Auxilio and Beyond Comparative Evaluation Usability and Design Guidelines for Head Movement-based Assistive Mouse Controllers MOHAMMAD RIDWAN KABIR MOHAMMAD ISHRAK ABEDIN RIZVI AHMED SAAD BIN

2025-04-27 0 0 3.72MB 29 页 10玖币

侵权投诉

Auxilio and Beyond: Comparative Evaluation, Usability, and Design Guidelines

for Head Movement-based Assistive Mouse Controllers

MOHAMMAD RIDWAN KABIR

∗

,MOHAMMAD ISHRAK ABEDIN

∗

,RIZVI AHMED,SAAD BIN

ASHRAF,HASAN MAHMUD, and MD. KAMRUL HASAN,Department of Computer Science and Engi-

neering (CSE), Islamic University of Technology (IUT), Bangladesh

Fig. 1. Auxilio provides accessibility to computer interaction for the upper limb disabled community, facilitating mouse cursor

movements and click actuation, solely based on head rotation and cheek muscle twitches.

Upper limb disability due to neurological disorders or other factors restricts computer interaction for aected individuals using a

generic optical mouse. This work reports the ndings of a comparative evaluation of Auxilio, a sensor-based wireless head-mounted

Assistive Mouse Controller (AMC), that facilitates computer interaction for such individuals. Combining commercially available,

low-cost motion and infrared sensors, Auxilio utilizes head movements and cheek muscle twitches for mouse control. Its performance

in pointing tasks with subjects without motor impairments has been juxtaposed against a commercially available and patented

∗Both authors contributed equally to this research.

Authors’ address: Mohammad Ridwan Kabir, ridwankabir@iut-dhaka.edu; Mohammad Ishrak Abedin, ishrakabedin@iut-dhaka.edu; Rizvi Ahmed,

rizviahmed@iut-dhaka.edu; Saad Bin Ashraf, saadashraf@iut-dhaka.edu; Hasan Mahmud, hasan@iut-dhaka.edu; Md. Kamrul Hasan, hasank@iut-

dhaka.edu, Department of Computer Science and Engineering (CSE), Islamic University of Technology (IUT), Board Bazar, Gazipur, Bangladesh, 1704.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not

made or distributed for prot or commercial advantage and that copies bear this notice and the full citation on the rst page. Copyrights for components

of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to

redistribute to lists, requires prior specic permission and/or a fee. Request permissions from permissions@acm.org.

Manuscript submitted to ACM

Manuscript submitted to ACM 1

arXiv:2210.04483v2 [cs.HC] 7 Oct 2024

2 Ridwan, et al.

vision-based head-tracking AMC developed for similar stakeholders. Furthermore, our study evaluates the usability of Auxilio using

the System Usability Scale, supplemented by a qualitative analysis of participant interview transcripts to identify the strengths and

weaknesses of both AMCs. Experimental results demonstrate the feasibility and eectiveness of Auxilio, and we summarize our key

ndings into design guidelines for the development of similar future AMCs.

CCS Concepts: •Human-centered computing

→

Pointing devices;Accessibility technologies;Accessibility technologies;

Pointing;Usability testing;•General and reference →Experimentation;Evaluation;Performance.

Additional Key Words and Phrases: assistive technology, assistive mouse controller, upper limb disability, wearable sensors, pointing

device

ACM Reference Format:

Mohammad Ridwan Kabir, Mohammad Ishrak Abedin, Rizvi Ahmed, Saad Bin Ashraf, Hasan Mahmud, and Md. Kamrul Hasan. 2024.

Auxilio and Beyond: Comparative Evaluation, Usability, and Design Guidelines for Head Movement-based Assistive Mouse Controllers.

1, 1 (October 2024), 29 pages. https://doi.org/10.1145/nnnnnnn.nnnnnnn

1 INTRODUCTION

Upper limb disability refers to the complete or partial loss of motor capability of the upper limb, potentially caused due

to — stroke [

], spinal injury [

], cerebral palsy [

], Amyotrophic Lateral Sclerosis (ALS)

[

], deformation of limbs at birth [

], amputation [

], etc., signicantly reducing the utilization of

the upper limbs in various motor tasks. The under-utilized residual capabilities [

] of the disabled upper limb might

hamper the lives of those aected in terms of both activity limitations and participation restrictions [

]. For example, in

patients with ALS, disability is characterized by motor dysfunctions while the brain and eye functionalities remain

preserved [

]. However, the same cannot be said for the motor capabilities of upper limb amputees [

