IEEE Systems, Man and Cybernetics Magazine - July 2018 - 11

(a)

(b)

(c)

Figure 4. Screenshots from the surgeon's POV on the Oculus rift display while executing the task of grabbing a

sphere with tweezers.

immersive 3-D system. We also observed that the majority
positioned cylinders without touching them, using the
of surgeons appreciated the simulator more than the 2-D
camera to look around in the environment. In the 3-D
system thanks to its ease of use. In particular, we noticed
case, we could also test the behavior of the Leap Motion: a
advantages of the ease of use on
moderator asked the surgeon to
visual interaction. It is comprehenswitch the camera source and persible because their professional
form an external task, move a real
experience is limited to using only
ball from one glass to another,
The proposed
common 2-D endoscopic systems.
before continuing the task executouchless approach
The surgeons' subjective contion. In both cases, the sphere was
siderations
about the real protocontrolled through the keyboard
was considered
type have been strongly positive.
arrow keys. Then, we asked the
sufficient and
They agree about the potentiality
surgeons to perform another test:
intuitive, while the
of this technology for real surgery.
extract a sphere from a surface
Qualitative results were gathered
with virtual surgical tweezers
Leap Motion sensor
about specific system features
(Figure 4) by means of a haptic
was regarded as one
according to the personal judgedevice (i.e., Oculus Touch) that
ment of each surgeon. We asked
simulates the tweezers. The surof the most reliable
them to complete a survey with
geons used their head for controltools involving hand
their feelings and votes about the
ling the endoscopic head and the
acceptance and usability of the
light in the environment. Each
tracking, having
proposed system. Different questask was monitored to compute
a high level
tions were inspired by the wellthe execution time and the surof accuracy.
known usability rules provided by
geons' mistakes. We have also colNorman's principles [11]. The surlected the personal judgments
vey helped us to retrieve data
about the 2-D and 3-D versions of
about comfort, control, and intueach task for each surgeon. Finalitiveness of the proposed new paradigm. For each paramely, we gathered qualitative considerations by surgeons
ter, we obtained remarkable feedback. The survey
about the overall comfort of the system.
significantly contributed to the proposed work and
Results
We amassed the results of 21 surgeons for each task in
each modality (i.e., 2-D and 3-D) and compared them. FigExecution Time and Mistakes
Task 1: Execution Time
ure 5 shows that for each task, we obtained a considerable
36.5
Task 1: Mistakes
35
reduction of time and mistakes from executing the operaTask 2: Execution Time
32
tions with the IVRE system with respect to the classical
Task 2: Mistakes
58
Task 3: Execution Time
43
2-D endoscopic approaches. The improvements go from
Task 3: Mistakes
43.5
32% to 58%, demonstrating that both precision and speed
0 10 20 30 40 50 60 70
have been consistently increased. These results show that
Percentage of Reduction from 2-D to 3-D Tasks
the proposed system could become a practical and innovative tool in the near future. Since the surgeons were accustomed to the classical 2-D surgical endoscopic systems,
Figure 5. a comparison of mistakes and execution
they experienced some disorientation when they tried our
time between 2-D and 3-D tasks.
Ju ly 2018

IEEE SyStEmS, man, & CybErnEtICS magazInE

Table of Contents for the Digital Edition of IEEE Systems, Man and Cybernetics Magazine - July 2018