Papers by Elizabeth Irving
Ophthalmic and Physiological Optics, Feb 26, 2007
Aims: To examine adaptation to progressive addition lenses (PALs) by investigating eye and head m... more Aims: To examine adaptation to progressive addition lenses (PALs) by investigating eye and head movement strategies of individuals naïve to PAL wear; to determine whether adaptation is a shortterm change in strategy or is maintained with continued PAL wear; to determine if adaptation strategies are different between PAL designs. Methods: Ten presbyopic individuals participated in a double-blind crossover study. Participants wore one PAL design for £4 weeks, had approximately 1 week without PAL wear and then wore the second PAL for £4 weeks. Eye and head movements were recorded at the beginning and end of each period of PAL wear in response to several visual tasks. Visual tasks were a flash discrimination task at 2 m and 40 cm, and reading text aloud. Quantitative analysis was undertaken for eye and head movement dynamics (latency, duration, peak velocity, etc.). Categorical analysis used eye and head movement metrics to classify participants as eye-movers, head-movers or mixed at each visit. Results: There was significant between-participant variability. A number of eye/head movement parameters exhibited carry-over effects. The number of vertical head movements for the distance task showed a significant increase between the first and second visit (F 1,9 ¼ 5.578; p ¼ 0.042), irrespective of lens design. Categorical analysis showed that participants employed task dependent strategies for all tasks. Participants that altered their strategy tended to include more head movements. Seven of the 10 participants preferred the second PAL worn. Conclusions: Participants recruited more head movements for flash discrimination tasks and for reading text during adaptation to PAL wear. The many and various eye and head movement parameters analysed were unable to discriminate differences dependent on PAL design.
Ophthalmic and Physiological Optics, May 1, 2010
The purpose of this study was to determine the reliability of the prism neutralized alternate cov... more The purpose of this study was to determine the reliability of the prism neutralized alternate cover test (ACT) used to measure near phoria in non-strabismic participants as measured by experienced and novice examiners and to compare the values to results obtained by eye tracking. Methods: Two experienced optometrists and two optometry students performed the ACT at 40 cm on 50 non-strabismic participants. The ACT was then repeated with the subjects wearing an eye tracker. The saccade amplitude of the eye movement recordings was used to determine the phoria. The reliability was determined using Bland-Altman analysis to determine the mean of the differences in the groups and the 95% limits of agreement. Results: The phoria range determined by the eye tracker was between 6 pd eso and 19 pd exo. The mean of the difference in values and 95% limits of agreement between the experienced examiners was 0.7 ± 3.0 pd; between the novice examiners was)0.1 ± 4.4 pd; between the experienced and novice examiners was 0.8 ± 3.5 pd; between the eye tracker and the experienced examiners was 0.6 ± 3.5 pd; between the eye tracker and novice examiners was 1.4 ± 3.6 pd. Conclusions: The mean differences of the ACT result between experienced examiners, novice examiners, and the eye tracker, were not clinically meaningful. However, the 95% limits of agreement were quite high (±3.0-4.4 pd). This information is useful for clinical decision making, student education and student evaluation.
ACM transactions on computing for healthcare, Dec 20, 2021
The standard approach to expert-in-the-loop machine learning is active learning, where, repeatedl... more The standard approach to expert-in-the-loop machine learning is active learning, where, repeatedly, an expert is asked to annotate one or more records and the machine finds a classifier that respects all annotations made until that point. We propose an alternative approach, IQRef, in which the expert iteratively designs a classifier and the machine helps him or her to determine how well it is performing and, importantly, when to stop, by reporting statistics on a fixed, hold-out sample of annotated records. We justify our approach based on prior work giving a theoretical model of how to re-use hold-out data. We compare the two approaches in the context of identifying a cohort of EHRs and examine their strengths and weaknesses through a case study arising from an optometric research problem. We conclude that both approaches are complementary, and we recommend that they both be employed in conjunction to address the problem of cohort identification in health research. CCS Concepts: • Social and professional topics → Personal health records; • Applied computing → Document searching; • Information systems → Retrieval effectiveness;
Journal of Vision, Jul 15, 2022
Investigative Ophthalmology & Visual Science, Jul 13, 2018
Investigative Ophthalmology & Visual Science, May 1, 2004
Optometry and Vision Science, 2012