], as

they are limited by their amputated body part(s). Research has shown that, unlike a normal person, the residual sensory

abilities of disabled individuals intensify over time, compensating for their lost ability [

]. Eventually,

they learn to utilize these abilities to accomplish dierent tasks in their daily lives [

]. Although physically

capable individuals can seamlessly use generic handheld pointing devices (e.g. an optical mouse) for interacting with a

computer, people with upper limb disability require Assistive Mouse Controllers (AMC) as an alternative input modality

for the same [

]. However, the usability and performance of such systems must meet a minimum standard to be

suitable for practical, everyday applications.

The ISO 9241-11 states the usability of any system or device as, “the extent to which a product can be used by

specied users to achieve specied goals with eectiveness, eciency, and satisfaction in a specied context of use”

[

]. Researchers have analyzed the feasibility and usability of dierent gestures and sensor technologies for developing

Vision-based [

], Electromyography (EMG)-based [

], Electrooculogram (EOG)-based

[

], and Wearables Sensor-based AMCs [

] to make computers accessible to individuals with

upper limb disability. However, a particular AMC technology may be convenient for people with a certain disability only.

For example, eye movement is one of the residual motor capabilities of patients with ALS [

]. Intuitively,

vision-based, EMG-based, or EOG-based AMCs may be more appropriate for them compared to those exploiting head

motion sensing technologies [

]. While the existing AMCs try to cater to dierent needs, they face limitations such as

but not limited to — sensitivity to lighting conditions, frequent calibration needs, noise interference, hygiene concerns,

and unreliable performance in uctuating environments, highlighting the need for more user-friendly and robust

alternatives. Hence, the design, development, and evaluation of alternative input modalities to facilitate computer

interaction for individuals with upper limb disability is still of interest to the research community [

]. Along with that,

Manuscript submitted to ACM

Auxilio and Beyond: Comparative Evaluation, Usability, and Design Guidelines for Head Movement-based Assistive

Mouse Controllers 3

the importance of analyzing the comparative performance, usability, and feedback of such an AMC from the users’

point of view cannot be overemphasized [45].

Throughout this study, we conducted a comparative analysis of the performance of Auxilio [

], a prototype of

a sensor-based head-mounted wireless AMC for people with upper limb disability that tries to address some of the

aforementioned limitations and shortcomings, against Smyle Mouse [

], a commercially available and patented

camera-based AMC in similar pointing tasks. Auxilio combines low-cost Commercial O-The-Shelf (COTS) Inertial

Measurement Unit (IMU) for controlling mouse cursor with absolute head movements and infrared sensors for actuating

mouse clicks with cheek muscle twitches. On the other hand, Smyle Mouse uses a camera to track head movements to

control a mouse cursor and detect a user’s smile to register mouse clicks. Smyle Mouse oers a separate UI that allows

the users to choose the type of event to be registered (left-click, right-click, double-click, drag, etc.) with a smile. We

acknowledge that comparing Auxilio against another sensor-based head-movement-controlled AMC would have been

ideal. However, despite our best eorts, we were unable to procure any such AMC in our locality. Hence, we compared

it against the Smyle Mouse, which is similar from the perspective of input modality and is a recognized commercially

available, and patented system.

Using these AMCs, a within-subject Point and Click experiment involving 10 participants without any motor

impairments was conducted featuring a balloon-popping game, Popper [

]. We believe that since Auxilio is fairly new

to the domain of AMC technologies, its performance, usability, and design limitations need to be analyzed before it

can be made available to its stakeholders. As evaluation involving individuals with upper limb disability is resource

intensive and might have restrictions from legal perspectives, we decided to carry out our initial investigation involving

participants without any motor impairment. Furthermore, we investigated the usability of Auxilio in comparison with

Smyle Mouse, leveraging the System Usability Scale (SUS) [

]. We followed it up with a qualitative

analysis of the interview transcripts of the participants to nd strengths, weaknesses, and future directives for similar

AMCs. To the best of our knowledge, this work is the rst of its kind in the literature.

In summary, the main contributions of this study in the domain of accessibility to computer interaction for the

physically disabled community are as follows:

(a)

We evaluated the feasibility and performance of Auxilio in comparison to the Smyle Mouse, a patented, camera-based

head-tracking AMC, through controlled pointing tasks.

