The Beyond 5G (B5G) Era of Next-Generation Digital Networks:

Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

Maradona C. Gatara

, Mjumo Mzyece

and Sijo J. Parekattil

3,4

Independent Researcher, South Africa

Business & Economics Department, Northwestern College, Orange City, Iowa, U.S.A.

College of Medicine, University of Central Florida, Florida, U.S.A.

Avant Concierge Urology, Winter Garden, Florida, U.S.A.

Keywords: Beyond 5G (B5G), Task-Technology Fit (TTF), Predictive Modelling, Haptic-Enabled Internet of Skills

(IoS), Remote-Robotic Surgery Applications, Human-in-the-Loop (HITL), Minimally Invasive Surgery

(MIS), Health Informatics.

Abstract: The coming Beyond 5G (B5G) era could mark a paradigm shift towards user-centric Quality of Experience

(QoE) centred network architectures. The infusion of QoE user requirements into network architectures will

be crucial for future ultra-reliable, ultra-low latency haptic-enabled Internet applications. One such application

will be the mission-critical use case of remote (tele-haptic) robotic surgery, signifying a transition towards

skillset delivery networks that will augment user task performance experience. In extending traditional

Quality of Service (QoS)-oriented networks to user focused QoE and with it, Quality of Task (QoT)

components, human users in a global control loop (such as robotic surgeons) will be capable of true-to-life

immersive remote task performance through the manipulation of objects in real-time, and of transcending

geographical distance. In this preliminary study using data elicited from 20 practising robotic surgeons (n =

20), we examine the emergence of a future B5G network and haptic-enabled Internet of Skills (IoS)

architecture, applied to the task-sensitive mission-critical use case of remote (tele-haptic) robotic surgery. We

conceptualise and demonstrate the use of non-linear Task-Technology Fit (TTF) predictive modelling to

empirically assess this futuristic use case, and in doing so, provide a novel QoE/QoT perspective of future

B5G communication networks.

1 INTRODUCTION

The emergence of Beyond 5G (B5G) networks such

as 6G networks (Giordani et al., 2020) and quantum

communication networks (Bassoli et al., 2021) offer

much promise. These digital networks of the future

will transcend the limits of current 5G network

technologies (Nawaz et al., 2019). Originally, the

traditional Internet was envisaged as a global

computer network, signifying a paradigm shift in 20

century economies (Shapiro & Varian, 1999). This

era brought forth the revolutionary Mobile Internet,

connecting billions of devices and computers,

disrupting whole 21

century economies and

industries (Dohler, 2018). In the present day, the

Internet of Things (IoT), predicted to tether trillions

of smart devices and positioned to redefine industries

of the coming decade, has come to the fore. These

Internets will, however, be overtaken by the

emergence of a haptic-enabled Internet whereby

highly responsive secure networks will support the

rendition of real-time haptic impulses remotely. This

would amplify the capacities of the IoT by

introducing a new element to human-machine

interaction via the development of immersive real-

time communications technologies (Pierucci, 2015).

In future Internets, haptics will take the form of two

key attributes: the transmission of touch and actuation

in real-time will extend traditional audio-visual

feedback of current systems via the support of both

tactile (cutaneous) and kinaesthetic modalities.

Firstly, the tactile (cutaneous) modality would render

data on the dimensions of surface, texture, and

friction. Secondly, the kinaesthetic modality would

relay data on force, torque, position, and velocity

112

Gatara, M., Mzyece, M. and Parekattil, S.

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote Robotic Surgery Applications.

DOI: 10.5220/0012384400003657

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2024) - Volume 2, pages 112-122

ISBN: 978-989-758-688-0; ISSN: 2184-4305

dimensions. With these transmission modes, human

users would be linked to remote environments with

more immersion. The sensations of sight and sound

augmented by audio-visual rendition and the

transmission of haptic impulses would be bi-

directional. Hence, touch would be detected by

imposing motion on an environment with feeling

rendered through a distortion or reactionary force.

Haptics will become critical to future Internet archite-

ctures with the emergence of the B5G era. A future

haptic-enabled Internet will shift conventional Quality

of Service (QoS) performance-related indicators

towards more dynamic, interactive and human-user-

centred Quality of Experience (QoE) and Quality of

Task (QoT) considerations (Gatara & Mzyece, 2023).

2 REMOTE ROBOTIC SURGERY

APPLICATIONS FOR

TELE-HAPTIC SURGICAL

TASK PERFORMANCE

B5G networks will enable the future Internet of Skills

(IoS) (Dohler, 2018). The performance of real-time

tele-haptic robotic surgery tasks is a mission-critical

application that leverages the ultra-reliable and ultra-

low latency requirements that will become

synonymous with B5G networks of the future. To

envision the connection between QoS, QoT, and QoE

components of a haptic-enabled IoS architecture, we

present this robotic telesurgery use case in Figure 1.

Figure 1: Haptic-Enabled Internet of Skills (IoS) for Tele-

Haptic Surgical Task Performance in Beyond 5G (B5G)

Networks.

