Recording began when the physician’s gaze shifted

towards the live monitor feed and ended when the

physician’s eyes shifted to any other object within the

room (Table 3).

Table 3: Time viewing monitor during each procedure.

Procedure

Number

Surgery

Time

Viewing

Time

28m

8m 11s

30m

12m 44s

20m

6m 15s

28m

4m 42s

Total

1h 46m

31m 52s

The State University of New York at Oswego and

the USF Health Morsani College of Medicine at the

University of South Florida in Tampa experimented

initially with two AR visualizations modalities:

• A micro-projector that could sit on a stable surface

and project an image onto the patient.

• A monocular headset that could be easily moved

around the physician’s line of vision.

Due to problems with both systems, it was

decided to take a step back, identify a different AR

syatem and undertake some formal and rigorous

Assessing the Feasibility of using Augmented Reality to Visualize Interventional Radiology Imagery

171

experimentation to empirically demonstrate the

benefits of introducing this technology into and

operating suite to view real time IR data.

3 EMPIRICAL TESTING

In deciding on the next steps to take, the team

investigated the work of the team from the MBRC,

viewing ultrasound visualisations on a Microsoft

Hololens (Salopek, 2018). After extensive

discussions with the medical faculty and surgeons in

Tampa, it was decided to take a similar path and

investigate the use of a Hololens to visualize real time

AR data during surgical procedures.

Initial trials were undertaken with the Hololens

allowing the surgeons to test the equipment and

discern whether they thought they could use it, or if

they would find it too bulky and obtrusive. Each

radiologist who tried it, gave a positive review, so it

was decided to proceed on this project using a

Microsoft Hololens.

Hence, the HMD that was selected for this phase

of the project was the Microsoft HoloLens (1st gen).

This was chosen due to its compact size, low weight

(579g), Wi-Fi and Bluetooth connectivity, see-

through holographic lenses, high output resolution

(1268x720), and battery life of approximately 2-3

hours. In addition, the HoloLens had the ability to

place projection-based AR anywhere within the

physician’s line of sight.

Following the problems experienced with the

Brother AiRScouter AR headset in a surgical setting

it was decided that the team should undertake some

empirical work to demonstrate the effectiveness of

this equipment in a surgical setting. It was decided to

run a sequence of experiments using an AR based

system, asking participants to undertake a series of

tasks that simulated the clinician viewing the imaging

feed, the patient [their task], and their hands with

minimal head movements. This study would then

compare the usage of an AR headset to complete the

tasks in order to evaluate whether the use of an AR

device would increase perceived accuracy, as well as,

speed efficiency of the tasks undertaken. We

hypothesize that the AR HMD will increase

efficiency, by decreasing timed tasks and neck/back

discomfort.

3.1 Experimental Method

These experiments attempted to test whether the

HoloLens will increase efficacy in the specific Tampa

operating suite under consideration. In this operating

suite, surgeons consciously keep their hands in the

right plane while frequently turning their head 60˚ to

80˚ away from the patient. A mock operating suite

was constructed to simulate the conditions under

which the surgeons operate (Figure 2).

During the experiments undertaken, participants

were requested to sit facing the workspace, monitor

A was set up above the workspace and was intended

only to be used in the high-level task phase of the

experiment. Monitor B is meant to imitate the VIR

monitor a surgeon would turn towards and was used

during both low-level and high-level tasks. Monitor

B was set up at an angle of 80˚ from the view towards

monitor A to simulate the operating suite conditions.

A video camera was set up to capture and record the

movements each participant makes while undertaking

the experimental tasks.

Figure 2: Experimental setup, a mock surgical suite.

The experiments were designed to require

participants to undertake tasks in which they would

need to coordinate their hand movements using

information displayed on a monitor and/or the AR

headset to determine which is more efficient and

allows for greater levels of focus on the task at hand.

