Research Progress in Artificial Intelligence for Central Serous Chorioretinopathy: A Systematic Review
Abstract
INTRODUCTION: This review synthesizes advancements in artificial intelligence (AI) applications for central serous chorioretinopathy (CSCR), analyzing challenges and outlining future research directions to guide personalized diagnostic and therapeutic strategies.
METHODS: A systematic literature search was conducted in the Web of Science database using a comprehensive AI-related keyword set (e.g., "deep learning," "neural networks," "computer vision") combined with "central serous chorioretinopathy." The search yielded 698 records, with 73 original research studies selected after excluding reviews and non-empirical work based on predefined criteria.
RESULTS: The application of AI in CSCR has progressed from disease classification to dynamic prognostic prediction, leveraging multimodal data fusion (e.g., optical coherence tomography [OCT] and OCT angiography [OCTA] and fundus fluorescein angiography [FFA]) to enhance both qualitative and quantitative diagnostic accuracy. AI models outperform clinical experts in classifying retinal disease subtypes and segmenting lesions. However, clinical translation faces infrastructural barriers (e.g., incompatible PACS systems) and limited physician trust due to "black box" decision-making. New approaches, such as explainable AI (XAI), are being integrated to enhance the transparency and clinical applicability of AI models. Key limitations involve single-center data dependency, interobserver annotation variability, and the inability of…
How strong is the evidence?
Evidence Hub reads this as a systematic review. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.