laitimes

In-depth analysis | Smart Healthcare Industry Research Report

author:SCI Academic Gang

Smart Healthcare will take the construction of a smart medical and health service cloud ecosystem as its vision and goal, integrate and gather resources from all parties in medical and health care through the overall unified medical and health cloud service platform, and provide residents and patients with high-quality and efficient medical and health services with full life cycle, online and offline integration.

1. Overview of the smart medical industry

1.1 Introduction to the concept of smart healthcare

The concept of smart healthcare: Based on innovative technology, with the purpose of solving the pain points of hospitals, patients and sub-health groups, regional public health, and pharmaceutical enterprises in the medical scene, the product portfolio according to the characteristics of the medical scene is used to solve the pain points in the scenario. At the current stage, the providers of smart medical products are mainly Internet technology companies, supplemented by some hardware manufacturers.

Smart healthcare scenarios are divided into smart medical diagnosis and treatment scenarios, smart hospital management scenarios, smart patient service scenarios, smart regional primary medical scenarios, and smart pharmaceutical enterprise scenarios. Its main smart medical products include: medical image screening products, clinical decision support (CDSS) products, intelligent review products, intelligent electronic medical record products, hospital navigation products and so on. In general, the concept of smart medical care is relatively broad, and the application scenarios and related products involved are also rich in types, and the theme is still the relevant application of Internet technologies and products such as artificial intelligence, Internet of Things, and 5G in the medical industry.

Figure 1: Research scope of smart healthcare

In-depth analysis | Smart Healthcare Industry Research Report

Source: Yiou Think Tank

1.2 Introduction and development history of smart medical policy

In the past five years, from the CPC Central Committee, the State Council to various ministries and commissions, a large number of relevant policies for the medical industry have been successively introduced, emphasizing the important supporting role of informatization and a new generation of information technology in the medical industry, and smart medical care has ushered in a policy-intensive period.

In-depth analysis | Smart Healthcare Industry Research Report

Figure 2: Smart healthcare-related policies (1)

Source: Huaan Securities Research Institute

Figure 3: Smart healthcare-related policies (2)

In-depth analysis | Smart Healthcare Industry Research Report
In-depth analysis | Smart Healthcare Industry Research Report

Source: Government website, Essence International

In-depth analysis | Smart Healthcare Industry Research Report

Figure 4: Smart medical industry policies have been introduced one after another

Source: Huaan Securities Research Institute

From a worldwide perspective, artificial intelligence (AI) began to become an independent research field in 1956, and before the 20th century, Chinese and foreign research on AI in the medical field focused on the clinical knowledge base, but because most of the clinical knowledge base must run on LISP devices. However, due to the fact that LISP equipment could not be networked and was expensive at that time, the clinical knowledge base was not widely used in clinical practice. From 2000 to 2015, foreign research focused on the application of AI outside the clinical knowledge base, such as the application of surgical robots and encouraging the development of electronic medical records. However, China's clinical knowledge base is still mainly based on the study of more types of diseases, and the development is relatively slow. From 2015 to 2017, due to the significant improvement of the accuracy of AI in image recognition, AI+ imaging has developed rapidly. Thanks to long-term research in the clinical knowledge base, CDSS products have matured. After 2018, China's AI+ medical care has entered a stage of stable development, and new products such as smart cases have been launched one after another.

Figure 5: Global smart healthcare development history

In-depth analysis | Smart Healthcare Industry Research Report

Source: iResearch Institute

Due to the late start of related underlying technologies in China, the development stage of the overall smart medical care is divided into three stages by the combination of time and innovative technology and medical scenarios:

The first stage: innovative technology + initial integration of medical treatment

At this stage, there are few combinations of innovative technologies and medical scenarios, and no mature products have yet appeared. At the national policy level, the pilot of innovative technology application is encouraged, hoping to promote its large-scale application in the medical field, and the financing is relatively early, and the start-ups are mainly angel rounds or A rounds. Mainly before 2016, there were fewer relevant policies.

The second stage: innovative technology + the birth of medical products

At this stage, the combination of innovative technology and medical scenarios is general, but smart medical care ushered in an outbreak, and national policies encourage medical institutions to introduce a variety of innovative technologies, such as artificial intelligence technology, cloud computing, big data, etc., to explore the establishment of a new medical system, during which mature smart medical products appeared. The main period is between 2016 and 2019, including the continuous promulgation of a number of smart healthcare-related policies.