In order to test whether the AR headset

visualization would outperform data displayed on a

static monitor (the control group), two different

skilled tasks were created, a high-level task and a

low-level task :

The high-level task is meant to require a sense of

direction, where navigation instructions come from

data and images displayed in a visualization that is

displayed separately to the task being undertaken.

This high-level task has the participants navigating

and completing a maze on Monitor A while the maze

IVAPP 2020 - 11th International Conference on Information Visualization Theory and Applications

172

layout is provided on either the Hololens or Monitor

B located at an angle of 80˚ on their right.

The low-level task was intended to be simpler than

the high-level task, but to involve a higher level of

haptic, physical manipulation – simulating the

physical operations a surgeon must perform while

operating. Here the participant will construct a Lego

object on the workspace, while the instructions are

provided on either the Hololens or Monitor B located

at an angle of 80˚ on their right side.

Each participant undertakes both tasks, one with

instructions being displayed on an AR headset, the

other getting instructions from monitor B. Half the

participants use the AR headset for the low-level task

and Monitor B for the high-level task, the other half

switch the visualization modalities.

The dependent variables within these

experiments are the time taken to complete the tasks,

the number of times the participants reference

monitor B, and the time’s spent looking at the

visualisations rather than at the task in hand.

This constructed environment will hopefully

simulate the radiology environment in such a way that

results from these tasks will not only correlate with

results found during testing in Florida but also

provide a justification for the introduction of this

technology into a surgical setting.

3.1.1 High-level Experiment

The high-level experiment utilized an immersive

dungeon game (Figure 3). Half of the participants

completed the high-level task using the HoloLens,

while the rest were asked to complete the tasking

using monitor to get instructions and navigation

information.

When using the AR HMD, the participant was

guided step by step, using several visual cues, to

complete the navigation task through the maze.

A series of contextual slides containing

navigation information were displayed and viewable

to the participant through the Hololens, this

information floated above monitor A, its position

varied slightly depending on the orientation of the

participant’s head while undertaking the task. In the

screen modality, the contextual navigation

information was displayed on monitor B.

When viewing through the AR HMD, the

participant was guided step by step, using several

visual cues, to complete the navigation task through

the maze.

A series of contextual slides containing

navigation information were displayed and viewable

to the participant through the Hololens, this

information floated above monitor A, its position

varied slightly depending on the orientation of the

participant’s head while undertaking the task. In the

screen modality, the contextual navigation

information was displayed on monitor B.

Figure 3: A participant using the game.

Figure 4: A Participant using the Lego.

3.1.2 Low-level Experiment

For the low-level LEGO task, participants are asked

to follow a step by step visual guide that, if followed

correctly, builds a LEGO bird in the workspace

provided in front of them (Figure 4).

A series of contextual slides containing construction

information were displayed and viewable to the

participant through the Hololens, this information

floated above monitor A, its position varied slightly

depending on the participant’s head orientation while

undertaking the task. In the screen modality, the

construction information was displayed on monitor

Assessing the Feasibility of using Augmented Reality to Visualize Interventional Radiology Imagery

173

Table 4: Experimental results, average completion times compared to time viewing visualization.

Game (Monitor B)

Game (Hololens)

Lego (Monitor B)

Lego (Hololens)

Average Time (s)

617.5

650.3

258.8

350.0

Number of Tasks

Time per Task (s)

36.3

38.3

18.5

25.0

Average View Time

2.87

4.35

3.31

4.07

Number of Views

Total View Time (s)

138.4

144.9

128.5

115.9

Viewing Ratio (%)

21.1

15.11

51.9

33.9

Table 5: Results of t Test comparing task times, number of views and view time (%) for the game task.

Modality

Mean

Std Dev

t Stat

t Crit

Total Time

Monitor

617.5

231.5

-0.28

2.23

0.78

Headset

650.3

169.4

Number

of Views

Monitor

38.8

26.8

0.79

2.22

0.44

Headset

29.5

10.0

Total

View Time

Monitor

138.4

173.2

-0.07

2.23

0.93

Headset

144.9

105.8

View Time

(%)

Monitor

19.8

22.6

-0.11

2.23

0.909

Headset

21.1

15.11

Table 6: Results of t Test comparing task times, number of views and view time (%) for the Lego task.

