2020 AI Healthcare Report

REPORT BY
Innovation Centre Denmark
China
AI healthcare
September 2019
Contents
PREFACE 3
1 AI POLICY DEVELOPMENT 4
2 HEALTHCARE AI IN CHINA 7
2.1 AI medical imaging 8
2.1.1 Painpoints 8
2.1.2 Application scenario 8
2.1.3 Advantages of development in China 9
2.1.4 Current industrial status and key players 9
2.2 AI assisted diagnosis 11
2.2.1 Painpoints 12
2.3 AI drug development 13
2.3.1 Painpoints 14
2.4 AI health management 15
2.4.1 Painpoints 15
2.5 AI disease prediction 17
2.5.1 Painpoints 17
2.5.4 Current industrial status key players 17
2.6 Key stakeholders 18
2.6.1 Governmental and academic key stakeholders 18
2.6.2 Industrial Key Stakeholders 20
3 HEALTH DATA 22
4 CONCLUSION AND STUDY TRIP 2020 24
REFERENCE 25
A PARTNERSHIP BETWEEN
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE PREFACE 2
AND THE MINISTRY OF FOREIGN AFFAIRS OF DENMARK
Preface
DENMARK recently released the ‘Danish Strategy for AI’, which aims to support the ability of Danish
business and industry to develop and offer world-class AI services to the domestic and international markets.
Healthcare is one of the prioritized areas.
CHINA is also placing strategic focus on AI in healthcare. Primarily, as a means to solve its domestic
problems related to the unbalanced distribution of healthcare resources and rise in non-communicable
diseases. The municipal government in Shanghai plans to invest $15 billion, more than many national
governments, demonstrating the strong drive for innovation.
Against this background, it is clear that China is not only an attractive market for AI healthcare because of its
market size, but also because of the vast amount of accessible health records, which can be used to develop
advanced algorithms.
With two main reasons, this report is drafted to give Danish audience an in-depth analyse of each segments
in AI healthcare in China, i.e. AI imaging, AI assisted diagnosis, AI disease prediction, AI health management
and AI drug development. Firstly, we lack the specialised knowledge to help experts working in specific use-
cases and different areas AI healthcare. Secondly, innovation and disruptive technology development in AI
healthcare is largely driven by data. To access patient data for research and product development,
identifying the right clinical partner and have close cooperation with hospitals is a must-have in China.
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE PREFACE 3
1 AI Policy development
Since 2009, China’s artificial intelligence (AI) policy has undergone five stages with changing keywords
which indicate the different priorities in different stage. The focus shifts from basic research in such
categories as Internet of Thing (IoT), information security and database in the early period, to big data and
infrastructure in the middle period, to AI itself and also intellectual property protection after 2017. In general,
China’s AI policy mainly focuses on six categories: “made in China”, innovation-driven development, IoT,
Internet+, big data, and scientific and technological R&D.
In July 2015, State Council issued Guiding Opinions on Vigorously Advancing the "Internet +" Action, which
for the first time prioritizes AI as one of the key tasks. The 13th Five-Year National Science and Technology
Innovation Plan called for China to seize the “high ground” in international scientific development, launching
a series of fifteen “Megaprojects for Science and Technology Innovation 2030” that both big data and
intelligent manufacturing and robotics are included.
In July 2017, State Council propagated Development Plan on the New Generation of Artificial Intelligence
which marks that AI became national strategy. As in November of the same year, the Ministry of Science and
Technology convened a high-level meeting that marked the official launch of the plan, standing up the New
Generation AI Development Plan Promotion Office. In support of the plan, the New Generation AI Strategic
Advisory Commission was also created at that time, convening senior academicians and experts from
prominent private sector players, including Baidu, Alibaba, Tencent, iFlytek, and Horizon Robotics. In
December, Ministry of Industry and Information Technology propagated Three-Year Action Plan for Bolstering
the Development of the Next-Generation Artificial Intelligence Industry in which built up the prior plan to
concentrate with greater specificity on objectives that will support the development of a world-leading AI
industry. The new plan recognized the importance of an AI industry “support system” to include a data resource
base with standard test data sets, cloud-based training frameworks, and initial test and evaluation systems.
In April 2018, the General Office of the State Council issued the Opinions on Promoting the Development of
“Internet plus Health Care” which primarily pertained to three aspects with the improvement of the “Internet
plus health care” service system and its support system. A total of fourteen recommendations were made to
encourage medical institutions to leverage the Internet and other information technologies to expand the
scope and content of health care services, and to construct an integrated online and offline health care
model covering the entire health care process. It also encouraged medical institutions to cooperate with
Internet companies for strengthening the integration of regional medical and health information resources,
with the use of big data to make epidemiological trend predictions and enhance the smart monitoring of
contagious diseases. It also encouraged the development of AI-based clinical diagnosis and treatment
decision support systems, the development of intelligent medical image recognition, pathological
classification and multidisciplinary consultation as well as intelligent voice technology applications in a variety
of medical scenarios for improving the efficiency of medical services. In the development of “internet plus”
medical services, the development of Internet hospitals relying on medical institutions will be allowed.
Medical institutions can use the internet hospital as the second name, and on the basis of the real hospital,
allow some common diseases and chronic diseases to be visited online.
In September 2018, the National Health Commission (NHC) released Measures for the Administration of
Internet Diagnosis and Treatment (For Trial Implementation), Measures for the Administration of Internet
Hospital (For Trial Implementation) and Specifications for the Administration of Remote Medical Services
(For Trial Implementation). Those documents clearly defined Internet diagnosis, which means medical
institutions could ask their own registered doctors to provide “Internet plus” family doctor signing services for
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE AI POLICY DEVELOPMENT 4
some common diseases and follow up consultations for chronic diseases by using internet and other
information technologies.
In order to better evaluate and standardize the digital services produced by hospitals, China implemented a
smart medical service grading system for the development of "smart hospitals," on March 2019. A typical
"smart hospital" features information-based service systems including a database of patients' medical
records, a registration system, a hospital navigation system and a logistics management system. By
introducing the grading system, hospitals will have a better idea about their advantages and shortcomings,
giving them the ability to better allocate their resources in building themselves into "smart hospitals,"
In parallel to these efforts at the national level, cities throughout China have started to develop and release
their own plans and policies for AI, including Beijing, Shanghai, Hangzhou, Zhejiang, and Tianjin etc.
Notably, Beijing plans to build a 13.8 billion RMB ($2.12 billion) AI development park that could host up to
400 AI enterprises. At the same time, Shanghai plans to establish a special fund, with a target size of 10
billion RMB in the first phase, to invest in AI development, and Hangzhou has launched its own AI park,
along with a fund that will invest 10 billion RMB in it.
Table 1 Summary of healthcare AI relevant policies in China
2 Healthcare AI in China
In China, the growing aging population and the uneven distribution of medical resources opens great needs
for AI in healthcare. At the same time, the big population number and the size of application market provide a
good basis for developing AI technologies. AI plays an important role in many aspects of medical services,
including clinical assistive decision, disease prediction, medical service, personalized treatment etc.
Different from many other developed countries, where family doctors serves as the gateway to specialist
hospitals, China doesn’t have hierarchical medical system, which means the patients can visit any hospital at
any time they want. The flow of visiting a doctor in hospital in China could be illustrated as below. AI can
target at each steps to provide solutions to improve the quality of healthcare service. The administration
department also rely on AI to innovate the development of Chinese medical industry.
Fig. 1 The flow of visiting a doctor and the role of AI.
Chinese AI companies are highly concentrated in Beijing, Shanghai and Guangdong. Among the world’s top
20 cities in terms of AI companies hosted, Beijing ranks first with 395 companies, and Shanghai, Shenzhen
and Hangzhou are also among the top 20. China’s AI companies mainly specialize in three categories—
voice, vision and natural language processing—with only a small percentage focusing on basic hardware.
AI is transforming China’s healthcare industry. Hospitals across the country have started adopting AI-
assisted medical products and services that include not just image-assisted diagnosis, but also range from
surgical robots to smart patient health monitoring and virtual medical assistants. New medical AI start-ups
are emerging and traditional healthcare companies and leading technology firms are betting big on AI
solutions.
China’s healthcare AI boom is also strongly supported by the central government that aims for China to
become the world’s leading AI innovation center by 2030. The application of AI in healthcare is a priority as
Chinese leaders look for ways to relieve the immense pressure that a rapidly urbanizing and aging
population is putting on the country’s healthcare system. The hope is that new AI-driven treatment methods
can address doctor shortages, high misdiagnosis rates, inaccuracies in early disease prediction, and much
more.
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE HEALTHCARE AI IN CHINA 7
Based on the statistics in the Bluebook of World Artificial Intelligence Industry 2018, Chinese AI related
companies focus on sectors including healthcare, finance, security, business, education etc., and the
percentage of companies in healthcare sectors is 22%, which rank as top 1 among other sectors. The
application areas of AI in healthcare sectors are mainly focusing on medical imaging, assisted diagnosis,
drug development, health management and disease prediction.
2.1 AI medical imaging