The third stage: the smart medical stage

At this stage, the combination of innovative technology and medical scenarios is relatively close, and the state gives guidance to deep learning-assisted decision-making medical device software, emphasizing its data security, algorithm generalization ability and clinical use risk, and Internet companies with technical advantages have joined the medical industry. Mainly from 2019 to the present, smart medical related policies are mainly based on product or specific medical scenario guidance

Second, the analysis of the smart medical market environment

2.1 Smart medical market environment

According to the National Bureau of Statistics, the national medical expenditure at the end of 2021 was 7,559.36 billion yuan, and China is the world's second largest medical market. Over the past few decades, China has invested heavily in healthcare informatization systems and generated a large amount of data, but most of the data is stored in isolation and unstructured form, and AI applications will further realize the value of data when it is effectively connected, standardized and analyzed by digital means.

In terms of the definition of digital health, in 2019, the World Health Organization issued the Global Digital Health Strategy (2020-2024), which for the first time determined the importance of digital health strategy, defined "digital health" as online medical services, telemedicine and mobile medical services, and promoted the concept of digital health to the world. Frost & Sullivan pointed out that digital health refers to the provision of services or products through digital technology to meet the health needs of individuals. China's digital health market mainly includes: digital medical service market; over-the-counter e-commerce market; health consumer goods e-commerce market; Other markets (IT infrastructure, marketing services, and others) While digital healthcare services are an important part of the digital health market, internet hospitals are the main platform for providing digital health services. In 2015, China established the first Internet hospital, Wuzhen Internet Hospital, setting a precedent for China's "Internet + medical health" innovation model and marking the beginning of the vigorous development of the digital medical service market.

In terms of market size, China's digital health market size was 218.1 billion yuan in 2019 and is expected to increase to 4,222.8 billion yuan in 2030, with a compound annual growth rate of 30.9%. The digital migration rate is expected to increase from 3.3% in 2019 to 24.0% in 2030. According to Frost & Sullivan, the market size of China's digital medical services market was 23.2 billion yuan in 2019 and is expected to increase to 739.5 billion yuan in 2030, with a compound annual growth rate of 37.0%.

Figure 6: China's digital health market size and digital migration

In-depth analysis | Smart Healthcare Industry Research Report

Source: Frost & Sullivan, Essence Securities Research Center

The healthcare big data solutions market refers to a market where service providers provide big data-driven solutions combined with advanced technology applications and medical insights to meet the informatization, digitalization and intelligent synthesis needs of various sectors of the healthcare industry, including hospitals, regulators, and policy makers, life sciences companies and individuals. The size of China's healthcare big data solutions market (based on sales volume of healthcare big data solution service providers) was RMB10.5 billion in 2019 and is expected to grow to RMB57.7 billion by 2024, with a CAGR of 40.5%. The overall penetration rate of medical big data solutions (sales revenue/total investment in China's medical informatization, %) is expected to increase from 7.2% in 2019 to 16.2% in 2024.

Figure 7: Total informatization investment in China's pharmaceutical industry by investment from 2015 to 2024 (estimated) (RMB billion)

In-depth analysis | Smart Healthcare Industry Research Report

In the entire medical information service terminal, the medical SaaS market is a sub-category with rapid growth. According to iResearch, the overall market size of China's medical SaaS reached 3.7 billion yuan in 2020, and will maintain a growth rate of more than 30% in the next five years. The digital transformation of the healthcare industry and the large demand for medical services are the keys to promoting the rapid development of healthcare SaaS.

China has a large patient base and chronic disease patients. The proportion of the population aged 65 and above is expected to rise from 13.5% in 2020 to 21.9% in 2030, and is expected to reach 318.1 million. In addition, with environmental pollution, irregular lifestyle, excessive stress load, etc., the number of chronic patients in China is increasing. In 2021, there were 140 million diabetics and 333 million hypertensive patients in China. The diagnosis and treatment of chronic diseases are characterized by long time, low medical risk, and frequent medication, and regular follow-up and disease records are more important.

Medical SaaS can not only build a corresponding system platform for the hospital to track and record the situation of patients with chronic diseases for a long time, but also provide online consultation services for retail pharmacies outside the hospital to solve the problem of patients purchasing prescription drugs frequently without a doctor's prescription. For this reason, chronic disease management is the most suitable digital solution for all health diseases. In addition, patients demand not only efficient diagnosis and in-hospital medical services, but also ongoing out-of-hospital management. Medical SaaS meets the needs of long-term care by providing chronic disease management solutions for both in-hospital and out-of-hospital institutions to track and record patients' health over a long period of time.

Figure 8: Size of China's healthcare SaaS market from 2018 to 2025

In-depth analysis | Smart Healthcare Industry Research Report

Source: iResearch, 36Kr Research Institute

2.2 Smart healthcare helps the medical industry

Pain point 1: Led by grassroots medical institutions, the distribution of urban and rural medical levels is uneven, and smart medical care helps to fair medical resources.

In 2018, there were 994,681 medical and health institutions in mainland China, of which primary medical and health institutions (including community health service centers (stations), township health centers, village health centers, and outpatient departments (institutes)) were the main ones, accounting for 94.9% of the total. However, there are only 33,009 hospitals (including general hospitals, traditional Chinese medicine hospitals and specialized hospitals), accounting for about 3.3% of the total; In addition, professional public health institutions (Center for Disease Control and Prevention, Specialized Disease Prevention and Control Hospitals (Institutes/stations), Maternal and Child Health Hospitals (Institutes/Stations), Health Supervision Institutes (Centers)) account for about 1.8%. Hospitals, which account for only about 3.3% of health care facilities, supply about 76% of the country's beds; According to statistics, in 2018, the average number of beds in medical and health institutions per 1,000 population was 6.03 (13.7 in Japan, 8.0 in Germany and 2.8 in the United States), the supply of beds is relatively insufficient, and the number of beds in medical and health institutions per 1,000 population is 4.56 more than in rural areas, and the allocation of urban and rural hospital bed resources is unbalanced, and the allocation of medical resources in the eastern and western regions is unbalanced.

We believe that in the future, we are expected to rely on smart medical technologies such as AI and Internet medical care to improve the imbalance of medical resources, thereby contributing to the fairness of medical care.

Figure 9: Primary medical institutions account for a relatively large proportion

In-depth analysis | Smart Healthcare Industry Research Report

Source: 2020 China Health Statistics Yearbook

Pain point 2: The pressure of medical insurance payment is gradually increasing, and smart medical care helps control medical insurance costs. In 2020, 1.36 billion people participated in the national basic medical insurance, and the participation rate in the past five years was basically stable at about 95%, and the medical insurance penetration rate was at a high level. At the same time, medical insurance expenditure has increased year by year, increasing its proportion of GDP, and increasing the pressure on medical insurance payment. In this context, the state has vigorously promoted a new medical insurance payment system based on payment by disease diagnosis related group (DGR) and payment by disease big data (DIP). We believe that behind this system, it is necessary to rely on a perfect electronic medical record system and massive medical big data support, which is expected to drive the rapid development of the smart medical industry.

Figure 10: Health insurance coverage has steadily increased

In-depth analysis | Smart Healthcare Industry Research Report

Figure 11: Health care spending and share of GDP have risen

In-depth analysis | Smart Healthcare Industry Research Report

Source: National Bureau of Statistics, National Health Insurance Administration

Pain point 3: Excessive medical treatment, excessive consumables, unreasonable resource allocation and other medical resources waste phenomenon is serious, under the premise of lack of medical resources, there is still a relatively serious waste of medical resources in the mainland, mainly reflected in three aspects: over-treatment, over-examination and over-seeking medical treatment.

Figure 12: Serious waste of medical resources in mainland China

In-depth analysis | Smart Healthcare Industry Research Report

Pain point 4: Under the influence of the new crown epidemic, the demand for online diagnosis and treatment has increased significantly. The new crown pneumonia epidemic has become a strong catalyst for the development of Internet medical care, and the country has comprehensively promoted Internet diagnosis and treatment to reduce the risk of cross-infection. With the implementation of policies and the change of people's concepts, compared with the previous offline outpatient clinics, the demand for Internet medical treatment such as online consultation and chronic disease management has increased significantly, which is conducive to the development of the smart medical industry.

2.3 Brief analysis of the supply and demand side of smart healthcare

Demand side: The demand side, led by hospitals, is increasingly recognizing innovative technologies

From 2014 to 2018, the budget investment of public hospitals for informatization construction increased year by year, releasing favorable opportunities for medical informatization. With the introduction of smart hospital and other policies, hospitals are paying more and more attention to information systems such as electronic medical records, clinical assisted decision-making, and big data construction and application, and the acceptance is increasing.

Figure 13: 2019-2020 China Hospital Informatization Status Survey - Priority Ranking of Hospital Information System Construction

In-depth analysis | Smart Healthcare Industry Research Report

Source: CHIMA, compiled by Yiou

Supply side: With the intensification of demand for smart healthcare, smart healthcare will become a new growth point for enterprises

Medical informatization enterprises: With the development of medical informatization in China, in the face of large hospitals participating in electronic medical record rating and interconnection, as well as the demand for new or upgraded hospital electronic medical records in more and more small and medium-sized cities, it will promote the increase of industry concentration, and medical informatization companies that seize the opportunity to expand their scale will have greater advantages.