Modality

Mean

Std Dev

t Stat

t Crit

Total Time

Monitor

258.8

58.5

-1.37

2.23

0.20

Headset

350.0

151.6

Number

of Views

Monitor

39.0

8.2

2.32

2.22

0.04

Headset

28.2

7.9

Total

View Time

Monitor

128.5

12.9

0.92

2.23

0.75

Headset

115.9

30.8

View Time

(%)

Monitor

51.9

13.3

2.79

2.23

0.02

Headset

33.9

8.5

B. Discussion with medical professionals indicated

that these two tasks would sufficiently replicate the

navigation and haptic skills required during a surgical

procedure.

3.1.3 Procedure

There were a total of sixteen participants in this study,

who each undertook both tasks using a different

information modality for each task. The average age

of the participants was 22 and 66% of the participants

were female. It was felt that although the number of

participants was small, the sample size was large

enough to provide meaningful results.

The data is discreetly arranged into four

experimental conditions, for a 2x2 independent

sample t test. The modality of the display method (AR

headset, monitor B) was the between-subjects factor

and the difficulty of tasks (high-level, low-level) was

the within-subjects factor. The rejection level for all

analyses was set at p = .05.

A questionnaire was provided post experiment

and completed by every participant. This

questionnaire measured device comfort, image

quality rating and device satisfaction. It also

measured the perceived speed, accuracy, and eye

fatigue that the participant experienced.

4 RESULTS AND DISCUSSION

A summary of the results of the experiment are shown

in Table 4. The average completion time of the low-

level Lego task was lower than the completion time

of the high-level Game task. This was due to the

nature of the tasks themselves. The Lego task

consisted of 14 separate steps whereas the Game task

had 17 individual steps.

Participants viewed the AR headset visualization,

on average, for longer than the image on the monitor,

even when view times included the times taken for

IVAPP 2020 - 11th International Conference on Information Visualization Theory and Applications

174

the participants to rotate their head to see the screen.

This could be due to the unfamiliarity of the

participants with viewing information in the Hololens

or perhaps it took participants longer to focus and

read information from the Hololens display.

The majority of participants looked at the monitor

more often than looked at the visualization on the AR

headset. It is difficult to understand why this occurred

it could perhaps be because participants are used to

glancing at a monitor, whereas looking at the

Hololens screen was a more thoughtful action.

Participants viewed the Game visualization, on

average, for less than the Lego instructions on each

view. The time taken viewing the Lego instructions

was high ranging from 34% of task time on the

Hololens to 52% of task time on the monitor. Again

this is thought to be primarily related to the nature of

the task where participants had to spend more time

understanding the complex construction information,

rather than quickly checking their location on a map.

The visualization viewing time for the game task

was similar on both the Hololens and monitor, with

the monitor viewing time being slightly less.

However, the average time spent viewing the Lego

visualization on the Hololens was much less, as a

percentage of task time, than the time spent viewing

Running a number of t tests on the metrics from

the game task shows little significance between the

monitor visualization and viewing the visualizations

on the AR headset or on the monitor (Table 5).

This indicates that during the game task when

using either an AR headset or a monitor :

• The difference in time taken to perform the game

task is not significant.

• The difference in number of views of the

visualization is not significant.

• The difference in the total time viewing the

visualization is not significant.

• The difference in percentage of time spent looking

at the visualization is not significant.

Running a number of t tests on the metrics from

the Lego task however, shows some significance

between viewing the visualizations on the AR headset

or on the monitor. This indicates that during the Lego

task when using either an AR headset or a monitor :

• The difference in time taken to perform the game

task is not significant.

• The difference in number of views of the

visualization is significant.

• The difference in the total time viewing the

visualization is not significant.