A significant number of patients return to optometric practice dissatisfied with their spectacles... more A significant number of patients return to optometric practice dissatisfied with their spectacles. An important question is whether any of these cases are preventable. There are several different clinical maxims that are used to modify the subjective refraction when determining the refractive prescription. These maxims aim to improve patient comfort and adaptation and thereby reduce patient dissatisfaction with new spectacles. They are not based on research evidence, but rather on expert opinion gained from clinical experience. The aim of this study was to retrospectively analyze a large number of case records of dissatisfied patients to assess the possible usefulness of the prescribing maxim "if it ain't broke, don't fix it." Three hundred eighteen non-tolerance cases from a university-based Canadian optometric clinic were categorized by a focus group of optometrists. Three prescribing categories were defined and comprised cases in which application of the proposed maxim may have prevented the recheck eye examination; a more limited application of the maxim for one working distance may have been appropriate; and finally scenarios in which the maxim did not work in that the practitioner was judged to have initially followed the maxim, yet patient dissatisfaction was still reported. The remaining unallocated records comprised prescribing situations outside the scope of this study. Approximately 32% of non-tolerance cases were judged to have been preventable by use of the proposed maxim. Furthermore, an additional 10% reduction in recheck cases may have been possible by a more liberal interpretation of the maxim. Conversely, 4% of cases were deemed to comprise scenarios in which the maxim was followed yet the patient returned later to report problems with their spectacles. The prescribing maxim "if it ain't broke,…
Brain Research, Feb 1, 2009
Pro and antisaccades are usually presented in blocks of similar type but they can also be present... more Pro and antisaccades are usually presented in blocks of similar type but they can also be presented such that prosaccade and antisaccade eye movements are mixed and a cue, usually the shape/colour of the fixation target or the peripheral target, determines which type of eye movement is required in a particular trial. A mixed-saccade task theoretically equalizes the inhibitory requirements for pro and antisaccades. Using a mixed-saccade task paradigm the aims of the study were to: 1) compare pro and antisaccades of children, 2) compare performance of children and adults and 3) explore the effect of increased working memory load in adults. The eye movements of 22 children (5-12 years) and 22 adults (20-51 years) were examined using a video-based eye tracking system (El-Mar Series 2020 Eye Tracker, Toronto, Canada). The task was a mixed-saccade task of pro and antisaccades and the colour of the peripheral target was the cue for whether the required saccade was to be a pro or an antisaccade. The children performed the mixed-saccade task and 11 adults performed the same mixed-saccade task alone and in a dual-task paradigm (together with mental subtraction or number repetition). A second group of 11 adults performed the mixed-saccade task alone. Children made mainly antisaccade errors. The adults' error rates increased in the mental subtraction dual-task condition but both antisaccade and prosaccade errors were made. It was concluded that the increased error rates of these two groups are reflective of different processing dynamics.
Gaceta óptica: Órgano Oficial del Colegio Nacional de Ópticos-Optometristas de España, 2010
PubMed, 2012
Myopia is characterized by an increase in axial length of the eye, but the reasons for the axial ... more Myopia is characterized by an increase in axial length of the eye, but the reasons for the axial elongation are still unknown. Higher intraocular pressure (IOP) has been associated with myopia and could be involved in eye enlargement. The purpose of this study was to investigate the effect of intraocular pressure on the geometry of the chick eye and to investigate whether an increase in IOP could cause the elongation of the eye. The IOP of ten 7-day old chick eyes was raised by injecting fluid into the eyes and the resulting deformation of the eyes was measured using digital cameras. In-vitro pressure-volume curves were obtained. The axial and equatorial strains (deformation normalized to the original dimension) were calculated. Our results showed that IOP increased exponentially with increasing injected volume. About 25 D myopia could be induced by the axial elongation created with an increase in IOP by 100 mmHg. As pressure increased from 0 to 140 mmHg, the chick eye elongated in the axial direction and initially contracted in the equatorial direction. The natural tendency of chick eyes is to elongate as IOP increases and this suggests that from a mechanical perspective IOP could play a role in myopia onset and progression. The results also suggest that oblate eyes might have higher risk of developing myopia.
Ophthalmic and Physiological Optics, Jul 26, 2017
progression control lens reverses induced myopia in chicks. Ophthalmic Physiol Opt 2017.
