斜视的尽早诊断和合理干预对改善患者预后具有重要意义，当前斜视的筛查和诊断主要依赖人工检查，存在人力资源不足和误诊、漏诊风险。近年来，人工智能在斜视领域的应用迅猛发展，涵盖斜视筛查、诊断、手术参数估计及预后预测等方面。基于视频、眼位照片及彩色眼底照相的深度学习模型在斜视筛查和诊断中显示出巨大潜力。尽管AI在斜视诊疗中取得了显著成效，但研究多排除复杂斜视类型，依赖静态、单模态数据，其实用性和普适性仍需进一步提高。未来，结合大模型技术和多模态数据的智能诊疗平台的建设将提升斜视的管理和眼保健水平，有助于实现斜视的个性化精准诊疗。
Early diagnosis and appropriate intervention of strabismus are crucial for improving patient outcomes.Currently, strabismus screening and diagnosis rely on manual examination, which is challenged by limited human resources and the risk of misdiagnosis.Recently, artificial intelligence (AI) has made rapid progress in strabismus, covering screening, diagnosis, surgical parameter estimation and prognosis prediction.Deep learning models based on video, eye and fundus photographs show great potential.Despite significant achievements, AI studies often exclude complex strabismus types and rely on static, unimodal data, which limits practicality.Future integration of large model technology and multimodal data into intelligent diagnostic platforms will improve strabismus management and eye care, enabling personalized and precise treatment.
刘陇黔,吴达文. 关注人工智能在斜视诊疗中的应用[J]. 中华实验眼科杂志,2024,42(12)：1079-1083.
DOI:10.3760/cma.j.cn115989-20240611-00149版权归中华医学会所有。
未经授权，不得转载、摘编本刊文章，不得使用本刊的版式设计。
除非特别声明，本刊刊出的所有文章不代表中华医学会和本刊编委会的观点。
文档即答
智问智答
机器翻译
你好，我可以帮助您更好的了解本文，请向我提问您关注的问题。
