A machine learning approach for retinal images analysis as an objective screening method for children with autism spectrum disorder.

BACKGROUND: Autism spectrum disorder (ASD) is characterised by many of features including problem in social interactions, different ways of learning, some children showing a keen interest in specific subjects, inclination to routines, challenges in typical communication, and particular ways of processing sensory information. Early intervention and suitable supports for these children may make a significant contribution to their development. However, considerable difficulties have been encountered in the screening and diagnosis of ASD. The literature has indicated that certain retinal features are significantly associated with ASD. In this study, we investigated the use of machine learning approaches on retinal images to further enhance the classification accuracy. METHODS: Forty-six ASD participants were recruited from three special needs schools and 24 normal control were recruited from the community. Among them, 23 age-gender matched ASD and normal control participant-pairs were constructed for the primary analysis. All retinal images were captured using a nonmydriatic fundus camera. Automatic retinal image analysis (ARIA) methodology applying machine-learning technology was used to optimise the information of the retina to develop a classification model for ASD. The model's validity was then assessed using a 10-fold cross-validation approach to assess its validity. FINDINGS: The sensitivity and specificity were 95.7% (95% CI 76.0%, 99.8%) and 91.3% (95% CI 70.5%, 98.5%) respectively. The area under the ROC curve was 0.974 (95% CI 0.934, 1.000); however, it was noted that the specificity for female participants might not be as high as that for male participants. INTERPRETATION: Because ARIA is a fully automatic cloud-based algorithm and relies only on retinal images, it can be used as a risk assessment tool for ASD screening. Further diagnosis and confirmation can then be made by professionals, and potential treatment may be provided at a relatively early stage.

Orthokeratology: clinical utility and patient perspectives.

Orthokeratology (ortho-k) is a special rigid contact lens worn at night to achieve myopic reduction and control. This review provides an overview on prescribing ortho-k, including clinical consideration on patient aspect and lens design; its clinical outcomes; and clinical efficacy and safety. Patient satisfaction was summarized. In order to achieve long-term healthy ortho-k treatment, it requires both patient and practitioners' care and rapport to maintain good ocular health and lens conditions.

High myopia-partial reduction ortho-k: a 2-year randomized study.

PURPOSE: To investigate if the combination of partial reduction (PR) orthokeratology (ortho-k) and spectacles for residual refractive errors in the daytime was effective to slow myopic progression in high myopic children. METHODS: High myopic children (aged 8 to 11 years) with spherical equivalent refraction at least -5.75 diopters (D) and myopia -5.00 D or more myopic were recruited and randomly assigned into PR ortho-k and control groups. Subjects in the PR ortho-k group were fitted with custom made four-zone ortho-k lenses with target reduction of 4.00 D for both eyes, and the residual refractive errors were corrected with single-vision spectacles for clear vision in the daytime. Control subjects were fully corrected with single-vision spectacles. Axial length of each eye of all subjects was measured with the IOLMaster at 6-month intervals by a masked examiner. This study was registered at www.clinicaltrial.gov with the identifier NCT00977236. RESULTS: Fifty-two subjects were recruited and randomized to the PR ortho-k and control groups. Twelve PR ortho-k and 16 control subjects completed the study. Compared with the residual refractive errors at the 1-month visit (after stabilization of ortho-k treatment), the median increase in noncycloplegic residual myopia at the 24-month visit was 0.13 D. In the control group, the median increase in myopia was 1.00 D at the end of the study. The mean ± SD increases in axial length were 0.19 ± 0.21 mm in the PR ortho-k group and 0.51 ± 0.32 mm in the control group (95% confidence interval, -0.55 to -0.12; unpaired t test, p = 0.005). CONCLUSIONS: This single-masked randomized study showed that PR ortho-k effectively slowed myopic progression in high myopes. Axial length elongation was 63% slower in PR ortho-k-treated children compared with children wearing spectacles.

High myopia-partial reduction orthokeratology (HM-PRO): study design.