• The difference in percentage of time spent looking

at the visualization is significant.

A possible explanation for the significance in the

different lower number of views was explained

above, where participants are used to glancing at a

monitor, whereas looking at the Hololens screen was

a more thoughtful action, taken as and when needed.

The difference in the view time can possibly be

explained by the time taken to rotate the head when

looking at the monitor. If this is removed from the

time looking at the screen, the two values would

become closer and the difference less significant.

While it is interesting to look at the time taken

with each modality, perhaps a more useful measure is

whether the Hololens worked as effectively as the

monitor as a visualization tool. Qualitative survey

results which were rated on a 5-point Likert scale :

• Participants rated the Hololens comfort at 3.2

• Image quality of the Hololens at 4.2

• Image quality of the monitor at 4.4

• Eye fatigue with the Hololens at 3.6

5 CONCLUSIONS

This study aimed to alleviate the risk of injury during

VIR operations, increase perceived accuracy and

precision, and potentially decrease the surgical

duration through supplementing the operating suite

with an AR based viewing system. It is believed this

would grant a clinician, a greater range of movement,

decreased cognitive load and improved focus.

We found evidence that shows a significance

regarding the number of times the participant’s

looked at the screen during the Lego task, which

required the participants to understand a range of

complex instructions (Table 6). Although the

participants viewed the visualization on the Hololens

less, they did look at the visualization for longer

periods of time. This could have been affected by a

participant’s familiarization with using a monitor.

The time spent viewing the visualization during

the game task was similar on both the Hololens and

monitor. However, the average time spent viewing

the Lego visualization on the AR headset was

significantly less, as a percentage of task time, than

the time spent viewing the monitor visualization. If

the time taken to rotate the head to the monitor is

factored in then the viewing time becomes equivalent.

The crucial aspect of this work involved

determining the effectiveness of the AR headset as a

replacement for the badly positioned monitor.

Participants rated the image quality of the

visualization on the AR headset slightly lower than

the monitor. A couple of participants complained that

Assessing the Feasibility of using Augmented Reality to Visualize Interventional Radiology Imagery

175

the HoloLens sat heavily on the bridge of their nose

during the experiment. This is perhaps an alternative

form of discomfort, replacing the back/neck strain

experienced by the surgeons.

A few participants mentioned that there was some

eye strain when trying to view the visualizations on

the AR headset. Another limiting factor was that the

virtual object rendered by the HoloLens may begin to

fade out if participants move their head or do not have

their view completely aligned. However, it should be

noted that the head movements required to correct

this problem involve only a few degrees of rotation.

However, it is perhaps important to note that

overall, ten participants greatly preferred the

Hololens to the monitor.

This experiment demonstrated the potential of our

hypotheses, that the implementation of an AR headset

as a visualization tool in a surgical setting could

increase the efficiency of timed tasks and decrease

neck/back pain among medical practitioners.

ACKNOWLEDGEMENTS

Many thanks to Bruce Zwiebel, MD and all the other

medical staff from the Morsani Center for Advanced

Healthcare in Tampa, Florida who donated their time.

We acknowledge the funding support of the

SCAC and the Festa Fellowship Programs at the State

University of New York.

Thank you to Brother International for providing

the Brother AiRScouter-200B for the initial trials.

REFERENCES

Ahlberg, G., Heikkinen, T., Iselius, L., Leijonmarck, C. E.,

Rutqvist, J., Arvidsson, D., 2002. Does Training in a

Virtual Reality Simulator Improve Surgical

Performance? Surgical Endoscopy, 16(1), 126-129.

Badiali, G., Ferrari, V., Cutolo, F., Freschi, C., Caramella,

D., Bianchi, A., Marchetti, C., 2014, Augmented

Reality as an Aid in Maxillofacial Surgery: Validation

of a Wearable System Allowing Maxillary

Repositioning. Journal of Cranio-Maxillofacial

Surgery, 42(8), 1970-1976.