Internet enterprises: Internet enterprises have shifted from information and connection platforms to directly participate in the key links of disease treatment, taking disease prevention as an example, which can play a huge role, thereby directly reducing medical insurance expenditure. This means that the value of Internet medical care will be further released, and the revenue capacity will be more imaginative.

ICT enterprises: The deployment of smart healthcare is accelerating the construction of a modern and refined hospital management system. With profound technology accumulation and industry practice experience, ICT enterprises continue to explore the frontiers of smart medical technology and practice, and promote the accelerated development of smart hospitals and the big health industry.

Smart Healthcare Startups: While improving their technical capabilities, entrepreneurial smart healthcare companies have been exploring commercialization in the past five years. Compare cooperation with B-end suppliers. Such companies are gradually setting their sights on hospital terminals. In the future, the main test for such enterprises is the development of sales channels.

With the intensification of the demand for smart healthcare, smart healthcare will become a new growth point for enterprises, and with the collaborative development and maturity of innovative technologies (such as cloud computing + AI, big data + AI, etc.), Internet giants have laid out the track of smart healthcare based on their own advantages.

In addition to large Internet companies, small and medium-sized enterprises have also ushered in an outbreak. Taking AI companies as an example, up to now, a total of 4 AI medical imaging companies have submitted prospectuses, one of which has been listed and another medical big data solution (using artificial intelligence technology) company has been listed. In addition, the number of financing in medical AI in Q3 2021 has reached 30, much higher than the 19 in the whole of 2020.

Figure 14: Number of financing in the medical AI segment in 2020vs. 2021

In-depth analysis | Smart Healthcare Industry Research Report

Third, the typical general direction application scenarios of smart medical care

3.1 Smart healthcare helps the medical industry

Smart Healthcare will take the construction of a smart medical and health service cloud ecosystem as its vision and goal, integrate and gather resources from all parties in medical and health care through the overall unified medical and health cloud service platform, and provide residents and patients with high-quality and efficient medical and health services with full life cycle, online and offline integration.

With residents' health and medical big data as the core, build a digitally-driven medical and health service system around residents' health management. Use the Internet of Things, industrial Internet, big data platform, data bank and other technical platforms to promote data collection, data integration, exchange and sharing, and realize data interconnection; Combining big data with machine learning, deep learning and other technologies, as well as evidence-based medicine, imaging omics and other disciplines, it provides data-based intelligent applications/services around the all-round medical and health service scenarios of residents, doctors, medical institutions, pharmaceutical companies and insurance institutions, which can optimize the residents' health management service experience, improve the diagnosis and treatment process, and enhance the efficiency of medical behavior. Data interconnection can optimize the experience of each application scenario, and the data generated by each application scenario can further enrich the data - forming a complete value closed loop.

Therefore, the main application scenarios of smart healthcare can be divided into user-centered health management centers, medical service centers and medical business centers that cover the entire life cycle of users. As well as data-based, digital-intelligence-driven medical and health service system business operation centers, regional medical collaboration, and elderly care service centers.

3.2 Health Management Center

The health management center is a service system that builds health management around health, sub-health and disease groups. Using information and medical technology to establish a set of perfect, thorough and personalized service procedures on the scientific basis of health care and medical treatment; Its purpose is to help healthy people and sub-healthy people establish an orderly and healthy lifestyle, reduce risk status, and stay away from diseases by maintaining health and promoting health; Once clinical symptoms appear, they will recover as soon as possible through the arrangement of medical services.

Figure 16: Intelligent health management business process

In-depth analysis | Smart Healthcare Industry Research Report

Source: China Resources Group

and building a common intelligent health management system in the medical and health industry based on cloud platform: using the Internet of Things and big data platform to integrate various heterogeneous data sources related to residents' health to form residents' health records; Using artificial intelligence technology and mobile technology to build health analysis and evaluation models, health prescription systems, tracking intervention systems, intelligent customer service, feedback education systems; Finally, the distributed deployment mode of cloud + pipe + end realizes a hierarchical and orderly health management system.

