VR-SFT Reproducing Swinging Flashlight Test in Virtual Reality to Detect Relative Aﬀerent Pupillary Defect

2025-04-26 0 0 510.06KB 12 页 10玖币

侵权投诉

VR-SFT: Reproducing Swinging Flashlight Test

in Virtual Reality to Detect Relative Aﬀerent

Pupillary Defect

Prithul Sarker, Nasif Zaman, and Alireza Tavakkoli

Department of Computer Science and Engineering, University of Nevada, Reno,

United States

prithulsarker@nevada.unr.edu

Abstract. The relative aﬀerent asymmetry between two eyes can be

diagnosed using swinging ﬂashlight test, also known as the alternating

light test. This remains one of the most used clinical tests to this day. De-

spite the swinging ﬂashlight test’s straightforward approach, a number of

factors can add variability into the clinical methodology and reduce the

measurement’s validity and reliability. This includes small and poorly

responsive pupils, dark iris, anisocoria, uneven illumination in both eyes.

Due to these limitations, the true condition of relative aﬀerent asymme-

try may create confusion and various observers may quantify the rela-

tive aﬀerent pupillary defect diﬀerently. Consequently, the results of the

swinging ﬂashlight test are subjective and ambiguous. In order to elimi-

nate the limitations of traditional swinging ﬂashlight test and introduce

objectivity, we propose a novel approach to the swinging ﬂashlight exam,

VR-SFT, by making use of virtual reality (VR). We suggest that the clin-

ical records of the subjects and the results of VR-SFT are comparable.

In this paper, we describe how we exploit the features of immersive VR

experience to create a reliable and objective swinging ﬂashlight test.

Keywords: Virtual Reality ·HTC Vive Pro ·FOVE 0 ·Swinging Flash-

light Test ·RAPD.

1 Introduction

Relative aﬀerent pupillary defect (RAPD) is a particular eye condition where

pupils constrict asymmetrically in response to light stimuli shone in each eye at

a time. This condition is also known as Marcus Gunn Pupil. RAPD is caused

by a unilateral or asymmetrical retinal or optic nerve disorder. RAPD is not a

disease itself, but the defect occurs due to lesions in the aﬀerent pathway which

is located behind the pupils. RAPD indicates either of the following: lesion of the

optic nerve, optic chiasm, glaucoma, retinal detachment, macular degeneration,

dense cataract, amblyopia etc. Even though RAPD is not life threatening in

all cases, this usually reveals disease in the pre chiasmal visual pathway [1].

A strong correlation is found between RAPD and multiple diseases, such as,

amblyopia [2], glaucoma [3], [4], retinal detachment [5], optic neuritis [6], [7]. A

arXiv:2210.06474v1 [cs.HC] 12 Oct 2022

2 Sarker et al.

study by Wilhelm et al. shows that they noticed asymmetries in the connections

between visual pathways and midbrain in about 2% of normal subjects [8].

The most common eye exam performed to detect RAPD is the swinging

ﬂashlight test (SFT) [1]. In normal eyes, the pupils constrict and dilate similarly

in response to the light stimuli. If the pupils do not respond identically, they are

diagnosed with RAPD. Neutral density ﬁlters (NDFs), cross-polarized ﬁlters, and

subjective grading based on the degree of initial constriction and subsequent re-

dilation of each pupil as the light is swung are diﬀerent ways for quantifying or

assessing RAPDs [9][10][11]. These approaches have been proven to be useful and

precise. However, human involvement makes these methods highly subjective; so

the validity of the measurements is questionable. There have been approaches to

reduce human dependencies when it comes to measuring RAPD. Most of these

strategies tackle this problem using pupillometer [12][4].

Virtual reality (VR) is a programmed environment where the objects and

images are made realistic-looking which gives the user the perception of reality,

and is one of the most promising technologies in terms of future development. In

this paper, we introduce a novel method to detect RAPD using VR and discuss

in detail about the methodology of our implementation and results. By using

virtual reality, we eliminate subjectivity of SFT, and control the illumination on

each eye precisely to quantify asymmetrical optic nerve disorder.

2 Literature Review

The variation in pupillary reactivity under bright and low light environments

was originally observed by Marcus Gunn in 1904 [13]. The disparity in pupil-

lary responsiveness at that time was regarded as peculiar. Later, Paul Levatin

in 1959 [13] demonstrated that the pupillary response had a speciﬁc cause. The

demonstration was made under the assumption that pupillary impulses or pupil-

lary escape caused by visible light are linearly proportionate to the visual stimuli

given. The authors introduced a novel test to identify pupillary escape under

various lighting conditions and optic nerve disease. This method was named

swinging ﬂashlight test (SFT). However, this approach alone was insuﬃcient to

determine the degree of the disease of the optic nerve [13].

