１１月２日，２０２０粤港澳院士峰会暨第六届广东院士联合会年会专题活动“ＡＩ＋医疗健康领袖峰会”在东莞举行。与会专家认为，人工智能医疗是医疗卫生技术和通用信息技术深度融合发展的产物，是医疗卫生领域未来发展和转型的方向。

“医疗智能化现在已经逐渐形成一种新的医疗模式。”中国工程院院士、国家新一代人工智能战略咨询委员会组长潘云鹤指出，“这个医疗模式首先是从智能设备开始的。智能设备如果和服务连在一起就变成一个云平台，云平台可能把大医院、中小医院、病人和居家使用者联系在一起，实现智能设备共享。将数据存在该云平台里，同时为中小医院和大医院进行服务。”

潘云鹤进一步阐述：“知识云平台可以和科研教学的知识结合起来，比如皮肤病的研究、教学等。云平台实际上是不断发展的教科书，还可以把药店、急救车连在一起，形成一个新的以知识为中心的医疗系统。”

在中国科学院院士、清华大学人工智能研究院院长张钹看来，医疗健康的智能化势在必行，但人工智能在医疗健康上的应用还存在不安全、不可信、不可靠和不可扩展等风险。“第三代人工智能是今后的发展方向，即建立可解释和鲁棒的人工智能理论，发展安全、可信和可扩展的人工智能技术，推动人工智能在医疗健康中的应用。”

人工智能在医疗器械领域有哪些应用？“从医疗器械的角度来说，几乎每一个领域、每一个类别都和人工智能有关系，比如有源类的医疗器械，检验的、手术的、可穿戴的这些装备，还有一大类是无源类的各种植入体，像人工关节、人工器官、心血管支架等，由于人工智能技术的发展，这些植入体的功能、性能都发生了非常大的变化，可以说是跨越式变化。”中国工程院院士、华南理工大学原校长王迎军认为，人工智能正在推动医疗器械行业的革命性转变，ＡＩ与医疗器械两个多学科交叉领域的深度融合发展，需要跨学科团队整体建设，包括人才培养以及深度融合的课题研究等。

“虽然我们关注于人工智能应用于临床诊断、治疗，即ＡＩ如何提高医生诊断效率，但不容忽视的还有用于训练ＡＩ的数据。这些数据应该更加完并定期更新，对于病人而言，信息的点、面以及长期  追踪很重要。”广东省人民医院院长余学清从临床角度指出，“我们需要在科学家协助下，获得完整、正确、快速的数据，但同时保障数据安全。”

对此，王迎军指出，ＡＩ＋医疗健康本身是一件医工交融的事情，需要医工紧密结合的团队来做。她强调：“无论是现代医学还是人工智能、医疗器械的发展，都需要一个更紧密的多学科团队、更早的融合，而不是你来做，我来用。”

来源：科技日报

On November 2, the 2020 Guangdong Hong Kong Macao academician summit and the sixth annual meeting of the Guangdong academician Federation "Ai + healthcare leaders summit" was held in Dongguan. Experts at the meeting believed that artificial intelligence medicine is the product of the deep integration and development of medical and health technology and general information technology, and is the direction of future development and transformation of medical and health field.

"Medical intelligence has gradually formed a new medical model." Pan Yunhe, academician of the Chinese Academy of engineering and head of the national new generation artificial intelligence strategy advisory committee, pointed out that "this medical model starts with intelligent devices first. If smart devices are connected with services, they will become a cloud platform. Cloud platforms may connect large hospitals, small and medium-sized hospitals, patients and home users to realize intelligent device sharing. The data will be stored in the cloud platform to serve small and medium-sized hospitals and large hospitals at the same time. "

Pan Yunhe further elaborated: "the knowledge cloud platform can be combined with the knowledge of scientific research and teaching, such as the research and teaching of skin diseases. The cloud platform is actually a textbook of continuous development. Finally, pharmacies and emergency vehicles can be connected together to form a new knowledge-based medical system. "

According to Zhang Bin, academician of Chinese Academy of Sciences and President of Artificial Intelligence Research Institute of Tsinghua University, the intellectualization of medical health is imperative. However, the application of artificial intelligence in medical health still has risks of insecurity, unreliability, unreliability and non scalability. "The third generation of artificial intelligence is the future development direction, that is to establish interpretable and robust artificial intelligence theory, develop safe, credible, reliable and scalable artificial intelligence technology, and promote the application of artificial intelligence in medical and health care."

What are the applications of artificial intelligence in the field of medical devices? "From the perspective of medical devices, almost every field and category is related to artificial intelligence, such as active medical devices, inspection, surgical and wearable equipment, and a large category of passive implants, such as artificial joints, artificial organs and cardiovascular stents. Due to the development of artificial intelligence technology, the functions of these implants are improved Great changes have taken place in energy and performance, which can be said to be leapfrog changes. " Wang Yingjun, academician of the Chinese Academy of engineering and former president of South China University of technology, believes that artificial intelligence is promoting the revolutionary transformation of the medical device industry. The in-depth integration and development of AI and medical device two interdisciplinary fields requires the overall construction of interdisciplinary teams, including personnel training and deep integration research.

"Although we focus on the application of artificial intelligence in clinical diagnosis and treatment, that is, how AI can improve the diagnostic efficiency of doctors, we should not ignore the data used to train AI. These data should be more complete, accurate and regularly updated. For patients, the point, area and long-term tracking of information are important. " Yu Xueqing, President of Guangdong Provincial People's Hospital, pointed out from a clinical point of view that "we need to obtain complete, correct and fast data with the help of scientists, but at the same time ensure data security."

In this regard, Wang Yingjun pointed out that AI + medical health itself is a matter of integration of doctors and workers, which needs a team of doctors and workers to work closely. "Whether it is the development of modern medicine, artificial intelligence and medical devices, we need a closer multidisciplinary team and earlier integration, rather than you do it and I will use it," she stressed

Source: Science and Technology Daily