See You/China Daily
For radiologist Hou Dalun, artificial intelligence has brought about major changes in his daily work.
Mr. Hou, head of the medical imaging department at Beijing Chest Hospital, said that since introducing the AI system in 2020, he has seen a significant improvement in the accuracy and efficiency of the department’s work.
This system helps radiologists analyze the results of computed tomography scans of lung nodules.
After years of refinement, the system can now analyze CT scans of a person’s lungs, often containing 300 to 600 slices, “within seconds” and provide accurate information about the size, shape and density of nodules, Hou said.
A radiologist will further evaluate the system’s initial results and make a recommendation as to whether surgery is necessary.
Before the system was used, it took more than 20 minutes to examine every slice of a patient’s lung CT scan, Hou said.
He said the system allows the department to complete interpretations of lung nodule CT scan results for about 600 patients per day.
Hou’s experience reflects part of China’s continued efforts to strengthen the application of AI in the medical field. Earlier this month, the country published guidelines to promote and regulate the integration of AI and healthcare.
The guidelines, issued by the National Health Commission Directorate General and four other departments, set out several goals, one of which is to make AI technologies, such as AI-assisted image diagnosis, available in most hospitals at the secondary level and above by 2030.
Chen Kuan, founder and chairman of the board of InferVision, a Beijing-based company that makes AI products to support diagnostic imaging and surgery, said the situation in China is “very good” for achieving the goal, as the government is encouraging hospitals to use AI technology.
The guidelines state that outstanding hospitals will be selected to collect high-quality medical imaging data, conduct research, and support the development and iteration of AI models at scale.
The quality of data and its annotations is the basis of the performance of AI medical devices, Chen said, adding that the products mimic the capabilities, standards and past experience of leading medical experts.
The guidelines outlined 20 key tasks to deepen the application of AI and also emphasized the need to leverage AI technology at the grassroots level.
It called for the development of AI applications to help primary care physicians diagnose and treat common illnesses, review prescriptions, and manage follow-up care.
The guidelines also call for smart management of chronic diseases and efforts to support residents in managing their own health.
Beiqijia Regional Medical Service Center, a primary care provider in Beijing, is one of the grassroots hospitals leveraging the power of AI to improve the quality of public health services.
The company’s Diabetes Collaborative Care Clinic uses an online system to track blood sugar levels for more than 800 diabetic patients.
When the system detects abnormal blood sugar levels, it alerts the patient and provides health management suggestions.
In addition, patients can upload photos of their meals to the system, which evaluates whether the patient’s diet is healthy and provides recommendations to improve their eating patterns and the order in which they eat their food.
Liu Huan, an engineer at the company who developed the system, said these features were made possible thanks to the large-scale language and multimodal model.
Du Sai, director of the clinic, said the system ensured comprehensive health management and timely intervention for patients while reducing the burden on primary care physicians.
Radiologist Hou said AI could play a positive role in the prevention and control of lung cancer, which is considered the most common cancer in China.
Hospitals in developing regions with a shortage of radiologists could use AI technology to perform rapid analysis of CT scan images, Hou said, adding that interim reports generated by AI could serve as a valuable early warning for those who need further intervention.
Wen Naijie, director of Beiqijia Regional Health Service Center, said that it is very necessary for grassroots hospitals to further introduce AI services to improve diagnosis and treatment capabilities.
To lay a solid foundation for the promotion of AI in the medical field, the guidelines called on relevant authorities to improve related infrastructure, computing power and algorithms, facilitate the supply of richer medical data, and strengthen the construction of pilot-scale testing bases.
The guidelines also emphasized the need to strengthen oversight of AI technology, from research and development to application.
The guidelines say stricter regulations are needed to ensure the quality and safety of AI medical services and protect people’s privacy and data security.
The National Health Commission added that the guidelines follow the principle that AI is an enabler rather than a substitute, and that such technology should be safe, reliable and under control.
Radiologist Hou agreed. He said the AI system the department uses to detect lung nodules is very efficient but has weaknesses.
He said the system sometimes sends false alarms and cannot estimate the risk of nodules as accurately as a radiologist, adding that the interim report it generates must be reviewed and approved by a doctor.
He said the system was and will continue to be an adjunct to radiologists, even if its shortcomings are successfully addressed, and stressed that AI should not replace human doctors in medical fields where “people’s lives are at risk.”
Currently, AI technology has limited ability to deal with “edge cases,” or unusual or extreme situations, said Chen, who noted that the technology company cannot replace the inspiration and creativity of human healthcare workers.
AI is like “a child that continues to grow but needs adult supervision,” he said.
China is stepping up efforts to respond to the accelerating evolution of AI and harness its potential.
The General Office of the National Health Commission and two other departments released a document last November listing 84 medical and health scenarios in which AI should play a bigger role, including image diagnosis, surgery, clinical consultation, medication, and medical insurance.
Zhu Keli, founding director of the China New Economic Research Institute, said the high-quality development of AI tools requires joint contributions from various stakeholders.
Zhu said the government needs to play a leading role by formulating long-term and favorable policies, building public data platforms, breaking down data barriers and strengthening regulations.
He said research institutions need to focus on improving algorithms and training models, and high-tech companies need to develop practical and easy-to-use products according to users’ needs, calling for cooperation among hospitals, research institutions and companies.
“The future of AI application in healthcare is promising. AI will become the core driving force to level up the industry,” Zhu said.
wangqingyun@chinadaily.com.cn