(b)

We conducted a comparative usability analysis of both AMC technologies using the System Usability Scale (SUS) to

highlight the usability challenges and advantages of each of them.

(c)

We carried out a qualitative analysis of user feedback to identify key strengths and limitations of the two AMC

types, providing recommendations and distilling design guidelines for similar AMC development.

2 RELATED WORKS

The recent technological advancements have shaped the design, and development of Assistive Mouse Controllers

(AMCs) for physically challenged individuals into a prominent research area [

]. Interaction data from such individuals

may be recorded either using computer vision, Electromyography (EMG), Electrooculogram (EOG), or wearable sensors.

In this section, we elaborate on the existing state-of-the-art AMCs and justify our scope and motivation behind this

study.

Manuscript submitted to ACM

4 Ridwan, et al.

2.1 Vision-based AMCs

Existing vision-based AMCs in the literature leverage facial or eye gaze features, collected from real-time video feeds

using an eye tracker, webcam, or other imaging sensors, and map eye gaze to screen coordinates for cursor control.

However, the user’s eye gaze needs to be calibrated before use. For mouse click actuation, dwell-time-based mechanisms

or gestures such as — eye wink,blink, or smile are among the common ones. Among the many studied works, researchers

have developed Smyle Mouse [

] that uses a generic webcam to register users’ head movement through nose

tracking for cursor movement and smile gestures for registering mouse click events. The system oers a calibration

phase where a user has to perform a series of gestures as instructed via a UI before actual usage. By default, the

smile gesture actuates a left mouse click. However, a user can customize the event (left-click, right-click, double-click,

drag, etc.) to be triggered with the gesture from a separate UI. Apart from the smile gesture, the system also oers

dwell-time-based click mechanism. Zhang et al. [

] have developed a software-based AMC leveraging eye gaze tracking

with an eye tracker to control the mouse cursor movement and dwell-time-based clicking method via a virtual UI for

Mouse/Keyboard simulation. The authors have evaluated their works through two experiments, a searching task, and

a web browsing task, utilizing Technology Acceptance Model (TAM) and System Usability Scale (SUS). Apart from

eye gaze tracking, researchers have also leveraged nose tracking for cursor control [

]. As opposed to eye trackers

or webcams, optical mouse or imaging sensors have also been used to track eye gaze for cursor movement [

However, a potential drawback of optical mouse sensors is the requirement of an additional light source to work

properly, as demonstrated by [

]. Concerning the methods used in the aforementioned works, virtual interface-based

methods for making computers accessible to the physically disabled are popular in the literature [

]. Although

common, in practice dwell-time-based click actuation suers from unwanted actuation of mouse clicks due to eye gaze

xation, generally known as the Midas Touch problem [

]. To address this issue, researchers [

] proposed a muscle or

eyebrow shrugging-based click actuation technique, implemented through the software packages Camera Mouse [

]

and ClickerAid [

]. Furthermore, Rajanna et al. [

] have also developed a system for people with arm or hand

impairment that uses eye gaze for pointing at a screen element while selection is actuated by exerting pressure on a

pressure sensor-based footwear.

A critical requirement for vision-based AMCs to work properly is to ensure proper lighting conditions for calibration

and accurate detection of facial [

] or eye gaze [

] features. For eye gaze-based AMCs, gaze tracking

may be challenging due to image resolution, dierent lighting conditions, the user’s dependency on eyeglasses due to

poor eyesight, and even the user’s skin complexion. In addition to that, human eyes are not the most accurate pointing

device [

]. Moreover, a human can only gaze at a single point at a given time, preventing the user from looking at

another region of interest without moving the mouse cursor. Most importantly, since eyes are used for both cursor

movement and click actuation, users might nd it dicult to execute both actions simultaneously when required, for

example, dragging an item. As stated earlier, a particular problem with dwell-time-based click actuation is the Midas

Touch problem [

], resulting in unwanted selection of UI elements. Another disadvantage of vision-based AMCs is

that existing eye trackers and webcams can not detect and track a user’s eye gaze beyond a particular distance from the

PC or workstation, forcing the user to maintain a particular distance from the PC or workstation.

