Development of an Optimal Algorithm for the Management of Patients With Retinal Pigment Epithelium Detachment in Neovascular Age-related Macular Degeneration Using Artificial Intelligence

Development of an Algorithm for Predicting Anatomical and Functional the Results of Therapy With Angiogenesis Inhibitors in Patients With Retinal Pigment Epithelium Detachments in Neovascular Age-related Macular Degeneration, Based on Primary Optical Coherence Tomography of the Macular Zone and Clinical Data.

The study involves the development of an algorithm for predicting anatomical and functional results of therapy with angiogenesis inhibitors in patients with retinal pigment epithelium detachments in neovascular age-related macular degeneration, based on primary optical coherence tomography of the macular zone and clinical data.

Patients were divided into 3 groups according to the results of therapy: adhesion of detachment, lack of adherence to detachment, rupture of detachment. For these groups, OCT images of the macular zone with maximum detachment before therapy are selected. These images, along with other clinical parameters, are input to the algorithm. The result is one of the 3 treatment outcomes listed above. The methods that will be used to develop the algorithm include methods for processing and transforming data, deep machine learning, metrics for calculating the accuracy of algorithms.

Inclusion criteria:

• Linear B - scan through the macular area with the longest detachment
• Other pathologies

Exclusion criteria:

• Images without detachment
• Images on which it is possible to diagnose the need for therapy only in the presence of additional factors not considered in the study.