The hottest medical AI is emerging. The expert is

2022-09-22
  • Detail

Medical AI is emerging. Expert: it is a doctor's assistant, but it still faces the bottleneck of development. Image provided by the organizer of this photo conference

medical treatment is an important application field of artificial intelligence. The application of intelligent medical means and technologies such as image-assisted diagnosis, electronic medical records, medical big data, etc. has been emerging in major domestic hospitals

in Shanghai, including the 10th people's Hospital, Ruijin Hospital, Renji Hospital and other comprehensive hospitals, artificial intelligence technology has been introduced since last year

what changes can medical AI bring to the traditional medical industry, and will it impact the career of doctors

at the "first China Medical Imaging AI conference" held in Shanghai from December 14 to 16, Liu Shiyuan, chairman of the conference, chairman of the China Medical Imaging AI industry university research and application innovation alliance, and director of the imaging medicine and nuclear medicine discipline of Shanghai Changzheng Hospital, revealed to pengpai that at present, the application direction of artificial intelligence in the field of medical imaging mainly includes three categories: disease screening, lesion mapping, three-dimensional imaging of organs, helping doctors improve the early diagnosis rate Reduce the misdiagnosis rate

however, he said that this technology still faces many bottlenecks, such as the stability of data needs to be improved, AI products are still slow to obtain the certificate issued by the National Drug Administration Department, and other issues related to medical ethics

many hospitals in Shanghai have applied AI technology

for the field of medical artificial intelligence, there have been many favorable policies in recent years. The outline of the "healthy China 2030" plan clearly points out the need to build a health information service system, and the "new generation of artificial intelligence development plan" recently released by the State Council also proposes to develop intelligent medicine

Zhou Gongwei, deputy director of the data statistics and information center of the National Health Commission, said at the conference that China has ranked among the six "digital health innovation centers" in the world. Surging noted that in recent years, a large number of medical technology companies have emerged in China, mainly focusing on imaging and virtual assistants to provide efficiency improvement tools for hospitals

in Shanghai third class a hospital, the application of AI has also increased in the past two years, especially in AI follow-up, AI film reading, AI tumor diagnosis and so on

for example, in August last year, "Watson tumor intelligent joint consultation system" was fully launched in Tongji University Cancer Center and Shanghai Tenth People's hospital. "Watson" is a medical artificial intelligence developed by IBM, which has been clinically applied in 14 countries before

after the doctor inputs the basic information of the patient, cancer stage, local recurrence, chemotherapy scheme, pathological stage, cancer metastasis and other information to "Dr. Watson", as long as more than ten seconds, "Dr. Watson" will give a treatment plan with high buoyancy, including the recommended use plan, the use plan that can be considered, and the use plan that is not recommended

in May this year, Ning guangtuan, an academician of the Chinese Academy of engineering and vice president of Shanghai Ruijin Hospital, should establish traceable management accounts for key raw and auxiliary materials and product sales accounts. The "ruining sugar aid" Ai doctor jointly developed by the team and Alibaba health AI laboratory also attracted national attention

in June, Renji Hospital Affiliated to Medical College of Shanghai Jiao Tong University officially launched AI follow-up assistant in the daytime operation ward, which can follow up patients after discharge, freeing medical staff from tedious work

in July, the first batch of intelligent mobile effis telemedicine robot system was put into use in Ruijin Hospital Affiliated to Jiaotong University Medical College, which can realize remote diagnosis and training

Liu Shiyuan believes that AI brings more freedom and possibility to doctors

expert: still facing the bottleneck of development

Liu Shiyuan, chairman of the conference, chairman of the China Medical Imaging AI industry university research and application innovation alliance, and director of the imaging medicine and nuclear medicine discipline of Shanghai Changzheng Hospital, said that when big data encounters artificial intelligence, the most promising breakthrough is in the field of medical imaging

"ai+ medical imaging" is a cutting-edge breakthrough technology that artificial intelligence has made in the field of image recognition, which is applied to the field of medical imaging, so as to improve the diagnostic efficiency and balance the material cost with the performance requirements and accuracy

Liu Shiyuan said that at present, the application direction of artificial intelligence in the field of medical imaging mainly includes three categories: disease screening, lesion delineation, three-dimensional imaging of organs, and has covered many diseases such as breast cancer, cardiovascular, skin cancer, pulmonary nodules and so on

"AI can help doctors improve the rate of early diagnosis and reduce the rate of misdiagnosis, which has practical applications in the fields of lung cancer screening, fracture, optic membrane disease and pathological diagnosis. There are also medical multidimensional measurement. Multidimensional quantification is the premise of precision medicine, and it can accurately understand the characteristics and focus of the disease." Liu Shiyuan said

but while embracing technology, Liu Shiyuan also believes that there are many problems with medical AI imaging products. First, most of the current imaging AI products are based on the model formed by the image annotation of a single disease, and there is no product that conforms to the clinical practical scene. In addition, AI products are often concentrated in a few diseases, which is difficult to cover all medical imaging problems

he also pointed out that AI brings many medical ethics problems, "if there are serious consequences such as misreading and omission during the follow-up of AI, resulting in the delay of the patient's condition, who should bear the relevant, AI research and development enterprises, medical institutions or doctors."

Liu Shiyuan also stressed that AI is a new category of medical management, and it is unclear how to define it, how to grade it, and how to test it. At present, the National Drug Administration has not issued the corresponding registration certificate for AI products, and clinical applications are carried out in the form of scientific research cooperation. After obtaining the registration certificate, how to enter the business model and how to charge for the products are still an unresolved problem

but as for "whether AI products can replace doctors", he pointed out that in principle, AI's big data, memory and computing capabilities will surpass human beings. Machines can indeed replace some repetitive and low-tech jobs of human beings, but the level of neural network development determines that machines cannot replace human beings, and human ethics, expression, communication and other abilities will not be replaced by machines

(original title: medical artificial intelligence is emerging, expert: he is a doctor's assistant but still faces the bottleneck of development)

Copyright © 2011 JIN SHI