AI could help radiologist analyse the medical images and design treatment programmes which are
personalised to the patients. AI may have multiple applications in medical imaging, such as imaging
acquisition, imaging processing, aided reporting, follow-up planning, data storage, etc.
AI medical imaging is one of the most common application of AI in healthcare industry in China. Currently, it
is the second-largest market segment of AI medical applications in China and was predicted to grow at a
growth rate of more than 40% and reach a scale of 2.5 billion US dollars in 2024. In addition, AI medical
imaging in China is considered by industry insiders to be the first market to commercialize AI in the
healthcare industry in China.
2.1.1 Painpoints
As one of the very important diagnostic procedure, medical imaging allows doctor to diagnose people more
accurately without being intrusive. The painpoints of medical imaging field in China are
 Doctor’s heavy workload in top-level hospitals. Take pulmonary nodule as an example, on average,
a top-level hospital will receive about 200 patients for pulmonary nodule screening, and each
patients will have 200 – 300 CT images. Each radiologist need to read 40,000 – 60,000 CT images.
The overwork has affected the image interpreting quality and the diagnostic accuracy of radiologists.
 There is a huge gap in the number of professional doctors in medical imaging in China. The dataset
of medical imaging increases 30% every year in China, which also account for 90% of digital data
volume for a hospital. In contrast, the numbers of radiologist increase by only 4% a year.
Under such circumstance, AI technology is expected to improve the diagnostic accuracy while shortening the
image reading time.
2.1.2 Application scenario

