Ping An Technology AI predicts flu outbreaks with 90% accuracy

1 min read

Ping An Technology’s intelligent disease prediction and screening models can now forecast the likelihood of an influenza outbreak with a more than 90% accuracy, according to a South China Morning Post report

Why it matters: Accurate and timely predictions of infectious disease outbreaks can be valuable to authorities working to prevent and control them more efficiently—ultimately saving lives.      

  • The technology arm of Ping An Insurance, one of China’s biggest insurers, Ping An Technology is also working on predictive models for diseases like diabetes and chronic obstructive pulmonary disease, which can provide insight into their epidemiology on a regional level. 

“Disease prevention and control has been a high priority for national health systems across the world. If we can spot the risk at a preliminary stage and make early treatment, it will not only alleviate the suffering of patients but also lower overall medical expenses.”

—Xiao Jing, chief scientist at Ping An Group

Details: The machine-learning based prediction models are currently exclusive to Shenzhen and Chongqing, and leverage environmental data, case reports from local hospitals, and historical data from regional health authorities.      

  • Dubbed the world’s first “macro-micro” method of disease prediction through AI and big data, the models can make an accurate prediction up to a week before an outbreak, the company said in a 2018 statement.
  • They can then be sent be sent as digital alerts to city authorities. 

Context: This isn’t the first time Ping An has explored the potential of AI in medical technology: Ping An Good Doctor, which was spun off from the insurance conglomerate in 2014, develops various AI-powered products, including telephone booth-sized clinics that collect a patient’s health data for a human doctor to review. 

  • Other companies, like YITU Technology and Insilico Medicine, are using AI to streamline medical processes that can be both time-consuming and technically difficult for humans to accomplish. 
  • YITU Technology’s AI cancer screening tool uses computer vision to quickly make diagnoses and recommend treatments.
  • Insilico Medicine’s new GENTRL tool promises to speed up and reduce costs for the slow and expensive drug development process by identifying new treatment candidates with AI similar to Google’s DeepMind.