IEEE Robotics & Automation Magazine - March 2022 - 74

Table 1. The average completion time in seconds.
Tools
Mean
W_DV
176.8
Standard deviation 79
W_3F
139
80.1
E_DV
154.3
56.6
E_3F
151.4
80.2
could not indicate a significant difference for one of the tested
tools due to high standard deviations. These results,
together with the UEF scores, highlight the need for a learning
period for both lay participants and experienced surgeons
when using more dexterous tools. There are strong
indications that more dexterous tools facilitated better
results in subsequent attempts when compared to a less dexterous
DV instrument configuration.
Table 2. The average number of collisions.
Tools
Mean
W_DV
27.8
Standard deviation 25.8
W_3F
62
52.7
E_DV
35.1
62
E_3F
31.9
80.2
Conclusions and Future Work
The purpose of this work was to assess the usability and
efficiency of an anthropomorphic, immersive teleoperation
concept in a virtual, minimally invasive surgical environment
to
demonstrate the dexterity limits and
teleoperation suitability in several
small-scale user studies. From our
SUS Scores
100
50
50 57.5 62.5
25
Users
Figure 13. The SUS scores of lay users (blue) and surgeons (red) for surgical task B.3.
80
65 52.5 52.5 65
80
95
62.5
32.5
preliminary set of trials, it was obvious
that adding dexterity to the
robotic surgical instruments does
not necessarily reap benefits in terms
of speed and efficiency when performing
simple pick-and-place tasks.
For the precision grasping task (surgical
task B.1), the DV grasper was a
better choice for most of our participants,
while the 3 F grasper seemed
80
90
70
60
50
40
30
20
10
only to add more mental workload and not enhance grasping
efficacy. Similarly, the presence of the E on the shaft
and the associated complexity of control did not provide
any performance improvement in surgical task B.2 but
facilitated better results with the more complex task (surgical
task B.3), which included object avoidance and reaching
behind virtual structures.
The SUS feedback from the first two trial groups
E_3F E_DV W_3F W_DV
NASA-TLX
Figure 14. NASA scores for the four shaft/end-effector combinations
when performing surgical task B.3.
Table 3. The surgeons' average scores.
Time (s)
Tools
Mean
Standard
deviation
Tools
Mean
Standard
deviation
E_3F
149.68
21.01
E_DV
196.51
E_3F
42.5
24.75
E_DV
48
W_3F
172.04
Collisions
W_3F
81
W_DV
114
145.67
W_DV
215.52
164.98
showed that participants found the anthropomorphic primary
control easy and intuitive to use, although unnecessarily
complex. However, in the last set of trials (surgical
task B.3), this approach proved to have an advantage over
the lower dexterous shaft and grasper. After initial control
difficulties in the first attempt mainly due to the constrained
motion of the tools and use of the clutch, task
performance improved in all subsequent attempts. The
results shown in the " Dexterous Instruments Comparison
(Surgical Task B.3) " section indicate that increased dexterity
of the primary/secondary teleoperation can provide
better efficiency for more complex tasks and potentially
reduce training times for surgeons, while improving
speed of movement and safety of the anatomical structures
(Tables 1 and 2). A larger number of participants over
a longer test period, primarily surgeons experienced in
robotic surgery, would help achieve more statistically consistent
results of this concept. As a future prospect, further
co-development of the concept with surgeons experienced
in robotic surgery would be useful to test this concept in
more realistic simulations as well as on actual robotic systems,
and to provide more insight into usability and
74 * IEEE ROBOTICS & AUTOMATION MAGAZINE * MARCH 2022
Score
Score
IEEE Robotics & Automation Magazine - March 2022

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2022
