Advancements in Deep Learning-Augmented Adaptive Optical Coherence Tomography (AO-OCT) for Enhanced Real-Time Retinal Imaging: A Narrative Review
Volume 19, Issue 1, 2025
Download| Author(s): |
Ansharah Hasib Ziauddin University (FESTM) Karachi, Pakistan, ansharahhasib28@gmail.com Mehwish Faiz* Ziauddin University (FESTM) Karachi, Pakistan, mehwish.faiz@zu.edu.pk Syeda Hafiza Afsheen Zafar Bahria University Health Sciences Campus, Karachi , Pakistan, afsheen.bumdc@bahria.edu.pk Nazia Ejaz Balochistan University of Engineering and Technology, Khuzdar , Pakistan, naziaejaz@buetk.edu.pk Aneela Kiran The begum Nusrat Bhutto Women University, Sukkur, Pakistan, aneelakiranansari73@gmail.com Shahzad Nasim The begum Nusrat Bhutto Women University, Sukkur, Pakistan, shahzadnasim@live.com |
|---|---|
| Abstract | The assimilation of deep learning and Adaptive Optics Optical Coherence Tomography (AO-OCT) has promoted the area of retinal imaging to a new direction with ultra-high-resolution, intelligent diagnostics. AO-OCT comprises Optical Coherence Tomography (OCT) and Adaptive Optics (AO), which synergistically allows high-resolution, three-dimensional imaging of retinal structures. This amalgamation can provide a detailed cellular-level image of the retina that enables very early detection and monitoring microstructural abnormalities. Deep learning approaches, specifically convolutional neural networks and vision transformers (ViTs)) has further unlocked the potential of AO-OCT systems. They can execute real-time motion artifact correction, improve the image quality in suboptimal imaging conditions and automate segmentation of complex retinal layers. Deep learning reduces the burden of manual interpretation and compensates for patient movement, which improves both the accuracy as well as efficiency of AO-OCT based diagnostics. Clinically, these advancements are critical for the early diagnosis and intervention of a number of retinal disorders like age– related macular degeneration (AMD), diabetic retinopathy (DR) and glaucoma. The ability to visualize and interrogate subtle cellular changes can provide important insights into the progression of disease before symptoms become clinically apparent; this has profound implications for prevention, diagnosis and treatment. This narrative review discusses how deep learning has recently gained ground in AO-OCT and offers an overview of its system architecture, diagnostic outlook, current challenges, as well as possible directions for AI-driven ophthalmic imaging. |
| Keywords | Deep learning; intelligent diagnostics; diabetic retinopathy; ophthalmic imaging; AO-OCT based diagnostics. |
| Year | 2025 |
| Volume | 19 |
| Issue | 1 |
| Type | Research paper, manuscript, article |
| Journal Name | Journal of Information & Communication Technology | Publisher Name | ILMA University | Jel Classification | - | DOI | - | ISSN no (E, Electronic) | 2075-7239 | ISSN no (P, Print) | 2415-0169 | Country | Pakistan | City | Karachi | Institution Type | University | Journal Type | Open Access | Manuscript Processing | Blind Peer Reviewed | Format | Paper Link | https://jict.ilmauniversity.edu.pk/journal/jict/19.1/7.pdf | Page | 47-50 |