British Society of Interventional Radiology (BSIR), 2017,

The Home of Image Guided Surgery, URL at

http://www.bsir.org/patients/what-is-interventional-

radiology/ (accessed 12/08/2019)

Diaz, R., Yoon, J., Chen, R., Quinones-Hinojosa, A.,

Wharen, R., Komotar, R., 2017, Real-Time Video-

Streaming to Surgical Loupe Mounted Head-Up

Display for Navigated Meningioma Resection, Turkish

Neurosurgery, 1.

Gallagher, A. G., Ritter, E. M., Champion, H., Higgins, G.,

Fried, M. P., Moses, G., Satava, R. M., 2005. Virtual

Reality Simulation for the Operating Room:

Proficiency-Based Training as a Paradigm Shift in

Surgical Skills Training, Annals of Surgery, 241(2),

364-372.

Imaging Technology News (ITN), 2017, Philips Announces

New Augmented-Reality Surgical Navigation

Technology, URl at

https://www.itnonline.com/content/philips-announces-

new-augmented-reality-surgical-navigation-

technology (accessed 12/08/2019).

Ishioka, J., Kihara, K., Higuchi, S., Nakayama, T.,

Takeshita, H., Yoshida, S., Fujii, Y., 2014. New Head-

Mounted Display System Applied to Endoscopic

Management of Upper Urinary Tract Carcinomas,

Brazilian Journal Urology, 40(6), 842-845.

Fritz, J., U-Thainual, P., Ungi, T., Flammang, A.J., Cho, N.,

Fichtinger, G., Iordachita, I.I., Carrino, J.A., 2012,

Augmented Reality Visualization With Image Overlay

for MRI-Guided Intervention: Accuracy for Lumbar

Spinal Procedures With a 1.5-T MRI System, American

Journal of Roentgenology 2012 198(3), W266-W273.

Keller, K., State, A., Fuchs, H., 2008. Head Mounted

Displays for Medical Use, Journal of Display

Technology, 4(4), 468-472.

Marker, D. R., Paweena, U., Thainual, T. U., Flammang, A.

J., Fichtinger, G., Iordachita, I. I., Fritz, J., 2017. 1.5 T

Augmented Reality Navigated Interventional MRI:

Paravertebral Sympathetic Plexus Injections.

Diagnostic and Interventional Radiology, 23(3), 227.

Mountney P., Fallert J., Nicolau S., Soler L., Mewes P.W.,

2014, An Augmented Reality Framework for Soft

Tissue Surgery. In: Golland P., Hata N., Barillot C.,

Hornegger J., Howe R. (eds) Medical Image Computing

and Computer-Assisted Intervention. Lecture Notes in

Computer Science, 8673, Springer, Cambridge.

Sacks, D., McClenny, T. E., Cardella, J. F., Lewis, C. A.,

2003. Society of Interventional Radiology Clinical

Practice Guidelines, Journal of Vascular and

Interventional Radiology, 14(9), S199-S202.

Salopek, J., 2018. Augmented and Virtual Reality in the IR

Suite, IRQuarterly, URL at https://connect.sirweb.

org/e-irq/participate/viewirqarticle?DocumentKey=4bf

c61a8-8b03-4e62-b78e-3e5a25fd88a4 (accessed

12/08/2019)

Shinohara, K., 2015. Ergonomic Investigation of

Interventional Radiology, Procedia Manufacturing, 3,

308-311.

Satava, R. M., 1993. Virtual Reality Surgical Simulator.

Surgical Endoscopy, 7(3), 203-205.

Wang, J., Suenaga, H., Hoshi, K., Yang, L., Kobayashi, E.,

Sakuma, I., Liao, H., 2014. Augmented Reality

Navigation with Automatic Marker-Free Image

Registration Using 3-D Image Overlay for Dental

Surgery, IEEE Transactions Biomedical Engineering,

61(4), 1295–1304.

IVAPP 2020 - 11th International Conference on Information Visualization Theory and Applications

176