Developing a diagnosis model for dry eye disease in dogs using object detection

Abstract

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, $$n$$ n = 17) and the number of 15 abnormal images(abnormal, $$n$$ n = 17), converted to JPEG images for labeling, learning, and model validation. The labeled image data were divided into a training image data set (normal, $$n$$ n = 15,276; abnormal, $$n$$ n = 26,196) to a validation image data set (normal, $$n$$ n = 6546; abnormal, $$n$$ n = 11,228). As a result of the experiment, the mean average precision ($$mAP$$ mAP ) 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.

Computer vision model for the detection of canine pododermatitis and neoplasia of the paw

“…Previous studies used image classification to label entire images, while the current approach uses object detection to not only locate, but also label individual objects within an image. A study reported a YOLOv5 model for dry eye disease in dogs with a very high mAP of 0.995 40 . Another group built a model for dermatophytosis, mange and fleas 38 .…”

“…A study reported a YOLOv5 model for dry eye disease in dogs with a very high mAP of 0.995. 40 Another group built a model for dermatophytosis, mange and fleas. 38 Another model evaluated 12 dog skin diseases.…”

A novel method of evaluating the non-invasive tear film break-up time and progression of corneal opacification in dogs using imaging video

“…The following method was used to evaluate NIBUT with the grid-line illuminator. Slit-lamp biomicroscope (GS LED Slit Lamp MW50D, Shigiya machinery works LTD., Japan) with an add-on custom-made grid-plate illuminator (a 42 cm × 33 cm plate with LED lights corresponding to a total of 19 squares × 15 squares of 2 cm 2 in size grid) was used to obtain a 20-s video of images from all patients ( Figures 1 , 2 ) ( 14 ). All videos were recorded prior to conducting any of the other examinations.…”

“…A modified version of the original NIBUT system developed by Mengher et al ( 13 ), which sheds a grid-shaped illumination on the corneal surface, has been used as the diagnostic method to evaluate the status of the ocular surface in this study. Previous studies on the use of grid-line illumination NIBUT system utilized an add-on illuminator plate over the commonly distributed ophthalmic slit-lamp biomicroscopes ( 14 ). The square grid lines of light reflected on the corneal surface represent the precorneal tear film (PTF), and the disturbance or disruption of these lines represents the break-up of the tear film ( 14 ).…”

“…Previous studies on the use of grid-line illumination NIBUT system utilized an add-on illuminator plate over the commonly distributed ophthalmic slit-lamp biomicroscopes ( 14 ). The square grid lines of light reflected on the corneal surface represent the precorneal tear film (PTF), and the disturbance or disruption of these lines represents the break-up of the tear film ( 14 ).…”