AI-based photoreceptor length segmentation and influence on disease progression and visual acuity changes in central serous chorioretinopathy (CSCR)

Abstract

Purpose : In central serous chorioretinopathy (CSCR), longer presence of subretinal fluid leads to growth of the outer segment of photoreceptors. However, it is unclear if this elongation leads to disease progression or permanent reduction of visual function. With AI-based biomarker annotation, we automatically analyzed these changes in a large cohort.

Methods : Retrospective analysis of 1685 examinations of 629 eyes of 317 subjects was conducted in patients with long time courses of CSCR. With the help of artificial intelligence, 1685 autofluorescence images were annotated for hyperfluorescence (HF) and 1368 optical coherence tomography (OCT) exams were annotated for subretinal fluid (SRF) and length of photoreceptors (LPR). The imaging modalities were then automatically registered onto each other. We then analyzed the presence of these biomarkers within 250 μm distance to the fovea.

Results : When SRF was present, elongated photoreceptors were significantly associated with reduced VA (p = 0.001). However, in patients where the fluid resolved, no negative permanent effect of elongated photoreceptors was found on final visual acuity (p=0.68). Also, no effect of LPR was found for growth of HF (p = 0.74).

Conclusions : AI-based biomarker detection and quantification can help to understand the disease. Although the perceptive function of the photoreceptor seems to be impaired by photoreceptor elongation, the elongation can resolve when fluid is resorbed and leaves no functional scarring. Further analysis of which factors lead to permanent visual impairment is needed.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.