Computer Methods in Biomechanics and Biomedical Engineering, Feb 1, 2012
The geometry of anatomical specimens is very complex and accurate 3D reconstruction is important ... more The geometry of anatomical specimens is very complex and accurate 3D reconstruction is important for morphological studies, finite element analysis (FEA) and rapid prototyping. Although magnetic resonance imaging, computed tomography and laser scanners can be used for reconstructing biological structures, the cost of the equipment is fairly high and specialised technicians are required to operate the equipment, making such approaches limiting in terms of accessibility. In this paper, a novel automatic system for 3D surface reconstruction of the chick eye from digital photographs of a serially sectioned specimen is presented as a potential cost-effective and practical alternative. The system is designed to allow for automatic detection of the external surface of the chick eye. Automatic alignment of the photographs is performed using a combination of coloured markers and an algorithm based on complex phase order likelihood that is robust to noise and illumination variations. Automatic segmentation of the external boundaries of the eye from the aligned photographs is performed using a novel level-set segmentation approach based on a complex phase order energy functional. The extracted boundaries are sampled to construct a 3D point cloud, and a combination of Delaunay triangulation and subdivision surfaces is employed to construct the final triangular mesh. Experimental results using digital photographs of the chick eye show that the proposed system is capable of producing accurate 3D reconstructions of the external surface of the eye. The 3D model geometry is similar to a real chick eye and could be used for morphological studies and FEA.
Journal of Cataract and Refractive Surgery, Feb 1, 2004
To determine whether exposing the extraocular muscles (EOMs) to lidocaine via retrobulbar injecti... more To determine whether exposing the extraocular muscles (EOMs) to lidocaine via retrobulbar injection for cataract surgery has a demonstrable negative effect on subsequent function of the muscle.
Movement Disorders Clinical Practice, Nov 20, 2016
Background: We recently reported that convergence insufficiency (CI)-type visual symptomatology w... more Background: We recently reported that convergence insufficiency (CI)-type visual symptomatology was more prevalent in participants with Parkinson's disease (PD), compared to controls. The objective of this work was to determine the prevalence of a confirmed clinical diagnosis of CI in PD, compared to controls. Methods: Participants with (n = 80) and without (n = 80) PD were recruited and received an eye exam. Published criteria were used to arrive at a clinical diagnosis of CI. The Convergence Insufficiency Symptom Survey (CISS-15) questionnaire was administered to each participant, with a score of ≥21 being considered positive for CI symptomatology. Student t test, chi-square, or nonparametric tests at the 0.05 level were used for statistical significance. Results: A total of 43.8% of participants with versus 16.3% without PD had a clinical diagnosis of CI (P ≤ 0.001). A total of 53.8% of participants with versus 18.8% without PD had scores on the CISS-15 of ≥21 (P ≤ 0.001). Conclusions: These results indicate that individuals with PD have a higher prevalence of CI and CI symptomatology than controls. These data provide evidence supporting the notion that treatment for symptomatic CI should be investigated in individuals with PD. Although the motor symptoms of Parkinson's disease (PD) have long been documented, nonmotor aspects are less well recognized in clinical practice. 1 Among those are the ocular/visual symptoms, which are quite prevalent in patients with PD. 2 Many visual problems affecting PD patients, such as the alterations in the blink pattern, 3 contrast sensitivity, 4 color vision, 5,6 vergence eye movements/ocular motilities, 7,8 flash 9 and pattern 10 electroretinograms, and visual evoked potentials 11 are well documented and have been linked, at least in part, to the dopaminergic deficit found in the disease. Many patients with PD suffer from eye strain while reading 3,12 that is sometimes accompanied by double vision. 1,12-14 These two complaints are often linked to convergence insufficiency (CI) with a decreased convergence amplitude. Although CI is described as being common in PD, 15 little data exist in the literature. In a 27-patient study, Repka et al. 16 reported that PD patients had significantly more frequent signs and symptoms associated with CI compared to age-matched controls. Around one-third of their PD patients had a clinical diagnosis of CI. Lepore 17 found that all PD patients with or without diplopia who could be tested had CI, and that those without diplopia had significantly higher convergence amplitudes and shorter disease duration. Other studies on PD often indicate that double vision is prevalent in these patients, 1,3,13,14 without mentioning the cause of diplopia. Recently, we reported that CI-type visual symptomatology was more
Experimental Eye Research, Jun 1, 2019