In this scenario, a robotic surgeon with the

requisite expertise will be the Human-in-the-Loop

(HITL) supported to perform tele-haptic surgery

tasks. A master (control) and assistant surgical robot

in a remote-controlled environment must be

connected through a reliable high-speed

communication network to render real-time control

commands and multi-modal sensory data. This

enhanced form of tele-haptic surgery will require

high-precision manipulation and meet stringent

latency, jitter, and packet-loss metrics. Therefore,

future B5G networks will be expected to more

consistently and reliably ensure the ultra-low latency

and ultra-reliable characteristics necessary for

seamless two-way haptic feedback. On this basis, in a

future Internet, surgeons will be able to extend their

physical skillsets over remote geographical distances

via a B5G-supported telecommunications network.

Consequently, current shortages of surgeons and high-

quality surgical care, and long-distance limitations in

travel would be greatly reduced. Furthermore, surgical

precision and patient safety would be enhanced.

3 TASK-TECHNOLOGY FIT

(TTF) THEORY AND

PREDICTVE MODELLING

FOR REMOTE ROBOTIC

SURGERY APPLICATIONS

The theoretical construct of Task-Technology Fit

(TTF) denotes the measurement of the degree to

which the functional capacity of a tool or system is

adequate for user needs or requirements (Goodhue,

1995; Dishaw & Strong, 1998). The theory of TTF

can be traced to the earlier theories of Cognitive Fit,

which suggests that effective, efficient problem

solving relies on matching characteristics of problem

representation and problem task (Vessey, 1991, 1994;

Vessey & Galleta, 1991), and Task-System Fit, which

is “the fit between task requirements and the

functionality of the IS [Information Systems]

environment” (Goodhue, 1992). A TTF conceptual

model of a haptic-enabled IoS is proposed and

illustrated in Figure 2 (Gatara et al., 2021).

Figure 2: Conceptual Task-Technology Fit (TTF) Model

for Quality of Experience (QoE) with Quality of Task

(QoT) Perspective of a Haptic-Enabled Internet of Skills.

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

113

The model in Figure 2 links task and technology

characteristics in (i) the master (control) domain and

(ii) the remote (controlled) domains. First, task

characteristics denote the most critical needs of the

human technology user. User needs can be specified

as surgeons’ most critical task demands in remote

robotic surgery (tele-haptic surgical task

performance).

To perform critical minimally invasive robotic

surgery tasks (grasping, palpation, and incision), the

user (surgeon) concurrently uses (i) a manipulator

(hand controller) and touch haptic device (remote

controller) as part of the Human System Interface

(HSI) in the master (control) domain and (ii)

manipulators (grasper, palpation probe, and end-

effector tip (cutter)) in the remote (controlled)

domain.

4 INSTRUMENT SCALE (ITEM)

MEASURES FOR

TASK-TECHNOLOGY FIT

(TTF) MODEL VARIABLES

The Task-Technology Fit (TTF) model developed for

this research links task and technology characteristics

in (i) the master (control) domain and (ii) the remote

(controlled) domains.

First, task characteristics denote the most critical

needs of the human technology user (Nance, 1992).

User needs can be specified as surgeons’ most critical

task demands in remote robotic surgery (tele-haptic

surgical task performance). For example, these

include (i) control movement (motion) of remote

assistant robotic arms (telemanipulators) e.g. to

manipulate a needle drive (end effector) tool (surgical

instrument) with wrist-like movements (1A), (ii)

visualisation (with magnification) of the operative

(surgical) field (area) e.g. for immersive stereoscopic

view and endoscopic three-dimensional (3-D) High-

Definition (HD) imaging (2A), and (iii) feeling and

control of grasping force when operating on patient

e.g. to displace tender organs (retraction) and soft

tissue (clutching) (3A). The items used to measure

these dimensions are detailed in Table 1.

Second, technology characteristics denote critical

support functions for the most critical needs of the

task performer (human user) (Dishaw et al., 2002).

For example, there are critical corresponding support

tools used by the surgeon including (i)

interchangeable needle driver (end effector) tool

(surgical instrument) attached to a lateral robotic arm

with functional support i.e. movement up to 7

Table 1: Measurement Items for the Task (Characteristics)

Construct (TC).

Variable Scale Item Source

TC 1A Control movement (motion) of

remote assistant robotic arms

(telemanipulators) e.g. to

manipulate a needle driver (end

effector) tool (surgical

instrument) with wrist-like

movements.

Saracino

et al.

(2019),

Yang et

al.

(2013)

TC 2A Visualisation (with magnification)

of the operative (surgical) field

(area) e.g. for immersive

stereoscopic view and endoscopic

3-D HD imaging.

TC 3A Feel and control grasping force

when operating on patient e.g. to

displace tender organs (retraction)

and soft tissue (clutching).

TC 4A Palpation manoeuvres when

operating on patient e.g. to detect

neoplastic lesions in solid organs

(hollow viscus).

TC 5A Incision (dissection) when

operating on patient e.g. to cut

soft tissue without damaging

embedded vessels and nerves.

TC 6A Suturing when operating on

patient e.g. to insert needle

(puncture tissue), loop the suture

thread (stitch), and tie the knot.

TC 7A Feel and reproduce true-to-life

(realistic) haptic feedback when

operating on patient e.g. to sense

kinaesthetic (force/joint-related)

and vibrotactile (cutaneous/skin-

related) sensations.