2.2 Electromyography (EMG)-based AMCs

Electromyography (EMG) signals refer to the measurement of very low electric potentials generated due to muscle

contractions with electrodes placed noninvasively on the skin, where the signal amplitudes are proportional to the

Manuscript submitted to ACM

Auxilio and Beyond: Comparative Evaluation, Usability, and Design Guidelines for Head Movement-based Assistive

Mouse Controllers 5

exerted muscle force [

]. For people with upper limb disabilities, EMG signals can be retrieved from the contraction of

the residual muscles [

], which can then be used to determine the type of intended motion. Researchers have explored

the feasibility and performance of a myoelectric cursor control for amputees with the help of a myoelectric armband

[

]. However, before the device could be used for mouse cursor control, the users had to go through a training phase

for device calibration and preparation for the subsequent test phase, aided by a guided UI-based calibration method.

For making computers accessible to people with high-level spinal cord injury, researchers in [

] have proposed two

cursor control methods, auto rotate and manual rotate, using a single-site surface EMG sensor. However, the proposed

method requires the usage of disposable Ag/AgCl center snap electrodes, which makes it a hindrance for daily tasks.

The aforementioned work was evaluated through a pointing task-based experiment utilizing Fitts’ law.

EMG signals are susceptible to external noises and are highly dependent on the exact and accurate placement of

electrodes [

] for accurate gesture recognition. One of the inherent problems of EMG technology is that the retrieved

signals are typically weaker and vary from person to person [

], thereby, requiring a user-specic device calibration

and gesture recognition, every time a user intends to use it as an AMC. Although deep learning techniques [

] have

been proposed for mitigating the need for user-specic device calibration and gesture recognition, the computational

expense for fullling the simple objective of an AMC seems unreasonable.

2.3 Electrooculogram (EOG)-based AMCs

An Electrooculogram (EOG) is used to measure the corneal-retinal Transepithelial Potential Dierence (TEPD) with

the help of noninvasive electrodes placed around the human eyes. TEPD is produced due to horizontal and vertical

movements of an eye [

], which can be utilized to implement mouse control and clicking mechanisms. However,

TEPD is likely to uctuate in dierent lighting conditions, making light adaptation and training an integral part of

EOG-based AMCs. EOG-based AMCs that leverage eye movements to extract relative gaze position on the screen have

been proposed [

], facilitating interaction with a computer for people suering from neurodegenerative

disorders. To facilitate typing through recognition of eye movement patterns, real-time EOG-based systems have been

developed [

] for people with ALS, which is a neurodegenerative disorder that does not aect the

brain functions or the eye movements. Such nature of the disease makes EOG-based systems best suited for people

suering from it [9,19].

Although EOG is a promising low-cost AMC technology that is still being researched, it is limited by its low spatial

resolutions, as it is dicult to estimate the absolute gaze position due to noise from nearby sources of bio-potentials

[

]. Apart from the high pre-processing complexities of EOG signals [

], their characteristics vary due to the

variation in the number, the type (dry or wet), the material, and the placement of electrodes [

Finally, from an ergonomic perspective, it is intuitive that continuous eye movements for controlling a mouse may pose

certain health issues, thereby, aecting the user’s performance and comfort.

2.4 Sensor-based Wearable AMCs

Apart from Vision, EOG, and EMG–based AMCs, sensor-based wearable AMC technologies have also been developed

for assisting people with upper limb disabilities, leveraging their residual motor functionalities as alternative input

modalities. Among these alternatives, head movement is a natural, eective, and the most common modality for moving

a cursor [

]. Other alternatives include but are not limited to tongue muscle movement [

], and Brain

Computer Interfaces (BCI) [57].

Manuscript submitted to ACM

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

AuxilioandBeyond:ComparativeEvaluation,Usability,andDesignGuidelinesforHeadMovement-basedAssistiveMouseControllersMOHAMMADRIDWANKABIR∗,MOHAMMADISHRAKABEDIN∗,RIZVIAHMED,SAADBINASHRAF,HASANMAHMUD,andMD.KAMRULHASAN,DepartmentofComputerScienceandEngi-neering(CSE),IslamicUniversityofTechnology(IUT),Bangl...

展开>> 收起<<

Auxilio and Beyond Comparative Evaluation Usability and Design Guidelines for Head Movement-based Assistive Mouse Controllers MOHAMMAD RIDWAN KABIR MOHAMMAD ISHRAK ABEDIN RIZVI AHMED SAAD BIN.pdf

共29页,预览5页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Auxilio and Beyond Comparative Evaluation Usability and Design Guidelines for Head Movement-based Assistive Mouse Controllers MOHAMMAD RIDWAN KABIR MOHAMMAD ISHRAK ABEDIN RIZVI AHMED SAAD BIN

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: