Ping An AI Disease Prediction Paper First Published on EBioMedicine of the Lancet

SHENZHEN, China

On September 21, 2019, “Forecasting Influenza Activity Using Self-Adaptive AI Model and Multi-Source Data in Chongqing, China”, jointly published by Ping An Technology, Chongqing Municipal Center for Disease Control and Prevention, Third Military Medical University, and Tsinghua University, was published on EBioMedicine of The Lancet, one of the four major international medical journals.

EBioMedicine is a publication focusing on human health and diseases, established by The Lancet and Cell. In June of this year, the latest impact factor of EBIOMEDICINE published by JCR (Journal Citation Reports) has reached 6.680, surpassing that of many prestigious medical journals. Its impact factor kept rising every year.

The Intelligent Disease Prediction Project, jointly established by Ping An Technology and Ping An Smart City, utilizes Artificial Intelligence technology to develop disease prediction models to assist in disease prevention and control at the city level. This project has been applied in Chongqing and Shenzhen, and won the IDC (International Data Corporation) 2019 Asia Pacific Smart City Award as the only winner in the Asia Pacific medical and health category.

The research team utilized multi-source data from Chongqing, including the health system’s influenza surveillance data, such as the percentage of influenza-like illness (ILI%), as well as exogenous weather data and public opinion data. The research integrated various cutting-edge artificial intelligence algorithms to innovatively construct a Self-adaptive AI Model (SAAIM) to retrospectively forecast the overall influenza epidemic trends in Chongqing from 2014 to 2017 and made an online prediction of influenza activities in Chongqing in 2018. The influenza activity can be accurately predicted one week in advance, and the validity of the model is verified through its practical applications.

On theoretical research, the model can accurately capture the irregular seasonal trends and analyze the influenza related features since SAAIM integrates multiple AI algorithms and time-shift parameters.

On practical applications, a timely and reliable influenza prediction model enables medical and health institutions to better prepare in advance for upcoming influenza epidemics, enhance the efficiency of disease prevention and control, and reduce the financial burden of government and individuals.

Smart City Disease Prediction Project

Collaborating with Chongqing Municipal Health Commission, Chongqing Municipal Disease Control Center, and Shenzhen Center for Disease Control and Prevention (CDC), the Smart City Disease Prediction Project makes full use of the big data technology, combines massive city-level data collected on the government platform, including weather data, public opinion, business experience and expert knowledge of local disease prevention and control, to build a disease prediction model. The accuracy of infectious disease and chronic disease prediction models has reached 90%.

In 2017, under the guidance of Chongqing Health Commission, Ping An and Chongqing Municipal Center for Disease Control and Prevention jointly released the first influenza and hand-foot-mouth disease prediction model based on big data and AI technology and the screening model for chronic obstructive pulmonary disease (COPD) risk factors. Since the application of the influenza prediction model in 2018, influenza epidemic trend can be predicted one week in advance with an accuracy of over 90%. It is the first infectious disease prediction model which has been put into use in China. The accuracy of the COPD risk screening model has reached 92%. Since December 2016, Ping An Technology and the Institute of Pathogenic Biology of Shenzhen CDC have jointly established an influenza prediction team to apply AI and big data technology to the Shenzhen Influenza Index, enabling the index to predict the prevalence of influenza in Shenzhen with an accuracy of over 90%.

Smart City Disease Prediction Project is a fruit of team PADIA (Ping An Decision Intelligence Assistant), led by Dr. Jing Xiao, Chief Scientist of the Ping An Group, and Dr. Liang Xu, one of the global “Innovators under 35” by the MIT Technology Review. PADIA is comprised of top overseas professionals includes alumni from top universities including Carnegie Mellon University, Oxford University, Cornell University and Tsinghua University. To this day, the team has applied for more than 150 national patents and has undertaken various governmental projects.

View source version on businesswire.com: https://www.businesswire.com/news/home/20190927005166/en/