Degrees of Freedom (DoF) (1B), (ii)

digitalstereoscopic camera (optic lens) with

progressive magnification up to 15 times (15x) (2B),

and (iii) interchangeable grasper tool (surgical

instrument) attached to a lateral robotic arm with

functional support i.e. laparoscopic forceps (5mm,

37cm) or fenestrated-grasper (3B). These identified

corresponding task (user need) and technology

(support function) characteristics (A and B pairs) will

be measured using five seven (7)-point Likert

measures on a scale from 1 (= to an extremely small

extent) to 7 (= to an extremely large extent) (Yang et

al., 2013). The items used to measure these

dimensions are detailed in Table 2.

HEALTHINF 2024 - 17th International Conference on Health Informatics

114

Table 2: Measurement Items for the Technology

(Characteristics) Construct (TC).

Variable Scale Item Source

TC 1B Control movement (motion) of

remote assistant robotic arms

(telemanipulators) e.g. via

interchangeable needle drive (end

effector) tool (surgical instrument)

attached to lateral robotic arm with

functional support i.e. movement up

to 7 DoF.

Saracino

et al.

(2019),

Yang et

al.

(2013)

TC 2B Visualisation (with magnification)

of the operative (surgical) field

(area) e.g. digital stereo scoping

camera (optic lens) attached to

lateral robotic arm with functional

support i.e. progressive

magnification up to 15x.

TC 3B Feel and control grasping force

when operating on patient e.g. Feel

and control grasping force when

operating on patient e.g. via

interchangeable grasper tool

(surgical instrument) attached to

lateral robotic arm with functional

support i.e. laparoscopic forceps

(5mm, 37cm) or fenestrated grasper.

TC 4B Palpation manoeuvres when

operating on patient e.g. via

interchangeable

laparoscopic/ultrasound probe tool

(surgical instrument) attached to

lateral robotic arm with functional

support i.e. single-use and

disposable with cross-section of less

than 15 x 10 mm (diameter of 5 to

12 mm).

TC 5B Incision (dissection) when operating

on patient e.g. via interchangeable

end-effector tip (cutter) tool

(surgical instrument) attached to

lateral robotic arm i.e. sterile Carbon

steel blade.

TC 6B Suturing when operating on patient

e.g. via interchangeable needle

driver (end-effector) tool (surgical

instrument) attached to lateral

robotic arm i.e. on CT-2 needles cut

to 6 inches (for placement 0-Vicryl

sutures).

TC 7B Feel and reproduce true-to-life

(realistic) haptic feedback when

operating on patient e.g. via force-

sensing for multiple degrees of

motion and force-awareness

(combined) i.e. sigma.7 haptic

(master) interface

(kinaesthetic/vibrotactile feedback).

The Use construct in Table 3 reflects the extent to

which the task performer has come to depend on the

technology tool and its support functions (Thompson

et al., 1991; Igbaria et al., 1997; Junglas et al., 2009).

Table 3: Measurement Items for the Use (Dependence)

Construct (UD).

Variable Scale Item Source

UD 1

I am very dependent on the use

hand telemanipulators (finger

controllers) to perform tasks

using robotic arms (with

attached surgical tools e.g.

needle driver).

Saracin

o et al.

(2019),

Yang et

al,

(2013)

UD 2 My work is highly dependent on

the use of hand telemanipulators

(finger controllers) to perform

tasks using robotic arms (with

attached surgical tools e.g.

probe).

UD 3 The use of hand

telemanipulators (finger

controllers) to perform tasks

using robotic arms (with

attached surgical tools e.g.

cutter) allows me to do more

than would be possible without

them.

The User Performance construct in Table 4 on the

other hand reflects the effectiveness, efficiency, and

quality with which tasks are completed using the

technology and its support functions to perform the

most critical tasks needed (Hiltz & Johnson, 1990;

Torkzadeh & Doll, 1999; Hou, 2012).

Five seven (7)-point Likert measures on a scale

from 1 (= to an extremely small extent) to 7 (= to an

extremely large extent) measure the Use and User

Performance outcomes resulting from the “Fit”

between Task and Technology characteristics

depicted in Figure 3. The presence of this “Fit” is

essential for optimal use and user performance

(Nance, 1992).

Figure 3: The Fit between Task and Technology

Characteristics.

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

115

Thus, task-technology fit (TTF) is examined for its

effects on Use and User Performance. The specific

items used to measure these dimensions are detailed

in Table 3 (above) and Table 4 (below).

Table 4: Measurement Items for the User Performance

Construct (UP).

Variable Scale Ite

Source

UP 1 The hand telemanipulators (finger

controllers) I use to control

assistant robot and perform tasks

using robotic arms (with attached

surgical tools e.g. grasper)

increases my productivity (easier

task execution

)

Saracino

et al

(2019),

Yang et

al (2013)

UP 2 The hand telemanipulators (finger

controllers) I use to control

assistant robot and perform tasks

using robotic arms (with attached

surgical tools e.g. probe) increases

my productivity (time reduction in

task com

letion

)

UP 3 The use of hand telemanipulators

(finger controllers) to perform

tasks using robotic arms (with

attached surgical tools e.g. optic

lens) decreases errors, increasing

quality (capability enhancement in

task execution).

5 DATA COLLECTION AND

DEMOGRAPHIC USER

PROFILE OF RESPONDENTS

We collected preliminary data from 20 practising

robotic surgeons (n = 20) via an electronic (online)

survey designed to elicit user responses.

