<![CDATA[Diabetic retinopathy (DR) and diabetic macular oedema (DME) are significant causes of blindness globally, resulting from diabetes-induced retinal damage. Early detection and timely intervention are critical to prevent irreversible vision loss in diabetic patients. Recent advances in imaging technology, particularly Optical Coherence Tomography (OCT), have provided a powerful tool for detailed visualisation of retinal structures, enabling the early identification of microvascular changes associated with DR and DME. This systematic review synthesises insights from 20 studies to explore the effectiveness of OCT in diagnosing, monitoring, and guiding treatment for these conditions. A comprehensive literature search was conducted across databases including Scopus, Scholar, and PubMed, using Boolean operators to combine keywords such as “OCT,” “OCTA,” “diabetic retinopathy,” and “deep learning.” Studies were selected based on criteria that included the use of OCT or OCTA in assessing DR or DME, providing statistical data on diagnostic accuracy and treatment response. The results indicate that OCT, particularly when paired with OCT Angiography (OCTA) and AI-driven analysis, significantly enhances the accuracy of detecting early microvascular changes in diabetic eyes. For instance, Zhang et al. (2021) found that OCTA could identify early retinal vascular alterations in diabetic patients with a sensitivity and specificity exceeding 90%. Additional findings reveal that OCT plays a crucial role in tracking disease progression and evaluating treatment efficacy, with improved visual outcomes observed in patients receiving anti-VEGF therapy monitored via OCTA. In conclusion, OCT has established itself as an invaluable tool in the management of DR and DME, enabling early diagnosis, precise monitoring, and tailored therapeutic interventions. Integrating artificial intelligence further augments OCT’s diagnostic capabilities, enhancing its potential to revolutionise diabetic eye care. However, accessibility and cost barriers remain challenges, emphasising the need for future research to focus on optimising AI models and expanding OCT accessibility in routine clinical settings.]]>
Copyrights © 2024
