Department of Human and Engineered Environmental Studies, Graduate School of Frontier Sciences, The University of Tokyo, Chiba, Japan.
Institute for Global Health Policy Research, National Center for Global Health and Medicine, Tokyo, Japan.
Biosci Trends. 2020 Nov 4;14(5):314-317. doi: 10.5582/bst.2020.03364. Epub 2020 Oct 25.
Fifth Generation (5G) mobile communications technology became available in Japan as of the end of March 2020. The Ministry of Internal Affairs and Communications (MIC) is proceeding with a plan to use 5G for a doctor-to-doctor remote diagnosis system. This remote diagnosis offers patients the benefit of receiving advanced medical care without having to travel long distances. The provision of a remote diagnosis will provide elderly patients in rural areas with an earlier diagnosis without burdening patients in Japan where the aging population and the uneven distribution of doctors are increasing. However, the system will increase the burden on specialists by expanding the doctor's catchment area. As a solution to that problem, deep learning-based artificial intelligence (AI) is expected to reduce the burden on doctors. In order to realize 5G- and AI-based real-time diagnostic support, diagnostic imaging using AI and an AI model that provides instructions are required. This is because ultrasonography and endoscopy, which can be used for remote diagnosis, do not acquire data on fixed areas like a CT or MRI scan. The AI model needs to instruct the doctor at the patient's home in order to collect appropriate information in accordance with the patient's symptoms and status. In order to build an interactive AI model, the interactions between doctors who are making a remote diagnosis should be recorded as training data and a 5G-based remote diagnosis system should be created. A remote diagnostic support system incorporating 5G and interactive diagnostic imaging incorporating AI will result in a system that places less of a burden on patients and doctors.
第五代（5G）移动通信技术于 2020 年 3 月底在日本投入使用。总务省（MIC）正在推进一项计划，将 5G 用于医生之间的远程诊断系统。这种远程诊断使患者无需长途跋涉即可享受先进的医疗服务。提供远程诊断将为农村地区的老年患者提供更早的诊断，而不会给日本的患者带来负担，日本的人口老龄化和医生分布不均问题日益严重。然而，该系统将通过扩大医生的服务范围来增加专家的负担。为了解决这个问题，基于深度学习的人工智能（AI）有望减轻医生的负担。为了实现基于 5G 和 AI 的实时诊断支持，需要使用基于 AI 的诊断成像和提供指令的 AI 模型。这是因为可用于远程诊断的超声和内窥镜不能像 CT 或 MRI 扫描那样获取固定区域的数据。AI 模型需要根据患者的症状和状况，指导患者家中的医生收集适当的信息。为了构建交互式 AI 模型，应将进行远程诊断的医生之间的交互记录为训练数据，并创建基于 5G 的远程诊断系统。结合 5G 的远程诊断支持系统和结合 AI 的交互式诊断成像将减轻患者和医生的负担。