Figure 17: Connected health feedback system

In-depth analysis | Smart Healthcare Industry Research Report

3.3 Medical Service Center (Internet Hospital)

The medical service center is an intelligent service that integrates online and offline with the goal of improving the patient experience. According to the actual medical needs of patients, promote the deep integration of information technology and medical services, and provide patients with personalized and intelligent services covering the whole process before, during and after diagnosis. Use Internet technology to optimize the medical service process and service model, including online services such as intelligent medical triage, waiting reminder, inter-clinic settlement, mobile payment, in-hospital navigation, push of examination and test results, mutual recognition of examination and test results, self-service printing and inquiry of outpatient emergency medical records, actively promote the construction and application of referral services, telemedicine, drug distribution, patient management and other functions, build online and offline integrated services, realize the organic connection between clinical diagnosis and treatment and patient services, improve the efficiency and quality of services, and improve patients' medical experience to improve satisfaction.

Figure 18: Medical service flow

In-depth analysis | Smart Healthcare Industry Research Report

Understand the patient's condition through multiple rounds of friendly intelligent consultation; Between the patient's colloquial expression of the main complaint and the standard medical terminology, professional medical NLP technology is used to achieve information transformation. Apply authoritative textbooks and massive literature to build a knowledge graph; Combined with the doctor's professional expertise and past diagnosis and treatment experience, portray a comprehensive, detailed and real-time portrait of the doctor; Apply the mechanism of multi-mode combination and mutual verification to accept the best results. Let the patient express discomfort in the simplest way, arrive at the most correct department, and find the most suitable doctor. In addition, it provides accurate in-hospital navigation services based on mobile terminals, integrates guidance and triage, and creates a mobile, self-service and intelligent in-hospital navigation and mobile guidance service system for hospitals, effectively improving patients' experience.

Figure 19: Improving service quality and efficiency with intelligent customer service

In-depth analysis | Smart Healthcare Industry Research Report

Internet hospital is a one-stop Internet medical center relying on physical hospitals, focusing on follow-up consultations and routine consultations, integrating consultation, prescription, payment and dispensing; From the perspective of the scope of diagnosis and treatment, Internet hospitals strictly prohibit the first consultation, mainly focusing on chronic diseases and some common diseases; From the perspective of organizational form, it must be based on physical hospitals and combined with the Internet; The service covers other offline services such as interpretation of examination reports, follow-up prescriptions, rehabilitation guidance and so on. The Internet hospital platform essentially uses Internet technology to develop a new medical service model, expanding the perspective of patient service from the hospital to the whole process, expanding the radius of medical services from the periphery to the region or even wider, and expanding the location of medical services from the hospital to inside and outside the hospital. Internet medical technology mainly includes Internet hospital platform and remote diagnosis and treatment platform. In addition to building an Internet-based technology platform, the prerequisite for its effective operation is more important than business design and efficient operation.

Build a universal Internet hospital platform to assist medical and health institutions with online capabilities, achieve effective integration and sharing of online and offline, and promote the extension of medical business.

Figure 20: Internet hospital platform architecture

In-depth analysis | Smart Healthcare Industry Research Report

Source: China Resources Group

3.4 Medical Business Center

The medical business center is actually to build an Internet hospital platform within the hospital, promote the "Internet +" of the physical hospital business, and carry out the optimization and upgrading of medical business and system around electronic medical records with the goal of empowering doctors: use the digital platform to reconstruct the existing functions as components; Use multi-modal flexible electronic medical record entry to improve the efficiency of doctors' medical record entry; Improve the standardization of medical care with dynamically updated clinical pathways; Adopt big data and artificial intelligence technology to build a clinically assisted decision-making and quality control system to improve the efficiency and quality of doctors' diagnosis and treatment.

In-depth analysis | Smart Healthcare Industry Research Report

Figure 21: Hospital Business Internet+

Source: China Resources Group

Clinical pathway management is the starting point to improve medical quality and standardize medical behavior

Clinical pathway (CP) is the best programmatic and standardized medical examination and disposal process specified with strict work order and accurate time requirements for a well-diagnosed disease or surgery, based on evidence-based medicine, with the purpose of expected treatment effect and cost control.

Formulation of clinical pathway: With the disease as the core, establish a set of basic process mechanisms for automatically formulating clinical pathways based on statistical principles, break through the limitations of traditional technical means and inability to quickly formulate clinical pathways in large quantities, analyze the best setting scheme of clinical pathways through big data processing methods and rapid processing of medical order data, and change the traditional manual way and empirical judgment-based methods of formulating clinical pathways.