The application of AI in medical imaging include
1. CT/PET image reconstruction. AI has the potential to improve image quality, and reduce the
radiation dose.
2. Read and analyze chest X-rays. AI can help doctors in medical images analysis for many disease
screening, and improve work efficiency and diagnostic accuracy.
3. Eye diseases screening. AI could improve the diagnosis accuracy of Glaucoma, Diabetic retinopathy
and age-related macular degeneration.
4. Brain MRI segmentation. Quantitative analysis of brain MRI is routine for many neurological
diseases and conditions and relies on accurate segmentation of structures of interest. Deep
learning-based segmentation approaches for brain MRI are gaining interest due to their self-learning
and generalization ability over large amounts of data.
5. Intracranial hemorrhage detection. AI algorithm was able to accurately identify acute intracranial
hemorrhage on head CT scans and even find some tiny hemorrhages that could be overlooked by
radiologists.
6. Target volume delineation for radiotherapy. AI algorithm is able to provide accurate, fast, and
automated contouring of primary tumor volumes on multiparametric MR images.
7. Bone fractures detection. AI would be capable to automatically detecting and classifying the bone
fractures and avoid missed diagnosis.
8. Breast cancer diagnosis. AI could improve the accuracy in detecting breast cancer.
9. Bone-age analysis. Each doctor needs around 1-2 hours to make an analysis, and with AI, it will only
need 15 – 30 min.
2.1.3 Advantages of development in China

The AI enterprises in United State cover a variety of fields in the medical care system, while in China, AI
enterprises have been mostly working on medical imaging due to the rapid growth of clinical demands, the
imbalanced distribution of high-quality healthcare resources and the lack of medical imaging doctors around
the country. AI in medical imaging has become the most commonly used applications in China, because of
several developing advantages of AI in the healthcare industry.
First, the medical imaging data is huge; more than 90% of the medical data comes from medical images. The
picture data structure is simple, which is convenient for machine learning. It is estimated that the volume of
medical data will reach 40 trillion gigabytes by 2020, 30 times that of 2010.
Another developing advantage of AI in China’s healthcare industry is about the fast iteration of big data, and
intelligent image diagnosis algorithm is relatively mature. In addition, medical imaging in China is strongly
supported by national policies. The government has issued several policies to increase support for domestic
medical imaging since 2013. In 2017, the total amount of financing in this field exceeded 1.7 billion RMB.
2.1.4 Current industrial status and key players

AI health market in China was about 20 billion RMB in 2018, a 53% increase over 2017, and investment in AI
in the healthcare industry in China is ranked as No. 1 in the world.
With regard to application, the AI medical imaging products mainly applied to disease screening, with a focus
on pulmonary nodule, diabetic retinopathy and stroke.
Pulmonary nodule screening

Lung cancer is the leading cause of cancer-related deaths in China, with over 690,000 lung cancer deaths
estimated in 2018. Products for pulmonary nodule screening is, no doubt, the most mature among other AI
powered medical products. By July 2018, there were over 20 companies in China claiming to have
developed products in this area. For now, AI enabled lung cancer or pulmonary nodule screening system
have been landed in many hospitals. However, all those initiatives focus on research cooperation with
hospitals, since it is still under discussion how to charge AI enabled product or service. Below table lists
some Chinese companies and their clinical partners in developing enabled lung diseases screening system.
Table 2 Chinese AI medical imaging companies
Diabetic retinopathy screening

Diabetic Retinopathy (DR) is the most common causes of vision loss in adults aged from 20 – 74. DR often
dose not have early warning signs, so it can cause rapid vision loss. Screening is recognized as an important
way of early identification of the eye with mild DR, which allows prompt intervention to prevent disease
progression. However, due to the lack of eye doctors and low awareness, the screening rate is less than
10%. AI is expected to perform the primary screening, which can largely increase the early screening rate in
China. Below table shows the leading companies in China developing AI enable DR screening system.
Table 3 Chinese AI companies developing DR screening system.
Target volume delineation

