EyeMark™ takes fully automated diabetic retinopathy analysis one step beyond imaging, grading and reporting. It is designed to automatically and accurately track disease progression from visit to visit to determine if a patient’s disease has increased, decreased or remained stable. This information augments the “Positive/Negative” disease assessment that is available today and is valuable for disease management.
Internally, the technology contains the EyeDiff™ engine which (i) automatically registers the images from different visits of a patient, (ii) detects and localizes the lesions present, and (iii) displays the lesion change visualization, along with lesion count and area statistics. Following this, the EyeMark Change Detection engine estimates microaneurysm (MA) appearance and disappearance rates (known as turnover rates) for use as a biomarker in monitoring the progression of DR. Early studies have shown the promise of EyeMark and the technology is under active development.
MA turnover is a biomarker
Microaneurysms (MA) are the first signs of DR and are therefore of significant interest. MA turnover indicates MA appearance and disappearance rates. Studies have shown that MA turnover can be used as a biomarker for monitoring the progression of diabetic retinopathy. High MA turnover rates possibly indicate higher risk for DR progression.
Beyond human precision
MA turnover measurement involves two steps: (i) careful alignment of current and baseline images, and (ii) marking of individual MAs. This process is very time consuming and prone to error, if done entirely by human graders. The problem is compounded by the variable image quality and inter- & intra-observer variability between screenings. EyeMark overcomes the above limitations by automating both the steps involved in MA turnover measurement: accurate image registration, and MA detection.
Technology behind EyeMark
EyeMark is powered by advanced, multi-scale image registration algorithms that accurately aligns images of a patient across multiple visits and is robust to variations in image appearance and quality. It also includes state-of-the-art deep learning based lesion detection algorithms and proprietary hybrid change detection techniques that uses multi-scale morphological filtering with deep learning. These technologies combined with an intuitive user interface make up EyeMark software.
How does EyeMark work?
A visiting patient is imaged by an eye care professional using a commercial fundus camera and the fundus images are input to EyeMark. The software automatically analyzes and grades the images (for current visit and all available previous visits), intuitively displays the lesion change visualization, and issues a simple, quantitative report.
EyeRead™ UWF is an autonomous tool for lesion characterization in ultra-widefield scanning laser ophthalmoscopy (UWF SLO) images.
In recent times non mydriatic UWF SLO imaging has been shown to be a promising alternative to conventional digital color fundus imaging for grading of diabetic eye diseases, with advantages including 130°-200° field-of-view showing more than 80% of the retina in a single image, no need for multiple fields, multiple flashes, or refocusing between field acquisitions, ability to penetrate media opacities like cataract, and lower rate of ungradable images. UWF SLO images are particularly suitable for detecting predominantly peripheral lesions (PPLs), which have been associated with higher risk of diabetic retinopathy (DR) progression. Accurate quantification of presence and extent of PPLs can only be done by a robust automated tool that is specifically designed for the pseudo-colored images of UWF SLO modality.
EyeRead™ UWF will automatically characterize lesions in pseudo colored UWF images while handling possible artifacts from eyelashes/eyelids and determine the lesion predominance in peripheral and central regions of UWF image. The ability to accurately quantify the presence and extent of predominantly peripheral lesions in UWF SLO images can enable clinicians to develop a more precise DR scoring scheme. This would help identify patients with higher risk of DR progression and onset of PDR, have a positive impact on diabetic patient management, and aid drug discovery research. Results from early prototypes of the tool have been published in Acta Ophthalmologica and presented in ASRS meeting.
Funded in part by prestigious National Institute of Health (NIH/NEI) grant EY028081
Eyenuk's AMD software is designed to be a fully automated retinal imaging analysis technology to identify age-related macular degeneration (AMD), the number-one cause of legal blindness in seniors. Our AMD tool will fully automate the screening, grading and reporting process for AMD, without need for a human grader and with high sensitivity and specificity. This will enable even a primary eye care provider or optometrist to screen for AMD, which is possible today only in a specialist’s office, offering the potential to identify and treat far more patients than is possible today.
The Our AMD uses proprietary image analysis technology for identification and quantification of drusen – deposits under the retina that are indicators of AMD – viewable in color fundus images. Promising early results were published in the IOVS Journal from ARVO.
Need for AMD Screening
Age-related macular degeneration (AMD) is the single largest cause for legal blindness among senior Americans. Non-neovascular or dry (atrophic) AMD is the primary form of AMD contributing to 85% of the AMD cases and may be asymptomatic or result in a gradual vision loss. The largest and most impactful dry AMD studies - Age-Related Eye Disease study (AREDS) and Age-Related Eye Disease study 2 (AREDS2) - have shown that vitamin supplements along with minerals and antioxidants slow progression of the disease in patients with intermediate AMD, and those with late AMD in one eye. But without a straightforward, reliable way to screen for these vulnerable patients, early detection and treatment will remain a problem.
Technology behind our AMD tool
Our image analysis algorithms represent cutting-edge of research in image processing, computer vision, and machine learning. Technological innovations like morphology-inspired filter bank descriptors combined with deep learning can automatically analyze fundus images to detect and localize lesions resulting from AMD.
Eyenuk's Glaucoma software is designed to automatically analyze retinal images to determine the presence of Glaucoma. Our Glaucoma tool can enable screening, grading, and reporting for Glaucoma directly at the point-of-care without the need for a human expert to grade the images. This will make glaucoma screening more accessible.
Today, testing for Glaucoma not only includes retinal images but also other procedures such as tonometry and perimetry testing. Our Glaucoma tool is designed to augment information from additional tests, if available, to provide improved performance.