Figure 22: Execution of the clinical pathway

In-depth analysis | Smart Healthcare Industry Research Report

With the national diagnosis and treatment norms as the standard, artificial intelligence and big data technology as the support, integrating evidence-based medicine and empirical medicine, collecting the data required for diagnosis and treatment through data interconnection with other medical information systems, and realizing fast and accurate intelligent comprehensive analysis and judgment through the intelligent reasoning engine of the data core layer, providing doctors and nurses with auxiliary services for intelligent consultation strategies, physical examination, auxiliary examination recommendations, reference diagnosis basis strategies and treatment suggestions.

Figure 23: Assisted decision-making system flow

In-depth analysis | Smart Healthcare Industry Research Report

Source: China Resources Group

In general, the medical business center is committed to realizing the whole process management of medical quality, based on the clinical medical knowledge base engine, to realize the transformation from "terminal quality" to "link quality", from "post-event reminder" to "instant feedback", from "transaction management" to "intelligent management", from "experience management" to "standardized management", from "single management" to "all-round refined management".

Figure 24: Medical Business Center

In-depth analysis | Smart Healthcare Industry Research Report

3.5 Business Operation Center

Business operation centers: DRGs, RPA, RFID

A system for medical service monitoring, analysis and performance appraisal based on DRGs.

DRGS (Diagnosis Related Group) refers to a system that uses the principles of statistical control theory to divide patients into several diagnostic groups for management according to factors such as age, disease diagnosis, comorbidities, complications, treatment methods, disease severity and outcome. DRGs are considered to be a "patient-centered" medical record combination system that comprehensively considers the severity and complexity of the disease, as well as the number of medical services and the intensity of medical resource consumption. DRGs incorporate workload performance management and incorporate relevant indicators into internal performance assessment and distribution, which can reflect the tilt towards clinical departments and important positions with high clinical front-line, high work risk and high technical difficulty, and fully reflect "good work, more reward, excellent performance and good reward".

Figure 25: DRGs system

In-depth analysis | Smart Healthcare Industry Research Report

The DRGs medical service monitoring and analysis system is divided into four roles, namely the president, the medical office, the medical insurance office, and the clinical department, and the corresponding system functions are set according to the needs of the role.

RPA is used to automate the repeated standardization process of hospitals, and the main applicable scenarios of RPA are: clear rules and logic, cross-system data integration, data collection, retrieval, aggregation, and daily repetitive work

Figure 26: Expected value of RPA implementation

In-depth analysis | Smart Healthcare Industry Research Report

The following are the specific scenarios of RPA implementation:

Procurement management: supplier access and supervision (automatic monitoring of supplier access and automation of regular supplier assessment and inspection; Company association analysis; Leather bag company testing; Dynamic supplier risk monitoring), dynamic monitoring of purchase price (realize dynamic procurement price monitoring, provide a timely, nationwide reference basis for hospital procurement, and compare the historical procurement data of the hospital)

Drug management: drug use monitoring (comprehensive drug monitoring according to year-on-year and chain analysis of drug consumption according to departments, doctors and other dimensions to reduce the proportion of drugs; Promote rational drug use), drug price monitoring (crawling external bidding and purchasing prices through crawler technology, creating a real-time drug price database, as a reference standard for drug price monitoring)

Consumables management: high-value consumables management (regularly check the use and charges of high-value medical consumables, monitor the use of high-value consumables), medical consumables monitoring (under the requirements of the Health Commission to reduce the proportion of consumption, assist hospitals to comprehensively monitor the use of consumables, reduce consumables consumption, and control costs)

Financial management: collection reconciliation (regular automatic reconciliation with the payer, social security/commercial insurance statements, patient information, electronic prescriptions, case homepage, etc.), payment review (pre-payment data preparation; Matching of contracts, orders, payment applications and information verification; Payment entry generation)

Independent supervision: audit trail discovery (analysis of data from different modules, use audit robots to carry out analysis in specific areas, identify high-risk areas and audit trails through analysis results), audit execution (sample sampling, data comparison, and analysis through robots to form analysis discrepancies for the audit team to perform further audit procedures. )

RFID realizes all-round and intelligent management of medical assets

Based on RFID and other Internet of Things technologies, the medical asset information is transmitted to the platform data center, so as to realize the hospital's multi-directional tracking and management of medical asset usage, life cycle, intelligent inventory, equipment location, safety maintenance and so on. The establishment of the platform will realize the modernization, scientific and precise management of hospital assets, solve unnecessary waste of human resources and costs, and manage more scientifically, accurately and effectively.