Radiotherapy is one of the three most important cancer treatments (the other two are surgery and
chemotherapy). Each patient has about 200 CT images, and doctor needs to mark the position of the tumour
on every image. AI could help doctors delineate the target areas automatically, and then deliver the
treatment plan. The doctor only needs to give the final confirmation of the treatment plan. Currently in China,
there are few companies focusing in target volume delineation, and two of them are listed in below table. In
August 2019, Linking Med, an AI company focusing in radiation oncology, received Class III medical device
license for its cloud-based product, which is of enormous importance as the first-of-its-kind license in China.
Table 4 Chinese AI companies developing radiotherapy planning system.
Besides, products for bone-age assessment, breast cancer screening, cervical cancer screening etc., start to
be landed in hospitals also. Some relevant companies are listed in below table.
Table 5 Chinese Companies focusing on other AI-enabled medical imaging solutions.
2.2 AI assisted diagnosis

AI assisted diagnosis includes medical imaging, electronic medical records, medical service robots, virtual
assistant. Medical imaging has been elaborated in above chapter, and in this chapter we will introduce other
applications, i.e. electronic medical record and virtual assistant.
Electronic medical record

The traditional electronic medical record (EMR) system in China doesn’t meet the needs of specialization of
disease data, structuring of medical case data, and cannot provide clinical decision support based on
medical records. AI can use natural language processing technology to standardize and structure medical
records and use speech recognition and speech synthesis to process large amounts of text entry.
Virtual assistant
Virtual assistant can provide real-time support to doctors. For example, patients need to know the necessary
information about the disease when they go offline for consultation, and these highly overlapping contents
take up a lot of doctors’ time. AI technology can help doctors respond to inquiries based on a large amount
of historical information, saving time and energy.
Fig. 2 A woman uses a screen on a robot developed by iFlytek at the outpatient hall in a hospital.
2.2.1 Painpoints
China has lagged behind in the adoption of electronic health records (EHRs) in healthcare institutions but the
use of EHR systems is now rapidly growing as a result of the new Medical Reform. The widespread adoption
of EHRs has made it easier for researchers to access and aggregate longitudinal patient data. China is likely
to be the country with the largest volume of EHR data because of the large population and the rapid
development and deployment of information systems across the country in recent years. However, there are
many factors influencing the quality of EHRs. The first is Chinese doctors’ heavy workload. On average,
Chinese doctors need to see over 100 patients each day in outpatient department, and each patient only has
3 – 4 min for consultation. It is very time-consuming for them to input each items carefully into the EHR
system. Secondly, EHRs are mainly used by tertiary and secondary hospitals (Tertiary hospital is the best
hospitals and primary is the lowest in China), and some primary hospitals in lower tier cities still use
handwriting for recording. In 2018, NHC evaluated the EHR systems of 4329 secondary hospitals and 1695
tertiary hospitals, and found that the average grade of EHR system in secondary hospitals is 1.35 and that in
tertiary hospitals in 2.81 (grade 0 – 8 and grade 8 represent the highest). Since 2009, relevant administrative
departments issued several Notices to promote the application of EHR system in hospitals, however, after 10
years, the quantity increases, but the quality is still poor. According to the Notice on Further Promoting the
Construction for the Informationization of Medical Institutions with Electronic Medical Records as the Core
recently issued by NHC, by 2020, all tertiary hospitals should upgrade their EHR system into grade 4 and
above. AI could use natural language processing technology to transfer voice messages into electronic
records and also to standardize and structurize stored records.
Due to the lack of triage system, long queue for registration and consultation, and patients’ knowledge
deficit, patients normally spend a long time to find out which clinical department is the right one to go and
which doctor should he/she visits. Based on facial recognition and natural language processing technologies,
auxiliary medical robot could provide basic information to patients, such as the visiting process, clinical
department’s introduction, preliminary triage etc., in order to response quickly to patients’ needs and reduce
the workload of the hospital.
In addition, due to the inconsistent health data and disconnection between different hospitals, patients
normally need to bring all their historical examination results and health data to visit the doctor, and for
doctor, it is waste of time to asking about patients’ history everytime. Virtual assistant is intended to help
doctors collect and analyse patients’ historical data in order to improve the consultation efficiency.

China is relatively more mature in auxiliary medical robot, and could be an advantage in developing other
medical robots. In electronic health record based on speech, due to the big needs in the market, many
Chinese companies put a lot of effort in developing advanced solutions.

In EHR field, the application could be intelligent voice recording, clinical decision supporting, natural
language recognition. For auxiliary medical robot, the application is very clear which is for general guidance
and preliminary triage. For virtual assistant, the primary application is for intelligence consultation.

