RESUMO
This study aimed to determine the correlation of the parameters that indicate the status of the ocular surface with the prognosis of corneal opacification. Fifty dogs (96 eyes) were examined using a grid-line illuminator (non-invasive tear film break-up time (NIBUT)). Thirty dogs (54 eyes) were included in the final analysis based on the criteria. The NIBUT and tear film break-up time (TFBUT) results of the eyes included in the study were divided into three groups: Group 1 (< 5 s), Group 2 (5 to <10 s), and Group 3 (≥ 10 s). The Schirmer's tear Test 1 (STT-1) results of the included patients were also divided into three groups: Group 1 (< 5 mm/min), Group 2 (5 to <10 mm/min), and Group 3 (≥ 10 mm/min). The corneal opacity grades are divided into four scores, ranging from 0 to 3. The corneal opacity grade score (COS) of 0 indicates a completely clear cornea or only a trace of opacity. COS of 1, 2, 3 indicate the presence of a prominent corneal opacity that does not interfere with the visualization of the fine iris details, the opacity obscures the visibility of the iris and lens details and severe obstruction of the intraocular structure visibility, respectively. The mean difference in COS during the follow-ups for each group of NIBUT were 0.61 ± 0.92 (n = 28), 0.10 ± 0.32 (n = 10), 0.19 ± 0.40 (n = 16). The NIBUT groups were significantly correlated with COS (p-value = 0.073) at a 10% level of significance. Post-hoc test at a 10% level of significance revealed significant correlations between Groups 1 and 2 (p-value = 0.041) and between Groups 1 and 3 (p-value 0.104). Although the TFBUT and STT-1 groups did not show any significant correlation with COS. Eyes with NIBUT of <5 s were found to have a significantly higher chance of increased COS compared with eyes with NIBUT of >5 s in the grid-line illumination plate NIBUT test. Among NIBUT, STT-1, and TFBUT, NIBUT was the only test that showed significant associations with the changes in COS.
RESUMO
The purpose of this study was to develop an object detection method for the diagnosis of dry eye disease (DED) in dogs. To this end, a methodology was designed to evaluate ocular surface video images using the YOLOv5 model, which is an object detection algorithm that has been widely used because of its simple network structure and fast detection speed. Because the cornea is a transparent organ, an illuminator plate with grid squares was used to provide grid lines, which were analyzed as the reflected straight lines of the light source representing the precorneal tear film (PTF) stability. The original video consisted of the number of 12 normal images(normal, [Formula: see text] = 17) and the number of 15 abnormal images(abnormal, [Formula: see text] = 17), converted to JPEG images for labeling, learning, and model validation. The labeled image data were divided into a training image data set (normal, [Formula: see text] = 15,276; abnormal, [Formula: see text] = 26,196) to a validation image data set (normal, [Formula: see text] = 6546; abnormal, [Formula: see text] = 11,228). As a result of the experiment, the mean average precision ([Formula: see text]) achieved 0.995. This study proposes a method to effectively determine ocular surface status in dogs by using YOLOv5 and concludes that an object detection model can be used in the veterinary field.