There were 19 male users (95%) and 1 female user

(5%), mostly aged 51 years and above (40%) and

between 46 and 50 years (35%). There were 18 right-

handed dominant users (90%), plus 1 left-handed user

(5%) and 1 ambidextrous user (5%). Additionally, 17

robotic surgeons (85%) were trained as Senior

Faculty versus 3 as Junior Faculty (15%). Also, most

of the robotic surgeons (65%) were reported to have

undergone more than 10 simulator hours.

Furthermore, 9 users (45%) were reported to have

expert microsurgery experience, whereas 4 users

(20%) were proficient. A further 11 users (55%) had

expert robotic experience, whereas at least 7 robotic

surgeons (35%) were expert-level laparoscopic

practitioners. Notably, 5 users (25%) reported

proficient videogame experience. The respondent

user demographic profile for this preliminary cohort

of practising robotic surgeons (n = 20) is provided in

Table 5.

Table 5: Respondent User Demographic Profile (n = 20).

Variable(s) Frequenc

Percent (%)

Gender

Male 19 95%

Female 1 5%

Total 20 200%

Missing 0 0%

Age

36-40 years 3 15%

41-45

ears 2 10%

46-50

ears 7 35%

51 years and

above

8 40%

Total 20 100%

Missin

00%

Hand Dominance

Right-Hande

18 90%

Left-Hande

15%

Ambidextrous 1 5%

Total 20 100%

Missin

00%

Training Level

Junior Facult

3 15%

Senior Facult

17 85%

Total 20 100%

Missin

00%

Simulator Hours

None 2 10%

Less than 5

Hours

2 10%

–

10 Hours 3 15%

More than 10

Hours

13 65%

Total 20 100%

Missing 0 0%

Microsurgery Experience

Novice 4 20%

Advanced

Beginne

0 0%

Competent 3 15%

Proficient 4 20%

ert 9 45%

Total 20 100%

Missing 0 0%

Robotic Experience

Novice 1 5%

Advanced

Beginne

1 5%

Competent 4 20%

Proficient 3 15%

ert 11 55%

Total 20 100%

Missing 0 0%

HEALTHINF 2024 - 17th International Conference on Health Informatics

116

Table 5: Respondent User Demographic Profile (n = 20)

(cont.).

Variable(s) Frequenc

Percent (%)

Laparoscopic Experience

Novice 6 30%

Advanced

Beginne

1 5%

Competent 4 20%

Proficient 2 10%

ert 7 35%

Total 20 100%

Missin

0 0%

Videogame Experience

Novice 6 30%

Advanced

Beginne

3 15%

Competent 4 20%

Proficient 5 25%

ert 2 10%

Total 20 100%

Missin

0 0%

6 MEASUREMENT

INSTRUMENT (CONSTRUCT)

RELIABILITY AND VALIDITY

A Partial Least Squares Structural Equation

Modelling (PLS-SEM) algorithm was run to estimate

parameters of measurement model constructs.

Confirmatory Factor Analysis (CFA) was conducted

to test construct measures for their internal

consistency, convergent, and discriminant validities.

PLS-SEM functions efficiently with small sample

sizes and attains high statistical power levels with

small sample sizes even when the data is non-

parametric or highly skewed (Hair et al., 2021), such

as the preliminary sample (n = 20) used for

preliminary nature of analysis in this study.

Composite Reliability (p

) scores for the dimensions

of Task, Technology, Use, and User Performance were

satisfactory. Composite Reliability (p

) ranged from

0.000 to 1.000, with higher values indicating higher

levels of reliability (Hair et al., 2021). In more advanced

research however, values between 0.700 and 0.900 are

generally considered as satisfactory (Nunnally &

Bernstein, 1994; Hair et al., 2021). The composite

reliability scores for each latent Task, Technology, Use,

and User Performance dimensions were found to be

satisfactory (greater than 0.700). Thus, internal

consistent reliability was established.

The descriptive statistics for these four

dimensions are presented in Table 6.

Table 6: Descriptive Statistics.

Variable Range Mean SD Skewness Kurtosis

Tas

5.55 4.721 1.404 -0.189 0.127

Technology 6.00 4.173 1.578 0.073 0.014

Use 4.00 5.360 1.125 0.023 -0.502

User

Performance

6.20 5.391 1.466 -1.095 2.038

Further, the Average Variance Extracted (AVE)

values for each of the Task, Technology, Use, and

User Performance constructs exceeded the prescribed

threshold of 0.500 (Hair et al., 2021). Thus, results

also reflected acceptable convergent validity.

The Task, Technology, Use, and User

Performance constructs were also tested for their

discriminant validity.

First, their indicator cross-loadings were

evaluated. The outer loadings on all indicators on the

associated construct did not score higher than any of

its cross-loadings (correlations) on other constructs.

Therefore, discriminant validity was established.

Results of indicator cross-loadings are presented in

Table 7.

Table 7: Cross-Loadings.