Since most of the AI-assisted diagnosis products are software and hardware integrated solutions, healthcare
AI companies in China cooperate with hospitals and continuously train models and optimize algorithms
through hospital desensitization medical records data. Hospitals can use the products mostly for free. In the
future, when the product becomes more mature, AI in diagnosis in China may be inclined to adopt the
service charge business model. The below table lists some key industrial players in assisted diagnosis.
Table 6 Chinese companies focusing on AI assisted diagnosis.
2.3 AI drug development

Medicine discovery is an extremely expensive process, on average new drug discovery takes 10 years with 1
billion dollars. AI has already been successfully applied in all stages in drug development: literature search,
identify target molecules, discover effective drugs, speed up clinical trials, and find biomarkers for
diagnostics.
2.3.1 Painpoints
At present, medicine discovery in China has several problems to be solved. Some of which can serve as the
directions of AI development. Specifically, medicine discovery in China always takes a long-time duration,
high costs, high-risk, and low rate of return. The integration of AI and drug discovery can greatly reduce time
and development costs.

AI can be applied in the whole process of drug development as showed in below figure, such as target
discover, compound synthesis and screening, crystal structure prediction, toxicity test, patient recruitment,
clinical trial design etc.
Fig. 3 The application of AI in new drug development.

Drug discovery applications accounts for over 35% market share of healthcare AI industry globally. In new
drug discover, Chinese pharmaceutical companies are still mainly focusing on generic drugs and improved
drugs, and foreign companies are focusing on innovative drugs. Regarding to the application of AI in new
drug discovery, Chinese companies are lagging behind. By October 2018, there were 6 Chinese companies
received investment successfully, and the total amount is 250 million dollars. Below table introduces the
Chinese companies in drug discovery.
Table 7 Chinese companies focusing on AI drug development.
2.4 AI health management

AI health management is a way to change from passive disease treatment into proactive self-monitoring, and
the main products are wearable devices. The market potential is big in China since the growing aging
population and the release of Two-Child Policy.
2.4.1 Painpoints
For health management in China, the painpoints include
 Firstly, smart wearable devices in traditional health management fail to solve data correlation.
Wearable devices only focus on data extraction, collection, and trend analysis, but fail to provide
users with health portraits and improve their health.
 Secondly, staff in the health management field are not that professional; most having a non-medical
background.
 Thirdly, there are growing population with chronic diseases in China. Hospitals, especially the tertiary
hospitals are not able to provide medical services especially when the patients have been
discharged. As planned by the national government, primary hospitals should take the main
responsibilities for chronic disease management for the citizens living nearby. However, due to the
quality of service in primary hospitals is low; people normally have very low trust to the doctors in
primary hospitals.

The massive big data in China is the prerequisite for AI development, especially in health management.
From the technology-driven perspective, AI can make personalized health management possible through
efficient calculation and accurate decision analysis. Even in the future, nutritionists and sports experts can
generate accurate health intervention plans based on the AI system of healthtech in China.
Most importantly, national government highly supports the health management for general population
especially after the release of Health China 2030.

The main application scenario of using AI in health management is chronic disease management, especially
diabetes and high blood pressure. Mother and Child Care is also a focused area as the two-child policy
released in China. The companies in these areas mainly use AI technologies to analyse individual data and
provide personalized health management plan.

China’s current amount of smart wearable device hardware is not high enough, and the accumulation of
disease-related data is insufficient. Therefore, the main application of AI in health management in China is
chronic disease management (such as diabetes and hypertension), maternal management, mental health
management, and population health management.
Table 8 Chinese companies focusing on AI health management.
2.5 AI disease prediction
In China, AI technology in disease prediction is mainly used for gene sequencing to forecast disease
occurrence. Current business models of genetic testing services of AI in China’s healthcare industry can be
divided into two types: a hospital delivery model and third-party testing model. Compared with the hospital
delivery model, the third-party testing model requires a higher threshold of qualifications, personnel, and
funds of medical inspection institutions.
2.5.1 Painpoints
The main painpoints for disease prediction, mainly referring to gene sequencing, business in China include
 The amount of genome data is huge, and manual experiments are time-consuming and labor-
intensive.
 Traditional gene sequencing costs too much.
 The general algorithm of gene sequencing diagnosis in different stages is ineffective and has low
accuracy.

The advantage of developing disease prediction technology and business in China is the increasing market
size and the development of supercomputer. According to the trend of disease incidence, cancer incidence,
and death rate have been continuously increasing in China in recent years. In 2015, the number of people
with cancer increases to 4.3 million, and 2.8 million died because of cancer. Disease prediction receives a lot
of attention. Data shows that from 2007 to 2016, China’s revenue from gene sequencing increased by 62.2
percent annually, reaching 5.06 billion RMB in 2016. It is estimated that by 2022, China’s gene sequencing
market will reach about 30 billion RMB. Additionally, if supercomputer’s powerful data processing ability is
applied to gene sequencing, it will greatly shorten the time of gene testing and improve efficiency.

