The greatest contribution that artificial intelligence could make to humanity might be in health care. According to the consultancy firm Frost & Sullivan, Artificial intelligence’s has the potential to improve medical treatment outcomes by 30-40 percent and reduce costs by as much as 50 percent.
This is particularly important for China, with its population of 1.4 billion people. Medical services can be scarce in China’s rural areas while, in urban areas, services are highly strained due to the sheer volume of patients. According to the latest data from the Organisation for Economic Cooperation and Development, China has 1.8 practicing doctors per 1,000 people, compared with 2.56 for the United States and 5.1 for Australia.
Adding further stress to China’s health care system is its aging population. According to the United Nations, China is aging more rapidly than almost any country in the world, due mainly to its previous one-child policy. By 2050, China’s population over 65 will reach around 330 million, roughly the current total population of the US.
China is not alone in its search for better health through AI. In a recent study by Accenture, the AI healthcare market in the US is predicted to reach US$6.6 billion by 2021 compared to US$600 million in 2014. However, China already has a smart healthcare strategy, an integral part of its overall AI strategic plan released in July 2017.
The plan calls for the development of a whole gamut of AI-related healthcare technologies, such as intelligent diagnosis, wearables, AI health monitoring, robot-assisted surgery, intelligent medical image recognition and medical genomics with a strong emphasis on elderly care.
Last year, the China Food and Drug Administration included AI diagnostic tools on its list of permitted medical devices. In May, China established a national Chinese Intelligent Medicine Association as a platform for research, exchange, and cooperation in AI for health care.
I believe there are three key areas: deep learning to analyze medical images, cognitive computing to capture and apply medical knowledge and AI analytics to provide continuous health monitoring.
The medical profession is particularly well suited to the use of Artificial intelligence’s. Medical doctors rely greatly on perceptual senses, like vision and hearing, to gather information about patient health. Artificial neural network approaches such as deep-learning are ideal for exactly this type of work.
For example, Google is experimenting with deep-learning in retinal images to provide early detection of diabetic retinopathy with accuracy on a par with experts. In China, researchers have used AI on eye scans to diagnose congenital cataracts as accurately as human doctors.
Using AI deep-learning to process medical images, such as CT scans and X-rays, is particularly hot among China’s start-ups. Radiology departments at top Chinese hospitals routinely handle tens of thousands of scanned images per day. AI deep-learning is already used to analyze and highlight abnormalities. Among big players, Alibaba’s health unit uses AI to interpret CT scans, and Tencent’s Mining uses AI to detect early signs of cancer.
The other skill important in medicine is the ability to learn, recall and apply vast amounts of medical textbook knowledge and keep up to date with the newest medical research/journals and pharmaceutical products. AI can use natural language processing and machine learning to read and understand millions of online documents, as well as millions of data points to help diagnose and recommend treatment.
Researchers in China are also using AI to capture general medical knowledge. For example, iFlyTek and Tsinghua University successfully created an AI system that not only passed last year’s Chinese medical licensing exam but also scored better than 96 percent of exam takers. The exam not only tested breadth of knowledge but also able to understand the intricate connections between facts and use them to make decisions.
Because of more affordable health care wearables that track activities and heart rate, consumers are taking responsibility for monitoring their own health. According to Tractica’s forecast, annual wearable device shipments will increase from 118 million units in 2016 to 430 million units by 2022.
This increased use of wearables means a lot of daily health data will be available online. Big data and AI predictive analytics can continuously monitor and alert users of abnormalities, and before the outset of more major medical problems.
Insurance companies, such as China’s Ping An Health, are starting to integrate wearables into their offerings. For example, customers who live healthier lifestyles get points and rewards. Wearables provide insurance companies with vast amounts of highly valuable customer biometric data, which can be used with AI to offer continuous monitoring and healthcare advice and also provide discounts to those with healthier lifestyles.
AI relies greatly on data for machine learning and predictive analytics, and China has no shortage, with its population generating massive amounts of real-time medical data. The Chinese population is eager to use technology and adopt AI. China is also unique in its approach to health care that leverages both Western and traditional Chinese medicine.
With an increased use of AI, combined with readily available medical and biometric data, China is on its way to providing quality personalized health care to more people at a lower cost, while keeping people healthier through continuous monitoring and alerts. With fewer people getting sick, the workloads for hospitals and medical staff will be reduced. A healthy nation is a wealthy nation, as the saying goes.
Dr. Andy Chun is an associate professor at City University of Hong Kong and convenor of the AI Specialist Group, Hong Kong Computer Society