Task Technol

Use User

Performance

TaC1 0.705 0.383 0.216 -0.191

TaC2 0.732 0.587 0.407 -0.017

TaC3 0.867 0.732 0.253 0.210

TaC4 0.611 0.543 0.209 0.316

TaC5 0.793 0.511 0.358 0.170

TaC6 0.817 0.539 0.340 0.183

TaC7 0.845 0.778 0.283 0.101

TeC1 0.664 0.618 0.391 0.097

TeC2 0.602 0.761 0.444 0.194

TeC3 0.768 0.875 0.445 0.308

TeC4 0.661 0.819 0.308 0.476

TeC5 0.546 0.818 0.497 0.486

TeC6 0.584 0.854 0.549 0.324

TeC7 0.569 0.835 0.440 0.412

UDe1 0.429 0.285 0.618 0.245

UDe2 0.196 0.456 0.855 0.471

UDe3 0.410 0.556 0.935 0.518

UP1 0.141 0.273 0.565 0.883

UP2 0.214 0.474 0.484 0.964

UP3 0.203 0.045 0.224 0.116

Second, the Fornell-Larker Criterion was used to

further establish discriminant validity. The square

root of the AVE for each of the Task, Technology,

Use, and User Performance variables was higher than

correlations between these constructs and other latent

variables. Therefore, discriminant validity was

further established. Results of the Fornell-Larker

criterion valuation with the square root of the

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

117

reflective constructs’ AVE on the diagonal, the means

and standard deviations of study constructs, and

correlations between the constructs in the off-

diagonal positions, are presented in Table 8.

Table 8: Fornell-Larker Criterion Results.

Mean

(SD)

Task Technology Use

User

Performance

Task

4.72

(1.40)

0.771

Technology

4.17

(1.58)

0.769 0.801

Use

5.36

(1.12)

0.398 0.552 0.814

User

Performance

5.39

(1.47)

0.181 0.431 0.529 0.758

Third, we further assessed discriminant validity using

the Heterotrait-Monotrait (HTMT) ratio of

correlations. Using HTMT as a criterion, all ratios

were found to be below the conservative threshold

value of 0.85, thus ascertaining the discriminant

validity of the Task, Technology, Use, and User

Performance measures. Results of the HTMT ratio

values for all pairs of constructs in the measurement

model are presented in Table 9.

Table 9: Heterotrait-Monotrait (HTMT) Ratio of

Correlations.

Task Technology Use

User

Performance

Tas

Technolog

0.868

Use 0.517 0.646

User

Performance

0.409 0.478 0.778

7 RESULTS: POLYNOMIAL

REGRESSION AND RESPONSE

SURFACE ANALYSIS

We modelled a relationship between Task and

Technology characteristics as independent variables

and Use and User Performance as dependent

variables, respectively, as a non-linear function. This

approach can have greater explanatory potential than

traditional moderated regression analyses. Moreover,

it can be used as an alternative method, as it outputs

more precise information on combinations

(interactions) of variables, beyond the results of more

conventional moderator analyses.

First, polynomial regression (Edwards, 1993) was

used to examine task and technology impacts on use

and user performance.

Latent variable scores obtained from PLS-SEM

analysis were used to compute Task (X) and

Technology (Y) characteristics, their interaction

(X*Y), and the quadratic terms (X

, Y

), in turn used

to predict Use and User Performance outcomes (Z) as

per the following polynomial equation [where b

denotes the respective beta coefficients for

corresponding X, Y, and Z terms, and e represents a

random disturbance term]:

Z = b

+ b

X + b

Y + b

+ b

XY + b

+ e (1)

where:

Z = Use or User Performance

X = The Task

Y = The Technology

The above variables were centred at their

midpoints i.e. ‘4’ for 7-point Likert scales. Centring

is recommended for polynomial regression analyses

(Edwards, 1994). Further, Aiken and West (1991)

suggested that centering reduces the likelihood of

collinearity. With the above formula, coefficients for

the terms X (b

), Y (b

), X

), XY (b

) and Y

)

were obtained.

Table 10: Polynomial Regression Results (Use).

Use

Predictor

Beta (β)

Standard

Error

Constant (b

) 1.222*** 0.336

Task

(

)

-0.215 0.580

Technolo

(

)

0.541 0.516

Tas

(

)

0.067 0.288

Task*Technology (b

XY) -0.074 0.171

Technology

) 0.074 0.359

= 0.333, F = 1.399

Table 11: Polynomial Regression Results (User

Performance).

User Performance

Predictor

Beta (β)

Standard

Error

Constant (b

) 1.121*** 0.376

Task (b

X) -1.046 0.649

Technolo

(

)

0.925 0.577

Tas

(

)

0.475 0.322

Task*Technolo

(

)

-0.254 0.402

Technology

) 0.072 0.192

= 0.508, F = 2.896

Second, Response Surface Methodology (RSM)

(Edwards, 2002) was used to plot three-dimensional

(3D) surfaces relating Task and Technology to Use

and User Performance.

Regression beta (β) coefficients resulting from

equation (1) as presented in Tables 10 and 11 above,

HEALTHINF 2024 - 17th International Conference on Health Informatics

118

were used to estimate stationary points (X

, Y

principal axes (p

, p

11,

20,

21)

, and shapes along lines

of congruence and incongruence (a

, a

Surface values for prediction of Use and User

Performance are shown in Tables 12 and 13.

Table 12: Response Surface Analysis Results (Use).