Companies dedicated in disease prediction can mainly be divided into two categories, one is controlling the
core technology for gene sequencing and develop gene sequencing equipment, and the other one is using
commercialized gene sequencing equipment to provide gene sequencing services to clinical institutions or
individual customers.
2.5.4 Current industrial status key players

Currently in China, there are 45 companies claiming to provide “disease risk prediction” service, 10% of them
have the capability to do gene sequencing. China is still lagging behind in the upstream of gene sequencing
regarding to technologies for equipment development. In the middlestream, a lot of companies are
competing intensively in providing gene sequencing services, and there is still no leading companies in this
area.
Table 9 Chinese companies focusing on AI disease prediction.
2.6 Key stakeholders

2.6.1 Governmental and academic key stakeholders
Shanghai Hospital Development Centre (SHDC)
SHDC was founded in 2005 by the municipal government of Shanghai as a non-profit corporation. It
manages the operation of 38 municipal tertiary hospital (almost all top-level hospitals in Shanghai) in
Shanghai, and controls budget, staffing, strategic planning, and health-record sharing of municipal public
hospital. SHDC is running a Hospital Link Project since 2006 with the main target to connect different
systems (including Picture Archiving and Communication System, Hospital Information System, Laboratory
Information System, Clinical Information System) in order to share data and information in-real time between
all hospitals in Shanghai.
Chinese Innovative Alliance of Industry, Education, Research and Application of Artificial

Intelligence for Medical Imaging (CAIERA)
CAIERA was set up in April 2018 with the commitment to effectively integrating various resources from
industry, education, research and application for healthy development of AI in medical imaging. The original
members of the Alliance include 120 well-known Tertiary hospitals in China, 55 enterprises of medical
imaging AI and 35 scientific research institutions. The alliance has issued a certain number of consensus
on relevant topics in AI, and one of them are The White Paper on Medical Imaging Artificial Intelligence in
China issued in March 2019, which plays a guiding role in understanding the market demands and
establishing standardized systems in the path of landing AI products in the field medical imaging.
Med-X Research Institute of Shanghai Jiaotong University

Med-X is China’s newest research institute combines engineering, life science and physical science to
cutting edge translational research to solve problems in biomedicine. The Institute consists of eight joint
research centers. They include: The Med-X-Ruijin Hospital microPET/CT Research Center; The Digital
Medicine Research Center of The Ministry of Education; The Med-X-Renji Hospital Clinical Stem Cell
Research Center; The Shanghai Med-X Medical Equipment Engineering Research Center; The Med-X-Sixth
Hospital Shanghai Medical Equipment Research Center; The Rehabilitation Engineering Research Center;
The Med-X-Mental Hospital Neuroimaging Center; and finally the Nano Biomedical Research Center.
Fig. 4 Organization structure of Med-X Research Institute
Shenzhen Institutes of Advanced Technology (SIAT) of the Chinese Academy of Sciences (CAS)
SIAT was jointly established by CAS, the Shenzhen municipal government and the Chinese University of
Hong Kong in February 2006. To meet the nation’s needs in healthcare and manufacturing, SIAT focuses on
emerging industries such as low-cost healthcare, service robots, electric vehicles, cloud computing, digital
cities, nano-medicine, new energy and new materials. SIAT has also established long-term cooperation with
many foreign academic and research institutions, including Stanford University (USA), Massachu-setts
Institute of Technology (MIT), NICTA (Australia), TRLabs (Canada), the University of Hamburg (Germany),
the University of Southampton (UK) and many other institutions.
Fig. 5 The focused areas of Shenzhen Institutes of Advanced Technology
Shanghai Institute of Materia Medica, Chinese Academy of Science

Shanghai Institute of Materia Medica (SIMM), Chinese Academy of Sciences (CAS), has the longest history
as a comprehensive research institution for drug discovery in China. Research priorities are given to treat
major diseases, such cancers, cardio-cerebrovascular diseases, neuropsychiatric diseases, metabolic
diseases, autoimmune diseases, and infectious diseases. SIMM is leading a national project to use AI as the
core technology for new drug development.
2.6.2 Industrial Key Stakeholders

