AI can solve China’s doctor shortage. Here’s how

If there is one country that has invested heavily in healthcare reform over the last few years, it is China. But as its population grows older, with already 300 million people suffering from chronic diseases, it seems almost impossible to keep up with the soaring demand for healthcare. According to the latest data from the Organisation for Economic Cooperation and Development, China has 1.8 practising doctors per 1,000 citizens, compared to 2.6 for the United States and 4.3 for Sweden. Can artificial intelligence (AI) relieve China’s overworked doctors of some of their burdens?

China’s ailing healthcare system

The hard-working medical professionals who keep China’s ailing healthcare system running could certainly use a helping hand. Overcrowding is the order of the day in the country’s urban hospitals, with a typical outpatient department in Beijing seeing about 10,000 people every day. The problem is exacerbated by the scarcity of medical facilities in rural areas, which causes people to flock to hospitals in nearby cities.

As the Future Health Index 2018 by Philips shows, the relatively low number of skilled healthcare professionals in relation to the size of the population is one of the main reasons why access to care in China lags behind most of the other fifteen countries surveyed.

Demographic projections give further reason for concern. The demand for care will only continue to grow as China is aging more rapidly than almost any country in the world. The United Nations estimate that by 2040, the country’s population over 65 will reach about 303 million, which is almost equal to the current total population of the United States.

However, there is also reason for optimism.

In its commitment to offer accessible and affordable care for all, the Chinese government is spearheading the development of healthcare technologies. And perhaps the most promising is AI.

The rise of AI

AI can help to make sense of large amounts of data, fueled by computing power that has risen dramatically over the last few years. That’s why China offers particularly fertile ground for AI development: with its 1.4 billion population, the country sits on massive troves of data.

Recognizing the country’s AI potential, the government has set out an ambitious plan to turn China into the world’s leading AI innovation center. Healthcare is one of the industries that is set to benefit from multi-billion dollar investments in startups, academic research, and moonshot projects.

This is not merely a vision, but a reality already in the making.

In 2017, the city of Guangzhou opened the first clinic in China to diagnose cataracts – an eye disease and major cause of blindness – with the help of AI. The tool is expected to save time by directing patients toward the best specialist for their needs, allowing more patients to be treated. Although further clinical trials are needed, it is a promising development, given the shortage of eye care practitioners in China.

Many similar initiatives are underway. According to Yiou Intelligence, a Beijing-based consultancy firm, some 131 Chinese companies are currently working on applying AI in healthcare.

A smart personal assistant for physicians

Speeding up the screening of medical images is just one of the ways in which AI could relieve China’s overburdened healthcare system.

As one Chinese radiologist said in an interview with the New York Times: “We have to deal with a vast amount of medical images everyday. So we welcome technology if it can relieve the pressure while boosting efficiency and accuracy.”

We should take these needs to heart, and focus on developing intelligent applications that ease the workload for physicians while improving outcomes for patients. Crucially, the goal should not be to replace physicians, but to augment them in their daily work, strengthening their role in the delivery of efficient and high-quality care.

For some, AI conjures up images of autonomous robots replacing human workers. But I believe that in healthcare, AI is best thought of as a smart personal assistant for physicians, which adapts to their needs and ways of working (‘adaptive intelligence’, as we call it at Philips). Viewed through that lens, AI will make healthcare more – not less – human.

Today, AI is already helping physicians with the analysis of medical images. As AI becomes increasingly sophisticated and is integrated with medical knowledge, it could support ever more precise diagnosis and personalized treatment plans. But in the short term, arguably the greatest gains are to be made in solving operational bottlenecks in hospitals – for example, by helping physicians get a quick overview of all clinically relevant information on a patient.

Patient data are usually stored in many disparate systems and formats. At Zhongshan Hospital in Shanghai, it can take a physician up to 20 days to manually extract all relevant information from 200 unstructured medical reports into one structured format.

By combining AI methods like natural language processing and machine learning with clinical knowledge, it is possible to collate all clinically relevant information in one dashboard. Physicians have to spend less time on capturing information from unstructured reports, and less time sitting in front of a screen to get a complete picture of the patient.

Improving care close to people’s homes

AI could also enable patients with chronic conditions to become better informed about their health and stay connected with professional caregivers.

According to the Future Health Index, adoption of telehealth in China is currently much lower than the 16-country average, but the Chinese population is open to the use of technologies that can supplement the care they currently receive.

For example, home health monitoring technology powered by AI could help the frail and elderly stay connected with professional caregivers, to ensure they receive timely care when needed. People with diabetes or hypertension could benefit from similar technology that allows them to track their condition via clinically validated sensors and devices.

Such initiatives would fit perfectly with the Chinese government’s ambition to improve care at the grassroots level to counter congestion in city hospitals. More widespread adoption of AI technologies should go hand in hand with investments in primary care facilities and internet connectivity in rural areas – making healthcare more equally accessible and affordable, and allowing people to enjoy a better quality of life close to their communities.

Looking further ahead, AI could also become pivotal in addressing lifestyle-related diseases such as obesity – a major health concern that affects about 1 in 8 people in China. Imagine people with high risk of obesity getting bespoke lifestyle tips via their smartphone. On a population level, data analyses could inform public interventions targeted at specific age groups or geographic areas. That is how big data could have big impact. As the Chinese government has outlined in its plans for a ‘Healthy China 2030’, the focus of the healthcare system will increasingly shift from treatment to prevention.

A call for collaboration

How to accelerate this journey towards more efficient, accessible and preventative care?

First, building a more robust data ecosystem should be a top priority. The quality of AI is only as good as the quality of the data you feed into it. China’s healthcare system would benefit from shared data standards, interoperability of systems, and improved data exchange; protected by top-notch security measures. The establishment of three digital national databases with health information by 2020 is an important step in this direction.

Second, data-driven approaches such as AI will only have the desired impact when combined with proven medical expertise. AI is only part of any solution; it is never a solution by itself. A deep understanding of the clinical context is indispensable. Any form of AI-assisted care must be centered around the physician and the patient, taking their needs as a starting point, and building on the wealth of human knowledge that is already available.

Third, AI-enabled tools must be rigorously tested against the highest regulatory standards. In healthcare, where lives are at stake, we need to deploy new technologies wisely and carefully. Only with proper clinical validation can we ensure responsible, safe and effective use of AI. Physicians as well as patients also require education on a tool’s strengths and limitations.

Fourth, collaboration between academia, startups, and established companies is of paramount importance. The challenges in China’s healthcare system are simply too big for any player to go at it alone. In this light, it is encouraging that the Chinese government has recently founded a collaborative platform to promote the exchange of ideas and kick-start new projects in intelligent medicine.

Finally, to ensure we are creating a future-ready healthcare system in China, we must address the shortage of talent at the intersection of medicine and data science. We should nurture and invest in developing people who combine medical know-how with a firm understanding of AI and other technologies. Ultimately, the sustainability of China’s healthcare system may lie in their hands.