Use

Stationary

Point

-0.572

(

-0.014

)

-3.941

(

-0.058

)

First Principal

Axis

Intercept (P

)

-4.570

(-0.064)

Slope (P

)

-1.099

(

-0.029

)

(-P

/(1+P

)

-46.138

(

-0.019

)

Second

Principal Axis

Intercept (P

)

-3.421

(-0.044)

Slope (P

)

0.910

(

0.030

)

Shape Along

Line of

Congruence (Y

= X)

Slope: a

)

0.326

(

0.639

)

Curvature: a

+ b

)

0.067

(0.350)

Shape Along

Line of

Incongruence

(Y = -X)

Slope: a

)

-0.756

(-0.463)

Curvature: a

(

+ b

)

0.215

(

0.097

)

Table 13: Response Surface Analysis Results (User

Performance).

User Performance

Stationary

Point

-1.167

(-0.003)

-8.481

-0.011

)

First Principal

Axis

Intercept (P

)

-8.818

(

-0.010

)

Slope (P

)

-0.289

(-0.018)

(-P

/(1+P

)

-46.138

(

-0.019

)

Second

Principal Axis

Intercept (P

)

-4.443

(

-0.014

)

Slope (P

)

3.462

(0.012)

Shape Along

Line of

Congruence (Y

= X)

Slope: a

)

-0.121

(-0.177)

Curvature: a

(

+ b

)

0.293

(

1.124

)

Shape Along

Line of

Incongruence

(Y = -X)

Slope: a

)

-1.971

(-0.993)

Curvature: a

+ b

)

0.801

(0.234)

The response for the Task (X) and Technology

(Y) predicting Use (Z) is shown in Figure 4.

Figure 4: Response Surface for Task-Technology Fit (TTF)

and Use.

The response surface for TTF effects on use was

saddle-shaped (stationary point: X

= -0.572, Y

= -

3.941). The first principal axis is not significantly

different [t = -0.029 (P

), t = -0.019 (-P

//P

+1)]

from the line of congruence (Y = X). Thus, a perfect

fit between the Task and Technology leads to

maximal use. The upward slope along the line of

congruence (Y = X) was negative but not significant.

The curvature along the line of congruence (Y = X)

was positive but not significant (a

= 0.293, t = 1.124),

indicating that the relationship between TTF and use

is linear. Therefore, the curvature along the line Y =

X does not significantly change for use. The

downward slope along the line of incongruence (Y =

-X) was negative but not significant (a

= -1.971, t =

-0.993). A lack of fit between the robotic surgery task

and support tools leads to a decrease in use. The

curvature along the line of incongruence (Y = -X) was

positive but not significant (a

= 0.801, t = 0.234),

further evidencing a linear association between TTF

and use.

The response for the Task (X) and Technology

(Y) predicting User Performance (Z) is shown in

Figure 5.

The first principal axis is not significantly

different [t = -0.018(p

), t = -0.019(=p

11+1

)] from

the line congruence (Y=X). Hence, a perfect fit

between the task and technology leads to maximised

user performance. The upward slope along the line of

congruence (Y=X) is negative and not significant (a

= -0.121, t = -0.177). The curvature along the line

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

119

Figure 5: Response Surface for Task-Technology Fit (TTF)

and User Performance.

of congruence (Y=X) was positive but not significant

= 0.293, t = 1.124), indicating that the relationship

between TTF and user performance is linear. This

indicates that the curvature along the line Y=X does

not significantly change for user performance. The

downward slope along the line of incongruence (Y=-

X) was negative but not significant (a

= -0.971, t = -

0.993). Hence, the lack of fit between the robotic

surgery task and support tools leads to a decrease in

user performance. The curvature along the line of

incongruence (Y=-X) was positive but not significant

= 0.801, t = 0.234), further indicating a linear

relationship between TTF and user performance. The

curvature along the line Y=-X did not, therefore,

change significantly for user performance.

The lateral shift (Atwater et al., 1998) in use and

user performance, in the surface along and

perpendicular to the line of congruence (Y = X) was

determined using the following equation:

– b

Lateral Shift = ——————————

2 (b

– b

+ b

)

where:

= The beta value for Task

= The beta value for Technology

= The beta value for Task

= The beta value for Task*Technology

= The beta value for Technolo

(2)

The lateral shift in use along the line of

congruence (Y = X) was positive (1.758), indicating

movement of approximately two units towards the

region where functional support levels surpass user

needs (Y > X). Here, the technology over-fits the task.

Hence, when the robotic surgery task and support tool

functions over-fit user needs, there is a sharp decline

in robotic surgeons’ dependence on use. Similarly,

the lateral shift in user performance along the line of

congruence (Y = X) was positive (1.230), indicating

movement of approximately one unit toward the

region where the robotic surgery task and support tool

functions over-fit user needs. Thus, when the robotic

surgery task and support tool functions over-fit user

needs, there is a sharp decline in the effectiveness,

efficiency, and quality, of robotic surgery task

performance.

8 DISCUSSIONS

In this paper, we investigated the potential transition

from technical system-oriented QoS to user-focused

QoE and QoT Internet configurations of the future.

We also explored the advent of an ultra-reliable and

ultra-low-latency B5G network and haptic-enabled

Internet. We applied this configuration to the use case

of remote robotic surgical task performance (tele-

haptic surgery applications) from the novel data-

driven evidence-based QoE/QoT perspective of

Task-Technology Fit (TTF) theory and predictive

modelling.

