AI-driven innovations in signal/image processing and data analysis for optical coherence tomography in clinical applications

Fueled by the explosion of algorithms and computational innovations, optical coherence tomography (OCT) has progressed rapidly in the last decade, towards faster and more accurate imaging and characterization of ocular, systemic, and chronic diseases. This chapter describes recent advances in signal and image processing and data analysis methods responsible for this translational impact of OCT, which has been mainly in the retina, but other applications are included. The tools developed and used to enhance, segment, and extract meaningful and quantifiable parameters from OCT images are described. Traditional image processing methods are briefly outlined and AI-based innovations are reviewed. The importance of open research, protocol harmonization, big data, and patient data privacy in driving further innovation is also discussed. This chapter does not provide an exhaustive review, but rather its purpose is to be illustrative of the ongoing research and translational work and encourage engineers, scientists, and clinicians to work together in this exciting field. Sufficient detail is given to enable newcomers to the field, both engineers and clinicians, to understand the challenges and opportunities.