Ocular surface tumors (OSTs) are rare but potentially life-threatening neoplasms encompassing entities such as ocular surface squamous neoplasia (OSSN), conjunctival melanoma, and lymphoma. Accurate diagnosis often requires expert ophthalmologists and pathologists, compounded by the reliance on advanced imaging modalities, with excisional biopsy being the gold standard. These limitations underscore the need for less invasive, accessible diagnostic approaches, where artificial intelligence (AI) holds significant promise. This review provides a comprehensive overview of AI advancements in OST management. It begins with definitions of AI and its key branches, followed by an examination of AI models applied to ophthalmic tumors using imaging data. Current developments in AI-related diagnostic tools for OSTs are discussed, highlighting their potential to enhance patient management, with classifications based on imaging modalities and specific OST types. Finally, the review addresses main challenges in AI implementation, including data limitations and ethical considerations, while outlining future directions to integrate AI into clinical ophthalmology practice. By bridging technological innovation with clinical needs, AI shows promise in OST diagnosis and management, ultimately improving outcomes in this challenging condition.
Ghanbari et al. (Mon,) studied this question.