The purpose of the study was to analyze vertical saccade parameters (latency, peak velocity, ampl... more The purpose of the study was to analyze vertical saccade parameters (latency, peak velocity, amplitude gain), and compare them to those of horizontal saccades in a cross-sectional study across the ages of the human lifespan. One hundred and thirty one participants (62 males) between the ages of 3 and 86 years made vertical prosaccades of 2−44 degrees in response to a dot stimulus projected on a screen. A subset of participants also made horizontal prosaccades of 2−60 degrees under the same conditions. The El-Mar eye tracker was used to record binocular eye movements. Measures of saccadic latency, peak velocity and amplitude gain were calculated for each participant. Differences between saccade parameters for upward & downward saccades were calculated. Vertical saccade parameters were evaluated as a function of age and age related differences between vertical and horizontal saccade parameters were determined. There was no significant difference between upward and downward saccades and no effect of age for either latency or peak velocity. Downward saccades had significantly higher gains than upward saccades (p = 0.0001) and this difference increased significantly with age (p = 0.001). Vertical saccadic latency initially decreased from about 400 msec at four years of age, remained stable for a period of time and then increases again in later life. The lowest peak velocities were found in participants under 20 and over 70 years of age, while the highest peak velocities were seen in participants between 20 and 60 years of age. The majority of vertical saccades were hypometric. Saccadic amplitude gains varied depending on both the stimulus size (p = 0.0001) and age (p = 0.0001) of participants. Vertical saccades are most accurate for small amplitudes and for participants between 20−30 years of age. Vertical saccades had significantly longer latencies than horizontal saccades (p = 0.0001) but there was no significant effect of age. Vertical saccades had lower peak velocities than horizontal saccades in very young children but this difference decreased with age (p = 0.0015). Large vertical saccades were more hypometric than their horizontal counterparts across all ages. The observed differences in saccadic parameters could be related to the different areas in the brain used for saccadic generation, different periods and/or mechanisms of development and senescence within the visual system and brain and/or the effects of differential use.
Vision Research, May 1, 1995
Lightweight translucent plastic goggles with convex or concave rigid contact lens inserts were ap... more Lightweight translucent plastic goggles with convex or concave rigid contact lens inserts were applied unilaterally to the eyes of young chicks. Convex and concave cylindrical lenses produced astigmatic refractive errors. The magnitude of the induced astigmatism was less than that of the inducing lens and varied with axis orientation. Decreased aperture size or interruption of the defocus resulted in a decreased response to refractive defocus. Slit apertures and spherical defocus produced variable amounts of myopia, hyperopia and astigmatism. Choroidal changes (increased thickness) were observed only in birds developing hyperopia or recovering from myopia. Chick Defocus Astigmatism Aperture size Aperture shape
A number of experiments have demonstrated that it is possible to induce ametropias in the eyes of... more A number of experiments have demonstrated that it is possible to induce ametropias in the eyes of young animals by distorting early visual experience. Much of the work has involved the use of chicks (Gallus gallus domesticus) because of their precocial nature and because they grow and develop rapidly. The earlier studies generally involved depriving the retina of a clear image by either suturing the eyelids together or applying some kind of translucent or light scattering device over the eye during development (e.g., Pickett-Seltner, et al., 1987). This kind of treatment invariably leads to varying levels of myopia, often referred to as form deprivation or experimental myopia, through the axial elongation of the eye.
Over the past three decades, refractive errors have been induced in a variety of vertebrate speci... more Over the past three decades, refractive errors have been induced in a variety of vertebrate species by various methods. By manipulating the visual environment during early visual experience, hyperopia, myopia, and astigmatism have been induced. When using defocus to induce refractive error, the error results from an adjustment in the emmetropization process in order to compensate for the induced blur. The rationale associated with lid suture, form deprivation, dark rearing, continuous light exposure, and dim light exposure is less clear.
Optometry and Vision Science, Feb 1, 1991
Experiments were conducted to characterize the distribution of the monocular components of fixati... more Experiments were conducted to characterize the distribution of the monocular components of fixation disparity, i.e., ascertain whether or not the fixation disparity measured by nonius alignments is equally divided between the two eyes or has an unequal distribution. Computer generated nonius lines were used to measure the monocular components of the forced vergence fixation disparity curve (FDC) for subjects with normal binocularity. Thirty-three percent of all subjects tested showed unequal distribution of their monocular components. The composite FDC's for subjects with unequal distribution of the monocular components tended to have steeper slopes than those with equally distributed components. It was concluded that equally divided monocular components are not a prerequisite for good binocularity.
Optometry and Vision Science, May 1, 1999
Uploads
Papers by Elizabeth Irving