平行眼:基于ACP的智能眼科诊疗

摘要

ACP理论是平行智能的核心,由人工社会( Artificial Societies, A)、计算实验( Computational Experiments, C)、平行执行(Parallel Execution, P)构成,在复杂系统建模与调控中发挥重要的作用.人眼作为一个典型的复杂系统,是接收外界信息的重要器官,拥有复杂而又精细的结构,同时又和身体其他器官有着密不可分的联系.文中将ACP理论引入到人眼的诊疗和护理中,提出平行眼的基本框架.人工眼( A)构建不断更新的虚拟人眼模型及其对应的由诊疗护理知识构成的虚拟眼科医生.在此基础上,计算实验(C)借助人工眼进行各种眼病发展预测并制定最优的治疗护理方案.最后平行执行(P)将虚拟世界和人工世界串联起来,利用实际人眼的数据不断更新人工眼模型,同时借助人工眼和计算实验引导实际人眼进行在线化、长期化的治疗护理.整个框架结合知识自动化、计算机视觉、知识图谱、机器学习等相关人工智能技术,力图有效提升对于人眼疾病的诊断精度,实现对于人眼的长期、精确、有效的监护.%The ACP theory, including artificial societies(A), computational experiments(C) and parallel execution( P) , is the core of parallel intelligence. This theory plays an important role in modeling and controlling complex systems. With a complex structure, the human eye is an important organ for human to receive the information of the world, and it also has close connection with other organs of the body. In this paper, the ACP theory is introduced into the diagnosis and treatment of the human eye, and the basic framework of the parallel eye system is proposed. The artificial eye system ( A) is built to continuously describe and update the virtual eye model and the virtual doctors represented with the medical knowledge of diagnosis and treatment of the human eye. The computer experiments( C) perform all kinds of experiments for predicting the development of the eye and evaluate the best diagnosis and treatment. Finally, the parallel execution( P) builds a bridge between the real world and the artificial world. It realizes the on-line, long-term guidance of the diagnosis process for the real eye, while the information of the artificial eye system is updated to follow the real eye. Many technologies are applied on the parallel eye system research to improve the efficiency of the diagnosis effectively, including knowledge automation, computer vision, knowledge mapping and machine learning. Based on the researches, the long-term, precise and efficient treatment for the human eye can be achieved.