Focusing on the application and challenges of artificial intelligence in uveal malignant melanoma

Uveal malignant melanoma is one of the common primary intraocular malignancies in adults.Its high concealment and significant metastatic potential lead to a high risk of blindness and mortality.With advances in machine learning and deep learning techniques, artificial intelligence (AI) has shown increasing promise for application in the diagnosis, management, and prognosis evaluation of uveal malignant melanoma.AI can thoroughly analyze the multi-modal data, such as clinical images, pathological images, and genetic data, and assist clinicians in diagnosis and treatment planning.AI analyzes ophthalmic photography and radiological image to assist in differential diagnosis, and predicts side effects and outcomes of radiotherapy to optimize treatments.AI constructs the models for accurate prognosis based on clinical features and digital pathology images, and its accuracy is comparable to that of gene expression profiling tests.The clinical application of AI in uveal malignant melanoma faces the challenges of data availability, technology limitations, and effective human-machine collaboration.However, with ongoing research in both uveal malignant melanoma and AI, AI is expected to improve the accuracy and efficiency of diagnosis, management, and prognosis assessment, ultimately improving patient outcomes.

KEYWORDS：