The analysis of non-linear impacts on use and

user performance represents a perspective of task-

technology equilibrium. This mechanism enables

more sophisticated and dynamic insights into the

effectiveness of TTF, and is useful for observing the

extent to which Information Technology (IT)

functions affect tool use and user performance levels.

Our findings show that when there is excessive

functional support for robotic surgery tasks, there is

an increasing likelihood of a lower dependence

among users, on using the technology whereby they

will more likely perceive that they deliver lower

quality MIS robotic surgery task performance, with

diminishing effectiveness and efficiency. This

finding represents an “IT surplus”, the supply of tool

functions that could exceed user task requirements

(Yang et al., 2013, p. 700). This is an extreme that

signifies a misfit, which can adversely affect task

productivity (Oh and Pinsonneault, 2007). Further, an

overfit can result in declining information

accessibility and processing performance, and has

been attributed to an excess of support functions that

can be termed as redundant (Jarvenpaa, 1989).

HEALTHINF 2024 - 17th International Conference on Health Informatics

120

9 IMPLICATIONS FOR

RESEARCH AND PRACTICE:

THEORETICAL AND APPLIED

CONTRIBUTIONS

From a more theoretical standpoint, an atomistic

approach (Yang et al., 2013), said to involve the

articulation and measurement of separate components

(p. 712) was used. This novel approach signifies a

more pragmatic, nuanced perspective of TTF

impacts. It can be applied to subsequent research

where the interaction effects of TTF warrant further

investigation. Moreover, the detailed analysis of use

and user performance effect differentials modelled

using three-dimensional (3-D) surfaces represents

richer insights into testing non-linear TTF.

From a more practical standpoint, the findings of

this study can serve as key guidelines with which to

enhance or reduce functional support related to

robotic surgery support tool use and surgeon user

performance. This can be an important benchmark

with which robotic surgery support tool designers can

calibrate the responsiveness of functional support to

user task needs. Further, the findings indicate that

excess or inadequate functional robotic surgery tool

support for surgeons’ user needs can lead to adverse

use and user performance impacts. Hence, robotic

surgery support tool designers must be acutely aware

of these task-technology differentials to attain a state

of congruence between supporting functions and

robotic surgeon’s needs.

10 CONCLUSIONS

In light of recent developments in ultra-reliable and

ultra-low latency communications that will come to

define next-generation digital networks, we

conceptualised the emerging transition from QoS-

centric content-delivery networks to QoE and QoT

focused skillset-delivery network configurations that

will typify closed-loop control architectures for

haptic-enabled and B5G Internets. We offer the novel

task-technology fit (TTF) conceptualisation and

predictive modelling and empirical analysis

perspective as a diagnostic tool. This vision of an

Internet of the future will involve the task performer

in a domain-specific technology user-focused context

(remote setting) performing tasks as the human-in-

the-loop (HITL), through immersive real-time

human-to-machine/robot (H2M/R) interactions.

Through this preliminary study, we examine the

mission-critical user scenario of tele-haptic (remote)

robotic surgery, expected to become a reality in the

era of B5G. With a haptic-enabled Internet and B5G

network to augment user skills, future robotic

microsurgeons will be ably supported to perform

seamless tele-haptic (remote) surgical tasks.

ACKNOWLEDGEMENTS

We sincerely thank the robotic surgeons who

participated in the study in conjunction with the

Robotic Assisted Microsurgical and Endoscopic

Society (RAMSES).

REFERENCES

Aiken, L. S., & West, S. G. (1991). Multiple Regression:

Testing and Interpreting Interactions. Sage

Publications.

Atwater, L. E., Ostroff, C., Yammarino, F. J., & Fleenor, J.

W. (1998). Self-Other Agreement: Does it Really

Matter?. Personnel Psychology, 51(3), 577- 598.

Bassoli, R., Boche, H., Deppe, C., Ferrara, R., Fitzek, F. H.

P., Janssen, G., & Saeedinaeeni, S. (2021). Quantum

Communication Networks. Springer.

Dishaw, M. T., & Strong, D. M. (1998). Supporting

Software Maintenance with Software Engineering

Tools: A Computed Task-Technology Fit Analysis. The

Journal of Systems and Software, 44(2), 107-120.

Dishaw, M. T., Strong, D. M., & Bandy, D. B. (2002).

Extending the Task-Technology Fit Model with Self-

Efficacy Constructs. Proceedings of the 8th Americas

Conference on Information Systems (AMCIS), August

9-11, 2002, Dallas, USA.

Dohler, M. (2018) The Future and Challenges of

Communications – Toward a World Where 5G Enables

Synchronised Reality and Internet of Skills. Internet

Technology Letters, 1(2), 1-3.

Edwards, J. R. (1993). Problems with the Use of Profile

Similarity Indices in the Study of Congruence in

Organizational Research. Personnel Psychology, 46(3),

641-665.

Edwards, J. R. (1994). The Study of Congruence in

Organizational Behaviour Research: Critique and a

Proposal and a Proposed Alternative. Organizational

Behaviour and Human Decision Processes, 58(1), 51-

100.

Edwards, J. R. (2002). Alternatives to difference scores:

Polynomial regression analysis and response surface

methodology. In F. Drasgow & N. W. Schmitt (Eds.),

Measuring and Analyzing Behavior in Organizations:

Advances in Measurement and Data Analysis (pp. 350-

400). Wiley.

