Outline:

• - Introduction of AI in Medical Device Industry
• - AI Technology in Medical Image
• - AI Technology in Clinical Assistance
• - Benefits and Challenges of AI Industry in China

Artificial Intelligence is the 4^th industrial revolution in the world, which is highly expected to add great significance to healthcare industry in every well-developed country, so as China. The healthcare segment is the top priority industry in China government’s plan. In 2019, ShangHai JiaoTong University published “the Whitepaper of AI Healthcare in China”, and addresses 4 main points fields of AI technology: Medical image, Clinical assistance, Medicine development, and Health management. I In current definition of China medical device regulation, only AI-based products used for imaging are identified as “medical device”. In this article, we will focus on how AI engage in medical imaging and clinical assistance, in addition, we discuss their expected benefits to China hospitals.

Figure 1. The Four Main Fields of Healthcare Industry.

AI Technology in Medical Image

AI medical imaging is a device which helps and supports doctors’ image diagnosis, and it covers the second largest pie of the entire global AI medical product market. The field of AI medical image analysis is expected to grow to 2.5 billion USD, covering 25% of total AI medical field. In China, the market for AI medical image analysis reached 9.6 billion CNY in 2016 and a year later reached 13 billion CNY. AI image analyzing products are expected solve following issues where China’s medical field faces: 

1. Lack of specialists/doctors: In China, the size of medical image data increases every year by 30% whereas the number of radiologists increases by 4% every year. Furthermore, the number of pathologists in China is lacking. One out of 70,000 people in China is a pathologist whereas in the United States one out of 2000 people is a pathologist. 

2. High misdiagnosis rate: The lack of doctors causes difficulties in working environment, which easily leads to misdiagnosis. For example, one level-three hospital receives 200 patients of tuberculosis every day and takes 200 to 300 pictures of CT scan for each patient. On average, one radiologist in one level-three hospital must see 40,000 to 60,000 pictures. It is said that in China general misdiagnosis happens at the rate of 27.8% on average. For malignant tumors specifically, misdiagnosis on average happens at the rate of 40%.

3. Image diagnosis takes time: For example, when one patient with a tumor comes up with 200 pictures of CT scan, a doctor needs 3 to 5 hours to detect the tumor, using a traditional method. After finding a tumor, a doctor needs to plan another scanning or come up with a surgery plan, considering the patient’s radioactive exposure.

To comply with rapid development of AI-assisted medical device, China’s Center for Medical Device Evaluation (CMDE) published the regulatory guidance “CMDE 2019 No.7 Order, Guidance of Deep-Learning Medical Devices Assisting on Decision-Making” in 2019. In this guidance, the definition of AI medical device is stated as: “Software with deep-learning technology that assists decision-making, based on medical device data or images that are generated by medical devices.”

Table 1. Chinese Companies in AI Imaging
Company Name	Main Field	Hospitals using their service
YITU Tech	Lung Cancer AI Imaging, Mammography AI Imaging, Breast Ultrasound Imaging	More than 100 AAA hospitals implement their system.
Wuhan Landing Intelligence Medical Co., Ltd	Fully Automatic Neck Cancer Screening, Fully Automatic Pathologic Cell Analysis	More than 400 hospitals all over the country
Huiying Medical Technology Co., Ltd.	Medical Image Cloud Platform, Big Data Analysis with AI Diagnosis Cloud Platform	More than 800 AAA hospitals
Infervision	AI Medical Platform, X-ray and CT Assistance, Tuberculosis Screening	25 top hospitals
Deepwise	AI-Assisting Diagnosis System, Cloud AI-Assisting System	Nearly 100 hospitals
VoxelCloud	Lung Cancer Diagnosis, Eye Fundus Diseases Inspection, Coronary CT Scan System	More than 100 hospitals

Source: [Whitepaper] Artificial Intelligence in Healthcare Industry in China: 2019

 

RECOMMEND READING

- ANALYSIS: Insight on Government Expenditures on Healthcare and Medical Device Trends in Asia - ANALYSIS: China Medical Device Market Overview and Importation Process

 

AI Technology in Clinical Assistance:

            AI diagnosis assistance provides supports in many kinds of aspects: electronic patient record, hospital guide and virtual assistance. In this section, we introduce these applications and the representative companies.

1. Virtual assistant: Patients usually consult the doctors about the healthcare or diagnostic related questions, but this interaction takes many doctor’s off time. The virtual assistant can answer the basic questions regarding the patient records to save many manpower on replying 

2. Electronic Patient Record: electronic patient record takes over hand-written electronic patient record. It helps doctors manage data and transmit it into quantified record of patients. Compared to the traditional patient records, patient’s condition is easier to record by inputting, and using the accumulated data of patients, AI electronic patient record assists diagnosis. 

3. Hospital Guide: In China, overcrowded hospitals are very common. This Ail hospital guide reduces burdens of workers at hospitals by triaging patients and being a medical reception. When this AI hospital guide device improved better, it is expected to take care of financial issues at hospitals.

Table 2. Chinese Companies in AI Clinical Assistance
Company Name	Main Field	Hospitals using their service
Cable Wen Bo know Technology Co., Ltd.	Cloud Patients Record:  tumors, blood, orthopedics, neurology, respiratory system and Psychiatry	400 AAA hospitals implement, and 3000 hospitals are one trial.
Unisound	Patient record by AI voice/sound input system	20 AAA hospitals implement, and 40 hospitals are on trials
SYNYI・AI	Patient record by AI voice/sound input system	20 AAA hospitals implements, and 40 hospitals are on trial.
LinkDoc	Electronic patient record, which assists clinical decisions	<10 hospitals are on trial.
iFLY TEK	Hospital guide	<10 hospitals are on trial.

Source: [Whitepaper] Artificial Intelligence in Healthcare Industry in China: 2019

Benefits and Challenges of AI Industry in China

Overall, China can provide good opportunities to develop AI technology with below beneficial conditions:

• - Big data of patients is available
• - Advanced arithmetic
• - Strong National Policy Support from the Government
• - Huge Capital Investment

Yet, At the same time, manufacturers should face several challenges when accessing  Chinese market.

• Cost: using AI technology in clinical is more costly than conventional diagnosis model, especially in China where healthcare insurance covers the whole nation.
• Infrastructure: Besides well-known urban cities, the infrastructures in most of hospitals are not ready to AI technology. Moreover, the data collected by these “BASIC” instruments at basic hospitals is expected to be low-quality resource for AI database.
• Business model: Medical device segment is a competitive market. China’s top digital players, BAT, has joined Chinese healthcare market few years ago and most of medical instruments are supplied by big medical enterprises, GPS(GE, Philips, Siemens). Finding out profitable business model is a key question for every medical device company.

QUALTECH has assisted several manufacturers to evaluate and plan for their AI-based medical devices to Asia market. Feel free to contact us if your company also has interest on China market.

 

RECOMMEND READING

- ANALYSIS: Insight on Government Expenditures on Healthcare and Medical Device Trends in Asia - ANALYSIS: China Medical Device Market Overview and Importation Process

References:

[Whitepaper] Artificial Intelligence in Healthcare Industry in China: 2019 Version, Issued by ShangHai University

The Guidance of Deep-Learning Medical Devices Which Assists Decision-Making in Clinical