Exploring the Potential of Deep Learning in Proptosis Detection
The field of ophthalmology has witnessed significant advancements in recent years, thanks to the integration of deep learning algorithms into healthcare applications. In this article, we will explore the evolution of a project called Eye Metrics, developed by AI Superior, which accurately assess eyelid positioning and identifies abnormalities related to proptosis while providing an efficient eye health analysis experience for healthcare professionals.
The initial challenge faced by AI Superior was to develop an application that could accurately assess eyelid positioning and identify abnormalities related to proptosis, a condition characterized by the forward displacement of the eye. In order to do so, this application needed to calculate key metrics relevant to the ophthalmological domain. These metrics included Margin Reflex Distance 1 (MRD1) and Margin Reflex Distance 2 (MRD2), which measure the distance from the centre of the pupil to the upper and lower eyelid margins, respectively. Eyebrow Height, which determines the distance between the centre of the pupil and the eyebrow. Proptosis distance, which gauges the distance between the centre of the pupil and the bone of the orbit on the lateral side and eye movement distances in various directions. By accurately calculating these measurements, the application would help assess eyelid positioning and identify any potential abnormalities, which is crucial in identifying and monitoring proptosis.
Solution by AI Superior
To address this challenge, AI Superior leveraged advanced deep learning algorithms to create Eye Metrics, a cross-platform web application accessible on mobile devices (iOS and Android) as well as desktop and laptop browsers. Eye Metrics enabled users, primarily patients, to capture facial images and calculate key metrics relevant to the ophthalmological domain. Moreover, Eye Metrics was able to continually improves its predictive models by incorporating new training samples.
Building upon the success of Eye Metrics, AI Superior ehanced the application by introducing additional functionalities for eye surgical assessment. These functionalities aimed to provide a more comprehensive and efficient eye health analysis experience for healthcare professionals. The enhanced version included two key features: batch uploads and live video recording. The batch upload feature allowed users to upload a series of images taken before and after surgery. These images were then analyzed to identify the delta between LAB values of the sclera. This automated comparison of image data streamlined the analysis process, saving valuable time and enhancing efficiency for doctors. In addition, this new version also included a live video recording feature that enabled users to record video footage and then estimate the delta between LAB values over a specified time interval, typically 1-5 minutes.
Outcome and Implications
Eye Metrics is poised to transform the way patients and healthcare professionals approach ophthalmological assessment. With its user-friendly interface and a model that outperforms Google’s face mesh detection, this application helps healthcare professionals identify proptosis and empowers patients to take a proactive role in monitoring their eye health