RetinAI’s AI Algorithms Enrich Clinical Research and Improve Patient Care
Bern, Switzerland, April 18 2023
RetinAI Medical AG (“RetinAI”), a leader in the management and analysis of healthcare data through advanced software and AI, is excited to announce the acceptance of five abstracts for poster presentations at the upcoming Association for Research in Vision and Ophthalmology (ARVO) conference taking place from April 23rd to 26th, 2023 in New Orleans. These abstracts highlight the use of RetinAI’s Discovery® platform ('Discovery') and its certified and research use only (RUO) AI models in recent studies across several retinal diseases. Discovery was used for acceleration of clinical trials, enriching clinical research and analysis and helping to drive better treatment decisions with robust analysis across patient populations and outcomes.
“By providing clinicians & physicians with extensively validated AI algorithms, we can improve patient care, help accelerate clinical trials, and drive better decisions for patient management" said Dr. Joseph Blair, ML Research Scientist at RetinAI.
One of the studies, CARDS (Multi Center Study to Validate AI for the Screening of Diabetic Retinopathy), was conducted across five different sites in collaboration with Novartis and the Fundación Ver Salud, focused on determining if a deep learning-based artificial intelligence algorithm called LuxIA in addition to RetinAI’s Discovery, could be used for the screening of more-than-mild Diabetic Retinopathy (mtmDR). The study demonstrated that the algorithm could be used as an automatic tool to screen patients with DR, increasing the testing feasibility for medical professionals in DR screening. This study validates the use of the LuxIA algorithm for mtmDR screening in a real-world primary care setting. The Discovery platform and LuxIA are a significant aid for both ophthalmologists and patients, enabling timely referral of diabetic retinopathy patients. Without proper screening and detection, many patients are at risk of blindness due to delayed treatments.
“With Discovery's algorithms, clinicians can get results within seconds, making it faster and easier to diagnose patients” said Dr. Sandro De Zanet, co-founder and Head of Research at RetinAI.
RetinAI’s AI algorithms were also used in a clinical study to assess the accuracy of artificial intelligence (AI)-based tools in quantifying retinal biomarkers from Optical Coherence Tomography (OCT) images as a more efficient alternative to a human, without loss of accuracy. As a result, a significant correlation between the volume estimated for pathological fluids by a human grader and the DL algorithm was observed, making the algorithm a valuable tool for clinical trials, enabling clinicians to access quantitative metrics that are otherwise too expensive or slow to receive in real-world practice. With RetinAI’s Discovery platform and AI models, clinicians can receive accurately segmented images in seconds, as opposed to hours with human graders.
Another study tested RetinAI’s advanced segmentation tools used to quantify layer thicknesses in a-scans to show differences in the presence of Geographic Atrophy (GA). Geographic atrophy (GA) is a late-stage form of age-related macular degeneration (AMD) that is characterized by the loss of retinal layers, including photoreceptors, the retinal pigment epithelium (RPE), and the choriocapillaris (CC). With Discovery's advanced segmentation tools, clinicians can now segment multiple layers and pinpoint where atrophy begins and resides. This technology has the potential to support doctors in understanding GA better and enable further research into the disease.
The company’s algorithms were also used in a study to assess biomarkers for choroidal neovascularization (CNV) development prediction in multifocal choroiditis (MFC) and punctate inner choroidopathy (PIC). Different morphological biomarkers, volumes of different fluid and lesion compartments as well as layer thicknesses were assessed and AI-based segmentation was manually corrected for all OCT scans, as needed. Lesion volume is an important predictive biomarker for future secondary CNV development in eyes with PIC or MFC. This finding highlights the importance of evaluation of volumetric data on OCTs in both the diagnostic and prognostic processes in patients with MFC and PIC.
A final study that will be presented, investigates the relationship between hyperreflective foci, subretinal hyperreflective amorphic material such as SHRM and Fibrosis and thickness of retinal layers and fluids on OCT in patients with intermediate and advanced dry and exudative AMD. OCT volumes were used, and automated segmentation was done on the B-scan level with two convolutional neural network models. Associations between measurements were calculated using Spearman correlation, with significance at levels of P-values < 0.001. The findings of this study highlight the importance of quantification of HRF and SHRM/FIB as biomarkers for AMD progression. The correlation between SRF and HRF in early AMD may point toward a more pronounced inflammatory state of these eyes. For dry AMD the increased amount of HRF could be a predictive factor for pronounced GA progression. These hypotheses are interesting and warrant further longitudinal assessment.
With these five studies, we want to help researchers and physicians learn more about ophthalmic diseases while finding ways to more efficiently look at biomarkers for disease. If you are willing to participate to these poster sessions, here is the agenda at ARVO 2023:
RetinAI Medical AG (RetinAI), founded in 2017, is a Swiss startup from Bern developing software solutions to make eye care institutions more efficient using advanced machine learning and computer vision. The company builds tools to collect, organize and analyze health data from the eyes, empowering healthcare professionals and patients with unique medical data analysis supported by AI. RetinAI's international team combines clinical, technical and scientific knowledge to foster the transition from reactive to preventive medicine for high-impact diseases such as Age-related Macular Degeneration (AMD), Diabetic Retinopathy or Glaucoma. www.retinai.com