PURPOSE: To report the study design and preliminary results of a pilot study, High Myopia-Partial Reduction Orthokeratology study. METHODS: Children with myopia of 6.00D or above and who satisfied the recruitment criteria were randomly assigned to partial reduction orthokeratology (PR ortho-k) and spectacle-wearing control groups. The myopia of the PR ortho-k children were partially reduced using custom made 4-zone ortho-k lenses of target 4.00D. Residual refractive errors were corrected with single vision spectacles. Control subjects were fully corrected with single vision spectacles. PR ortho-k subjects were also required to return for assessment after the first overnight lens wear, and one week and one month after lens wear. RESULTS: Fifty-two eligible subjects were randomly assigned to PR ortho-k group (n=26) and control group (n=26). The median age of each group was 10.00 years. The median (range) subjective myopia of the right eye at baseline was 6.41D (5.00-8.00D) and 6.22D (6.00-8.00D) for PR ortho-k and spectacle groups, respectively (p>0.05). Nineteen (79%) PR ortho-k subjects achieved successful lens fit at the one month visit and the median myopic reduction was 3.75D in the right eye. The incidence of (mild) corneal staining in PR ortho-k subjects reduced from 30% at the first overnight lens wear to 16% at the 1-month visit. Corneal pigmented arc was observed in 32% of PR ortho-k subjects at the 1-month visit. CONCLUSIONS: PR ortho-k was successfully applied to high myopic children with no significant changes in ocular health/best corrected visual acuity after one month of lens wear.

Visual outcome of Soflens Daily Disposable and Soflens Daily Disposable for Astigmatism in subjects with low astigmatism.

PURPOSE: To compare the visual outcome of Soflens Daily Disposables and Soflens Daily Disposable for Astigmatism (Bausch & Lomb, Rochester, NY, USA) when worn by subjects with low astigmatism. METHODS: Forty-one subjects (aged 19 to 40 years) with myopia between 0.50 D to 6.00 D and astigmatism not more than -1.25 D in at least one eye were recruited and fitted with spherical and toric soft contact lenses in this double-masked, non-dispensing cross-over study. High and low contrast logMAR visual acuities were measured over the lenses. At the end of the study, each subject was asked to complete a questionnaire on vision, initial comfort and ease of handling of the two lens types. RESULTS: Monocular and binocular high and low contrast logMAR visual acuities were significantly better with the toric lenses compared with the spherical lenses (p < 0.01). Results from the questionnaire showed a significantly higher grade for vision with the toric lenses than with the spherical lenses (p = 0.003) but no significant differences in initial comfort and ease of handling between spherical and toric lenses (p > 0.21). CONCLUSIONS: The results of the present study show that both subjective and objective vision in myopic subjects with low astigmatism improves significantly when corrected with toric lenses compared with spherical lenses.

Practitioners' analysis of contact lens practice in Hong Kong.

BACKGROUND: The aims of this survey were to appraise contact lens practice and care systems used in Hong Kong between March 2007 and February 2008, and to evaluate the attitude of practitioners towards prescribing different types of contact lenses. METHODS: Questionnaires were sent out to 797 optometrists licensed to fit contact lenses in Hong Kong in March 2008. RESULTS: The response rate was 27%. About 20% of contact lens patients (as reported by the respondents) were new wearers. Over 90% of contact lens patients were wearing soft contact lenses. Among soft contact lenses, the use of daily disposable and silicone hydrogel lenses had increased from 27% and 3% in 2002 to 37% and 21% in 2008, respectively. Less than 10% were rigid contact lens wearers. For astigmats fitted with soft lenses, only about 46% were fitted with toric lenses. For presbyopes using contact lenses, the majority (57%) were fitted with single vision lenses with over-spectacles, followed by monovision (29%). The most commonly recommended soft lens disinfecting system was multipurpose solution (74%). Less than 20% of the respondents reported confidence in prescribing orthokeratology lenses. CONCLUSION: Soft contact lenses continued to dominate the contact lens market in Hong Kong, with silicone hydrogel and daily disposable lenses becoming increasingly popular. The use of toric and bifocal lenses was still not popular among practitioners. Multipurpose solutions were still the care systems of choice by practitioners in Hong Kong.

Comparison of near heterophoria tests under varying conditions on an adult sample.

PURPOSE: This study compared the near phoria measurement using the Bernell muscle balance card with and without prism neutralization, using both trial frame and phoropter correction, and compared with the conventional Maddox rod method. METHODS: Forty young normal Chinese adults had their near phoria measured with trial frame correction using the conventional muscle balance card method (method 1). Any deviation was compensated with a prism bar as an alternative approach (method 2). The conventional Maddox rod method (method 3) was also carried out for comparison. These three methods were repeated with phoropter correction and considered as methods 4, 5 and 6. RESULTS: The phorias obtained from these six methods were not significantly different from each other (repeated measures anova, p > 0.05). More than half of the subjects were exophoric. Although the difference in phoria was not significant, phoria measurement using phoropter correction yielded a greater coefficient of variation. CONCLUSIONS: Near phoria measurement using the muscle balance card conducted with trial frame correction was less variable, and was also more natural and similar to a real reading situation. The use of prism for compensation did not affect the phoria results. Exophoria seems to be more common than esophoria in young Chinese adults.