Artificial Intelligence and Oct Angiography in Full Thickness Macular Hole. New Developments for Personalized Medicine

Background: To evaluate the predictive performance for 1-year visual acuity of an artificial intelligence (AI) based model applied to optical coherence tomography angiography (OCT-A) vascular layers scans from eyes with full thickness macular hole (FTMH). Methods: In this observational, cross-sectional study, 35 eyes of 35 patients with FTMH were analyzed by OCT-A before surgery. Superficial vascular plexus (SVP) and deep vascular plexus (DVP) high quality images were collected for the analysis. An AI approach based on convolutional neural networks (CNN) was used to generate a continuous predictive variable based on both SVP and DPV. Different pre-trained CNN networks were used for feature extraction and compared for predictive accuracy. Findings: 35 patients affected by FTMH were analyzed with a mean 1-year BCVA score distribution of 56.63 (±19.6 SD). Among the different tested models, the Inception V3 network applied on the combination of deep and superficial OCT-A images, showed the most significant differences between the two obtained image clusters: BCVA C1= 49.10 (±18.60 SD) and BCVA C2= 66.67 (±16.00 SD) (p=0.005). Interpretation: The AI based analysis of preoperative optical coherence tomography angiography (OCT-A) images of eyes affected by FTMH may be useful support systems in setting up predictive of visual acuity recovery. The combination of preoperative SVP and DVP images showed a significant morphological predictive performance for visual acuity recovery. Funding: Universita Cattolica del Sacro Cuore contributed to the funding of this research project and its publication. Declaration of Interest: The authors have no conflict of interest to disclose. Ethical Approval: This prospective, observational, cross-sectional study evaluated 35 eyes of 35 patients affected by FTMH and evaluated with preoperative OCTA. This study adheres to the tenets of the Declaration of Helsinki (52nd WMA General Assembly, Edinburgh, Scotland, October 2000), and written informed consent was obtained from all patients prior to participation in the study. Institutional Ethics Committee approval was obtained from the Fondazione Policlinico Universitario “A. Gemelli” IRCCS of Rome, Italy(ID number: 3680).