Figure 27: Hospital fixed assets

In-depth analysis | Smart Healthcare Industry Research Report

3.6 Regional medical coordination

Regional medical collaboration: Build a telemedicine platform, effectively promote regional medical collaboration to build a cloud-based telemedicine cloud platform, realize the support of remote consultation, two-way referral, remote appointment, video conference, remote specialty diagnosis, distance education, remote digital resource sharing and other services, and promote the H2H2C medical service model.

Figure 28: (Big H)2 (Small H)2(C) Service

In-depth analysis | Smart Healthcare Industry Research Report

Its functions mainly include: (offline, interactive) remote consultation, two-way referral, remote appointment (expert appointment, examination appointment, test appointment), video conference, remote specialist diagnosis (imaging, pathology, ECG, ambulatory ECG, ambulatory blood pressure, surgery\ultrasound\endoscopy, monitoring), distance education, remote digital resource sharing.

Regional medical collaboration: technology-driven efficient collaborative use of information technology for regional medical resources, connecting medical resources (including doctor resources, medical equipment resources, medical data, medical knowledge, etc.) in the region, realizing resource sharing and collaboration between medical institutions, implementing the hierarchical diagnosis and treatment model and two-way referral system, while improving the medical level and service quality of each hospital, fundamentally solving the problem of difficult and expensive medical treatment for patients.

Figure 29: Regional medical collaboration

In-depth analysis | Smart Healthcare Industry Research Report

Regional medical collaboration: The telemedicine platform architecture relies on Internet of Things technology, cloud computing, 4G/5G, remote sensing/telemetry/remote control, mobile Internet, AR\VR and other technologies, and builds a cloud telemedicine platform around electronic medical records, including telemedicine applications, telemedicine operations, and customer service ends.

Figure 30: Telemedicine

In-depth analysis | Smart Healthcare Industry Research Report

The telemedicine platform also includes: distance education, remote academic exchange, remote surgery, and remote diagnosis and treatment.

Regional medical collaboration: Cloud PACS provides seamless offline and online medical image sharing services to build a cloud-based telemedicine cloud platform, which supports remote consultation, two-way referral, remote appointment, video conferencing, remote specialty diagnosis, distance education, remote digital resource sharing and other services, and promotes the H2H2C medical service model.

Figure 31: Diagnostic imaging flow

In-depth analysis | Smart Healthcare Industry Research Report

Regional medical collaboration: AI empowers the quality and efficiency of medical services

With the continuous advancement of computer technology and medical imaging technology, medical imaging has gradually developed from auxiliary examination methods to the most important clinical diagnosis and differential diagnosis methods of modern medicine. However, there are many problems in the field of medical imaging, including unbalanced supply, high misdiagnosis and missed diagnosis rate, and slow reading by imaging doctors. The combination of artificial intelligence and medical imaging can solve the problems in the field of traditional medical imaging.

3.7 Elderly Care Service Centers

Elderly care service center: build a smart elderly care service model of "system + service + elderly + terminal"

Using advanced information technologies such as the Internet of Things, Internet, mobile Internet technology, intelligent call, cloud technology, GPS positioning technology, etc., we will build a smart elderly care cloud service platform, covering various forms of elderly care such as institutional pension, home care, and community day care, and create a smart elderly care service model of "system + service + elderly + terminal". Through cross-terminal data interconnection and synchronization, the platform connects various departments and roles to form a complete closed loop of intelligent management, realize the information interaction between the elderly and their children, service institutions and medical staff, effectively monitor the physical state, safety situation and daily activities of the elderly, and timely meet the needs of the elderly in life, health, safety, entertainment and other aspects.

Fourth, AI empowers typical small field application scenarios of smart healthcare

4.1 CT image recognition

AI+CT image recognition has a wide range of applications. The main product forms of AI+CT imaging include: image analysis and diagnosis software, CT image 3D reconstruction system, target automatic delineation and adaptive radiotherapy system. Through intelligent CT image recognition, it can complete case screening, intelligent analysis and diagnosis, and assist clinical diagnosis and treatment decision-making. From the perspective of application scenarios, it mainly includes chest, limb joints and other parts, breast, heart and lung, coronary artery, bone and other organ tissues, and has a wide range of applications.