Chinese Tech Giants
The Chinese Tech giants, e.g. Tecent and Alibaba, have begun to see the health care system’s challenges
as an opportunity to leverage their “anything, anytime, anywhere” consumer-focused approach to capture a
new market with digital health care. While physicians, hospitals, and other providers may have more
experience working within a heavily regulated environment and an ability to deliver specialized, high-acuity
care, these tech companies benefit from a deep understanding of consumers, enormous financial resources,
and the ability to innovate and scale their technology rapidly. This has positioned them well to meet the basic
health care needs of a significant proportion of the population.
Tecent whose WeChat communication platform has more than 1 billion users, is pursuing a “Smart Hospital”
strategy to enable patients to schedule appointments with specialists, conduct virtual visits, and access
personal health information such as diagnostics, imaging reports, and prescriptions. Tencent is also making
inroads into ambulatory care, building a network of primary care and ambulatory surgery centers in eight
cities, as well as communication, payment, and referral services for providers and patients.
Alibaba users of the popular payment platform Alipay, known to the Chinese public as a fast and safe
financial transaction platform, can schedule appointments. Alibaba then leverages its logistics capabilities to
easily deliver those medications from partner pharmacies in less than 24 hours. Seeking to capture more of
the value chain, Alibaba established its Tmall pharmacy division to distribute over-the-counter drugs and
medical devices to consumers.
Ping An Technology
Ping An Technology, a subsidiary of Ping An Group uses AI, cloud and other cutting-edge technologies to
develop and operate mission-critical platforms and services that support financial services, medical health,
smart cities etc. Based on deep learning technologies, Ping An Technology built a prediction model using
case reports from participating hospitals, historical data from regional health authorities and meteorological
and environmental statistics to predict flu outbreaks ahead of time with accuracy rates of over 90% in cities
of Chongqing and Shenzhen. The results were published recently in EBioMedicine with the title “Forecasting
Influenza Activity Using Self-Adaptive AI Model and Multi-Source Data in Chongqing, China”.
Ping An Good Doctor, which is also under Ping An Group, is China’s largest health care platform with
consultations, medical appointment services, certain health checks, has begun installing unstaffed, AI-
enabled kiosks, called “One-Minute Clinics”, in communities and work sites around the county. Users sit in a
small booth and talk with an “AI Doctor” about their symptoms and medical history; the virtual doctor then
makes a diagnosis and treatment recommendations. If necessary, the patient can talk to a physician by
video, and the kiosk’s smart medicine cabinet can dispense more than 100 different medications. In the first
2 months of operation, the one-stop clinics, which are open 24-7, has reportedly served 3 million consumers.
Fig. 6 Ping An Good Doctor’s One-Minute-Clinics are now installed in eight provinces across China.
3 Health data
With the largest population in the world, China has a large amount of data in health sector. According to the
statistics, by 2020, Chinese medical data will increase to 35ZB, which accounts for 20% of medical data in
the world. However, the quality of medical data is still poor due to the system separation between different
hospitals, even between different departments. In addition, the medical records in some primary hospitals
and hospitals in small cities or rural areas are still based on hand-writing, which lowers the value of those
medical data. How to structurize and standardize those medical data is still big issue in China. Government
hopes by using new technologies in data collection, storage and analysis, that this data can become more
valuable.
AI’s inherent dependence on data is at odds with the government’s goal of protecting sensitive medical data.
Data sharing is essential to ensuring the wide and safe application of AI-driven medical products. Datasets
used to train AI systems need to be drawn from a wide variety of hospitals and patients. For this reason,
regulators have recently taken steps to improving China’s currently underdeveloped and unstandardized
electronic resource infrastructure. The NHC has issued policy documents to promote the use of electronic
medical records across hospitals and assigned local regulators to assess how well they are being adopted.
However, under China’s Cybersecurity Law and related implementation measures, much of the data
generated in a medical context are classified as sensitive, potentially subjecting them to strict data
localization requirements or security reviews. Initial healthcare-specific guidelines are requiring medical
institutions to store patient information on secure servers within China and to apply for security evaluations if
they wish to transfer it overseas.
China also lacks a robust regulatory framework for AI-powered software and devices. Much like in the United
States and other parts of the world, many products in China are a long way from being commercialized. Yet,
the lack of unified industry standards for risk classification and evaluation of medical AI is still causing
significant delays in product registrations. Regulators are taking a distinctly conservative approach, with
current medical device catalogs favoring the highest risk classification (Class III) for products or solutions as
soon as they incorporate deep learning with explicit diagnosis functions. Many of the 50+ start-ups in China
that have invested heavily in algorithm-driven diagnostic products may not survive due to the complexities
and high costs of medical device registration. Increasingly, analysts suspect that as much as 90 percent of
all Chinese AI start-ups could fail amid fierce competition, pressure to commercialize, and a wider economic
slowdown.
To solve the problem, led by the NHC and the National Medical Product Administration (NMPA), healthcare
regulators have been actively working on creating classification catalogs, clinical trial guidelines, and
technical review documents specifically for medical AI. In February 2019, the Center for Medical Device
Evaluation (CMDE) issued evaluation guidelines for medical devices using AI for clinical decision-making. In
July 2019, led by CMDE from NMPA, Cooperation platform for AI medical device innovation was officially
established, which consists of government, associations, medical institutions and universities. In this
cooperation platform, several working groups and expert committees have been created with the specific
task of creating further standards and guidelines for AI medical device approval.
In the cooperation platform there are 3 out of 10 working groups focusing on database, which shows that
CMDE pays a lot of attention on clinical data as the key element for AI technology.
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE HEALTH DATA 22
Fig. 7 Cooperation platform for AI medical device innovation.
Table 10 The key responsibilities of each working group in cooperation platform.
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE HEALTH DATA 23
4 Conclusion and study trip 2020
The aging and growing populations, the prevalence of chronic diseases and exponential advances in
innovative digital technologies have and will continually alter and transfer the way healthcare and eldercare
are delivered. China’s medical AI booms in recent years, which is very much aligned with the general aim of
Chinese government, which is to become the world’s leading AI innovation centre by 2030. The favourable
policy environment, financial support, as well as the huge market size for AI healthcare development in
China encourage more and more foreign companies to look for opportunities in China. Danish government
also just released the ‘Danish Strategy for AI’ and healthcare is one of the prioritized areas in 2019. Denmark
has excellent AI research community and strong institutions for turning research into business application.
ICDK Shanghai will leverage the opportunity and established network to promote more Sino-Danish
cooperation in AI healthcare and assist Danish AI healthcare companies to find business possibilities in
China.
In order to achieve more Sino-Danish cooperation in AI healthcare, encouragement of open communication

