How Machine Learning Is Helping Save the Lives of Cancer Patients in China

Faster and more accurate diagnosis of lung cancer is helping save lives in China.

Dr. Yang Yang is a very busy man. Although he agreed to a video interview, he doesn’t really have time to talk. Cameras and lights are set up in a small brick-walled room at the end of a narrow hallway in the largest pulmonary hospital in Shanghai, but we are unsure when he will show up to discuss how it came about that he is using machine learning to diagnose lung cancer in his patients.

AI Saving Lives in China

Click the button below to load the content from Youtube.

Video by John Hunt

Race Against Time

Lung cancer patients have a very good survival rate if the cancer is diagnosed correctly in its early stages. Because of the large number of patients they screen each day, pulmonary doctors in China find themselves in a race against time.

Here is why: Lung cancer is still the most commonly diagnosed cancer globally and the leading cause of cancer death. Each day in China, nearly 11,000 people are diagnosed with cancer, and lung cancer accounts for the largest proportion. There has been a sharp rise in the incidence of lung cancer over the past 15 years and unless something changes, lung cancer mortality in China could increase by as much as 40 percent by 2030.

A lung cancer screening typically involves making CT scans and then examining the images for known patterns or nodules. It takes a specialist with a very good eye to recognize the patterns early. Lung cancer symptoms are similar to numerous other disorders, so CT scans may not even be ordered. Even if they are, doctors often do not detect the nodules, which are hard to see when the cancer is easiest to treat. It is a classical use case for technology, and that’s where Dr. Yang and his friend Flat Chen enter the frame.

Challenge Meets Opportunity and Argus is Born

Dr. Yang and Chen have both experienced Shanghai’s transformation to the business powerhouse it is today, and their career paths have been linked to this dramatic 30-year rise. After becoming friends during an MBA program in Shanghai, Dr. Yang turned to medicine while Chen started his career in the technology industry with SAP in China. Chen’s drive to apply new technologies to solve real-world business problems has propelled his career.

While meeting over a beer in 2017, Dr. Yang told Chen that he was looking for a faster and more accurate way to screen his patients, and wanted to know if they could join the race to develop a better machine learning solutions for screening CT scans. Chen took the problem back to SAP to see if there was a team that could apply machine learning to the X-ray screening process.

Applying his skills as developer, product manager, and entrepreneur, Chen gathered a team of developers under the leadership of colleague Zion Chen at SAP Labs China. Together they began working on a proof of concept for Dr. Yang. Interest in contributing to the project was high as many employees knew someone who had been touched by lung cancer.

The team developed a prototype that was able to pinpoint signs of lung cancer in an CT scan. They submitted their prototype to the demo day at SAP Labs China and came out on top of the competition, which led to even greater resources and support from the management team at SAP’s development center in Shanghai. The solution known as Argus has now become one of the top contenders in the social entrepreneurship initiative SAP One Billion Lives, which aims to solve the world’s biggest problems.

“Working in SAP provides me with the opportunity to work with a lot of passionate people, addressing one of the world’s biggest pain points,” says Chen.

The Argus prototype has helped Dr. Yang and his colleagues to gain important initial experience applying machine learning to screen patients: “We have a working prototype that has already surpassed my expectations. The solution not only helps us analyze more patients in a shorter period of time, it also improves the accuracy of detection. It’s really powerful. We now trust it, and think it’s the future.”

More collaborations like the one between Dr. Yang and Chen can help doctors more effectively screen more cancer patients to save lives.

Flat Chen is the lead of the Intelligent Enterprise Solution Ecosystem in APJ and Greater China, based at the Shanghai office of SAP Labs China.  


Joanne Chen, head of communications for SAP Labs China, also contributed to this story.