Keeping the deﬁcits of traditional SFT in mind, Thompson et al. introduced

a more reliable technique to measure the relative deﬁcit of the eyes with neutral

density ﬁlters (NDF) [9]. The ﬁlters dim the light from the ﬂashlight. Based on

the reduced light impulse to each eye using the NDF, the authors proposed to

quantify the relative aﬀerent defect by the log units of the neutral density ﬁlters.

The swinging ﬂashlight test is repeated by placing diﬀerent values of NDF in

front of the relatively normal eye until the pupillary reaction is found to same. In

1993, Bell et al. proposed to quantify RAPD by critically observing the pupillary

reaction under diﬀerent lighting conditions, and identiﬁed the period of time for

constriction and dilation of the pupil [11]. The results of this procedure are highly

subjective because no measurement tool was utilized. This technique is still the

most used one for ﬁguring out RAPD [11].

VR-SFT: Reproducing Swinging Flashlight Test in Virtual Reality 3

Kawasaki et al. in 1995 used a computerized binocular infrared video pupil-

lometer to record pupillary reaction and experimented with various duration and

intensity of light illumination on each eyes [12]. The authors found out strong

linear relationship between diﬀerence in contraction amplitude or pupil diame-

ter (y-axis) and the diﬀerence of light illumination for each eye (x-axis). They

performed linear regression to locate the point where the line intersects with the

x-axis containing the illumination levels. The intersection point is the relative af-

ferent pupillary defect score, also known as RAPD score. This intersection point

is considered because equal and balanced pupillary responses to stimulation of

the right and left eyes are produced by the log-unit attenuation.

To make the procedure objective, RAPDx device is introduced [4]. The de-

vice is intended for recording and analyzing pupil responses to various stimuli

making quantiﬁcation of RAPD easy. After the recognition of RAPDx device,

accurate evaluation of relationship between RAPD and other diseases has been

possible. There has been research on patients with optic nerve disease to evalu-

ate progression of RAPD [14]. Satou et al. continued this research and concludes

that an absolute RAPD score of 0.2 or less can be considered as healhy whereas

an absolute RAPD score of 0.5 log units or over means RAPD is present [15].

With the success of deep learning, Temel et al. proposed a transfer learning

based approach to capture pupil diameter from video dataset [16]. For the ex-

periment and data collection, the authors developed a headset that can perform

automated SFT. From the video, they used a pupil localization algorithm to

locate the pupil and measure the pupil diameter, and ﬁnally compared left and

right pupil diameter to detect RAPD.

Recent improvements in virtual, augmented, and mixed reality (VAMR) have

led to widespread acceptance and commercial success which reﬂect in vision

assessment practices. Perceptual Modeling in Virtual Reality (PMVR) [17] and

Visual Acuity in Virtual Reality (VAVR) [18] conduct their experiments using

commercially accessible VAMR technology. In this paper, we propose a novel

objective method to replicate swinging ﬂashlight test using virtual reality for

accurate pupil measurement and detect RAPD. No comparable VR research has

been done, as far as we know. We propose and validate our approach on two

diﬀerent brands of VR headset.

3 Methodology

In traditional swinging ﬂashlight test, the participant and the examiner are

seated side by side in a dimmed room, and the participant is asked to look

at a point at a far distance to remove the eﬀects of accommodative response.

When the examiner thinks that the participant’s pupils are completely dilated,

a ﬂashlight is directed towards one of the eyes of the participant. After the ﬂash-

light is turned on, the pupils of a normal participant is constricted. Within few

seconds of the illumination, the pupils become adjusted to the light illumination.

Then the ﬂashlight is swiftly moved to the other eye to observe the response of

the pupils. This process is repeated multiple times, and the examiner takes note

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：

VR-SFT:ReproducingSwingingFlashlightTestinVirtualRealitytoDetectRelativeAerentPupillaryDefectPrithulSarker,NasifZaman,andAlirezaTavakkoliDepartmentofComputerScienceandEngineering,UniversityofNevada,Reno,UnitedStatesprithulsarker@nevada.unr.eduAbstract.Therelativeaerentasymmetrybetweentwoeyescanbed...

展开>> 收起<<

VR-SFT Reproducing Swinging Flashlight Test in Virtual Reality to Detect Relative Aﬀerent Pupillary Defect.pdf

共12页,预览3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！

VR-SFT Reproducing Swinging Flashlight Test in Virtual Reality to Detect Relative Aﬀerent Pupillary Defect

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: