**2.2 The development of optical coherence tomography angiography**

Although OCT can provide information about the structure of the retina and the optic nerve, it is unable to assess the vascular impairment and degradation known to play a key role in glaucoma pathophysiology. To fully assess damage in glaucoma patients, additional details about the retina and optic nerve vasculature are needed.

Initial work by White et al. showed that SD- OCT could be used to probe the retinal vasculature and was expanded upon by Zhang et al. who used a technique called swept source (SS) OCT to visualize blood flow in 2005 [14, 15]. This was accomplished by repeatedly scanning a transverse cross-sectional area of the retina and analyzing the scan differences caused by movement [6]. Since the presence of blood flow creates movement which affects the reflectance in each scan, successive scans or "angiograms" can be used to probe the underlying vasculature of the eye.

In 2018, Nesper et al. demonstrated the complexity and connectivity of parafoveal capillary plexuses and identified the location of specific arterioles seen in prior histological studies [28]. This enhanced understanding of vasculature and their perfusion greatly improved the assessment of many ophthalmological diseases, including glaucoma.

Work in the past decade has focused on accurately delineating blood vessels by employing sophisticated algorithms. In 2012 Reif et al. introduced a phase-based algorithm known as optical microangiography (OMAG), which uses intensity and phase information from repeated B-scans to delineate blood vessels [19]. In the same year, Spaide et al. showed that an amplitude-based method called split-spectrum amplitude decorrelation angiography (SSADA) could delineate blood vessels by measuring the decorrelation between two consecutive B-scans [6]. Research has found SSADA to be superior in the detection and connectivity of microvascular networks compared to other algorithms [18]. Current investigations of OCTA technology aim to implement novel deep learning (DL) and machine learning (ML) techniques to reduce or remove image artifacts to enhance OCTA accuracy.