Gatara, M. C., Mzyece, M., & Parekattil. S. J. (2021).

Towards a Fifth (5th) Generation (5G) Network and

Haptic-Enabled Internet for Remote Robotic Surgery:

The Beyond 5G (B5G) Era of Next-Generation Digital Networks: Preliminary Study of a Task-Technology Fit (TTF) Model for Remote

Robotic Surgery Applications

121

A Task-Technology Fit (TTF) Model. Proceedings of

the 45th Meeting of the Wireless World Research

Forum (WWRF), January 18-21, 2021, Kuala Lumpur,

Malaysia.

Gatara, M. C., & Mzyece, M. (2023). From QoS to QoE

plus QoT in Beyond 5G Networks. Journal of Mobile

Multimedia, 19(1), 135-146.

Giordani, M., Polese, M., Mezzavilla, M., Rangan, S., &

Zorzi, M. (2020). Toward 6G Networks: Use Cases and

Technologies. IEEE Communications Magazine, 58(3),

55-61.

Goodhue, D. L. (1992). User Evaluations of MIS Success:

What are we Really Measuring?. Proceedings of the

25th Hawaii International Conference on Systems

Sciences (HICSS), January 7-10, 1992, Hawaii, USA.

Goodhue, D. L. (1995). Understanding User Evaluations of

Information Systems. Management Science, 41(12),

1827-1844.

Hair, J. F., Jr., Hult, G. T. M., Ringle, C. M., & Sarstedt, M.

(2021). A primer on partial least squares structural

equation modeling (PLS-SEM) (3rd ed.). Sage

Publications.

Hiltz, S. R., & Johnson, K. (1990). User Satisfaction with

Computer-Mediated Communication Systems.

Management Science, 36(6), 739-764.

Hou, C.-K. (2012). Examining the Effects of User

Satisfaction on System Usage and Individual

Performance with Business Intelligence Systems: An

Empirical Study of Taiwan’s Electronics Industry.

International Journal of Information Management,

32(6), 560-573.

Igbaria, M., Zinatelli, N., Cragg, P., & Cavaye, A. L. M.

(1997). Personal Computing Acceptance in Small

Firms: A Structural Equation Model. MIS Quarterly,

21(3), 279-305.

Jarvenpaa, S. L. (1989). The Effect of Task Demands and

Graphical Format on Information Processing

Strategies. Management Science, 35(3), 285-303.

Junglas, I., Abraham, C., & Ives, B. (2009). Mobile

Technology at the Frontlines of Patient Care:

Understanding Fit and Human Drives in Utilization

Decisions and Performance. Decision Support Systems,

46(3), 634-647.

Nance, W. D. (1992). Task/Technology Fit and Knowledge

Worker Use of Information Technology: A Study of

Auditors. Ph.D. Dissertation, University of Minnesota,

Minnesota, USA.

Nawaz, S. J., Sharma, S. K., Wyne, S., Patwary, M. N., &

Asaduzzaman, M. (2019). Quantum Machine Learning

for 6G Communication Networks: State-of-the-Art and

Vision for the Future. IEEE Access, 7, 46317-46350.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric

Theory (3rd ed.). McGraw-Hill.

Oh, W. and Pinsonneault, A. (2007). On the Assessment of

the Strategic Value of Information Systems:

Conceptual and Analytical Approaches. MIS Quarterly,

31(2), 239-265.

Pierucci, L. (2015). The Quality of Experience Perspective

Toward 5G Technology. IEEE Wireless

Communications, 22(4), 10-16.

Saracino, A., Deguet, A., Staderini, F., Boushaki, M. N.,

Cianchi, F., Menciassi, A., & Sinibaldi, E. (2019).

Haptic Feedback in the da Vinci Research Kit (dVRK):

A User Study Based on Grasping, Palpation, and

Incision Tasks. The International Journal of Medical

Robotics and Computer Assisted Surgery, 15(4), 1-13.

Shapiro, C., & Varian, H. R. (1999). Information rules: A

strategic guide to the network economy. Harvard

Business School Press.

Thompson, R. L., Higgins, C. A., & Howell, J. M. (1991).

Personal Computing: Toward a Conceptual Model of

Utilization. MIS Quarterly, 15(1), 125-143.

Torkzadeh, G., & Doll, W. J. (1999). The Development of

a Tool for Measuring the Perceived Impact of

Information Technology on Work. Omega, The

International Journal of Management Science, 27(3),

327-339.

Vessey, I. (1991). Cognitive Fit: A Theory-Based Analysis

of the Graphs Versus Tables Literature. Decision

Sciences, 22(2), 219-240.

Vessey, I. (1994). The Effect of Information Presentation

on Decision Making: A Cost-Benefit Analysis.

Information and Management, 27(2), 103-119.

Vessey, I., & Galletta, D. (1991). Cognitive Fit: An

Empirical Study of Information Acquisition.

Information Systems Research, 2(1), 63-84.

Yang, H.-D., Kang, S., Oh, W., & Kim, M. S. (2013). Are

All Fits Created Equal? A Nonlinear Perspective on

Task-Technology Fit. Journal of the Association for

Information Systems, 14(12), 694-721.

HEALTHINF 2024 - 17th International Conference on Health Informatics

122