and better local understanding of Chinese AI healthcare ecosystem is the pre-requisite. ICDK Shanghai is
planning to organize a fact-finding tour to China in 2020 for relevant Danish audiences and stakeholders
from industry, academy and authorities. During the trip, we plan to visit and arrange in-depth discussion with
Chinese leading AI healthcare companies, top-level researchers, authorities and other key stakeholders in
order to gain insights into the Chinese regulatory, business and academic environment. For more
information, feel free to contact us.
CONCLUSION AND STUDY
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE 24
TRIP 2020
Reference
[1] China Institute for Science and Technology Policy at Tsinghua University. China AI Development Report.
2018
[2] Artificial Intelligence Institute, Shanghai Jiaotong University et al. White Paper of China Healthcare AI.
2019
[3] Xudong Zhang. Annual Report on Medical Artificial Intelligence in China. 2019
[4] China Academy of Information and Communications Technology. Blue Paper of World Artificial
Intelligence Industry. 2018
[5] Chinese Innovative Alliance of Industry, Education, Research and Application of Artificial Intelligence for
Medical Imaging. White Paper of Chinese medical Imaging AI. 2019
[6] ifenxi: Report on Chinese Medical Imaging Industry. 2019
[7] Yue I Cheng, Michael P A Davies, Dan Liu, et al. Implementation planning for lung cancer screening in
China. Precision Clinical Medicine. 2019(2):13–44.
[8] http://news.rfidworld.com.cn/2019_03/4afb67743d39f624.html
[9] https://www.hit180.com/37980.html
[10] https://www.yigoonet.com/article/22454019.html
[11] https://thediplomat.com/2018/02/chinas-ai-agenda-advances/
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE REFERENCE 25
Innovation Centre Denmark Shanghai
Dr. Yu Qi, Innovation Officer (Health & Life Science)

Direct phone: +86 136 0186 6475
E-mail: [email protected]
Yu Qi has a science PhD in molecular medicine and experiences in both academic

and industry research. As Life Science Innovation Officer at ICDK, she assists
Danish companies and organizations who wish to expand their research and
technologies to the Chinese market.
Work Experience / Core Competencies: Market research, technology scouting,

strategic partnership development within the Life Science and Healthcare sector,
facilitating and developing Sino-Danish partnerships in both academic and industry
fields.
THE MINISTRY OF HIGHER EDUCATION AND SCIENCE REFERENCE 26

2020 AI Healthcare Report

Uploaded by

Copyright:

Available Formats

2020 AI Healthcare Report

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2020 AI Healthcare Report

Uploaded by

Copyright:

Available Formats

REPORT BY

Innovation Centre Denmark

Fig. 1 The flow of visiting a doctor and the role of AI.

2.1 AI medical imaging

2.1.2 Application scenario

2.1.3 Advantages of development in China

2.1.4 Current industrial status and key players

Pulmonary nodule screening

Diabetic retinopathy screening

Table 3 Chinese AI companies developing DR screening system.

Target volume delineation

Table 4 Chinese AI companies developing radiotherapy planning system.

Table 5 Chinese Companies focusing on other AI-enabled medical imaging solutions.

2.2 AI assisted diagnosis

Electronic medical record

2.2.2 Advantages of development in China

2.2.3 Application scenario

2.2.4 Current industrial status and key players

Table 6 Chinese companies focusing on AI assisted diagnosis.

2.3 AI drug development

2.3.2 Application scenario

Fig. 3 The application of AI in new drug development.

2.3.3 Current industrial status and key players

2.4 AI health management

2.4.2 Advantages of development in China

2.4.3 Application scenario

2.4.4 Current industrial status and key players

Table 8 Chinese companies focusing on AI health management.

2.5.2 Advantages of development in China

2.5.3 Application scenario

2.5.4 Current industrial status key players

2.6 Key stakeholders

Chinese Innovative Alliance of Industry, Education, Research and Application of Artificial

Med-X Research Institute of Shanghai Jiaotong University

Shanghai Institute of Materia Medica, Chinese Academy of Science

2.6.2 Industrial Key Stakeholders

Table 10 The key responsibilities of each working group in cooperation platform.

In order to achieve more Sino-Danish cooperation in AI healthcare, encouragement of open communication

Dr. Yu Qi, Innovation Officer (Health & Life Science)

Yu Qi has a science PhD in molecular medicine and experiences in both academic

Work Experience / Core Competencies: Market research, technology scouting,

You might also like