Figure 32: AI+CT image recognition products approved and registered by the National Medical Products Administration (three categories)

In-depth analysis | Smart Healthcare Industry Research Report

Source: Huaan Securities

China's image analysis data is growing rapidly, radiologists are facing a shortage, and AI+CT image recognition technology has great potential for development. At present, the image data of China's CT industry is exploding, and the market size will reach 2.47 billion yuan in 2019, with a CAGR of about 30%. On the other hand, at present, there are only more than 80,000 radiologists in mainland China, but the annual diagnostic workload has reached 1.44 billion images, combined with the annual growth rate of China's medical imaging data of 30%, and the corresponding annual growth rate of radiologists is only 4%, there is a huge gap between the two, AI+CT image recognition is expected to make up for this gap, and the development potential is huge.

Figure 33: Radiologists and CT image growth projections

In-depth analysis | Smart Healthcare Industry Research Report

Figure 34: Market size of CT examination industry in China (billion yuan)

In-depth analysis | Smart Healthcare Industry Research Report

4.2 CDSS assists clinical diagnostic decision-making

Clinical Decision Support System (CDSS) generally refers to a computer system that provides auxiliary support for clinical medical decision-making based on artificial intelligence deep learning algorithms. CDSS inputs medical-related guideline literature, expert consensus and electronic medical record data, and outputs models in clinical diagnosis through big data analysis and neural network operations based on artificial intelligence, so as to assist doctors in providing clinical diagnosis of relevant cases.

The main functions of CDSS to assist clinical diagnosis decision-making include: auxiliary diagnosis, treatment plan recommendation, similar medical record recommendation, and medical order quality control.

A Smart medical policy continues to be favorable, and the future CDSS development prospects are good: According to the data released by the Eggshell Research Institute, the bidding of projects involving CDSS products or services reached 42 in 2019, and with the favorable smart medical policy, CDSS projects will continue to increase in the future.

Figure 35: CDSS project bidding from 2015 to 2019

In-depth analysis | Smart Healthcare Industry Research Report

4.3 Retinal image recognition helps the diagnosis and treatment of chronic diseases

Retinal imaging is widely used in clinical practice. The retina is the only part of the body that can directly observe blood vessels and nerve cells in a non-invasive way. The risk of chronic diseases can be detected, diagnosed and evaluated by observing changes in blood vessels and nerve cells in the fundus, including: eye diseases: diabetic retinopathy, pathological myopia, retinal vein occlusion, glaucoma and age-related macular degeneration; Other chronic diseases: hypertension, diabetes, ICVD, Parkinson's disease and anemia. A number of artificial intelligence retinal image recognition medical device products have been registered worldwide.

Figure 36: AI retinal image recognition product/software

In-depth analysis | Smart Healthcare Industry Research Report

The market potential of artificial intelligence retinal medical image recognition technology is huge.

Medical and health scenarios: Compared with traditional retinal imaging methods, artificial intelligence retinal image recognition technology has the advantages of high diagnostic efficiency and high diagnostic accuracy, which can help doctors shorten image analysis time, effectively reduce medical costs, and alleviate the imbalance of medical resources.

Big health scenarios: Artificial intelligence retinal medical image recognition technology can provide customized products and services for health customers to meet their health risk assessment and management needs.

Figure 36: AI Retinal Medical Imaging Market Size (2019 to 2030 (estimated)

In-depth analysis | Smart Healthcare Industry Research Report

5. Pain points and prospects of smart medical industry

5.1 Online Healthcare

Industry pain points:

  1. Medical demand is considerable, but the market acceptance of online healthcare is low, and enterprises lack a clear profit model.

From the current situation of domestic medical treatment, the market structure of supply is less than demand is the direct factor of the formation of difficult and expensive medical treatment, and medical demand is relatively rigid demand, coupled with the government's regulation and supervision of the medical economic system, market demand will not change too much due to the shortage of resources and price changes, under the trend of chronic disease aging, the scale and structure of rigid demand are gradually expanding, medical demand only increases, and the continuous growth of medical demand is also the driving force of the derivative online medical industry.

However, the resource-oriented characteristics of the medical industry establish high barriers for start-ups and cross-border enterprises, and security and authenticity are important factors for consumers in the medical industry, and these two characteristics are precisely the shortcomings of the Internet industry. Online medical treatment starts from free diagnosis and treatment consultation, in the free consultation stage, consumers out of curiosity or profit attitude to contact online medical APP, but consumers do not give enough trust to the online consultation platform, online medical APP has not become a substitute for physical hospitals, so it is difficult for online medical enterprises that switch to a charging model to develop a larger market in diagnosis and treatment service business, and key charging items can only rely on registration business and hospital information system services. Many Internet medical companies still have not found a reasonable profit model.

(2) Data sharing barriers limit the depth of development of the online medical industry to a certain extent.

The data of each individual in China's medical system is decentralized, and it is difficult to form an effective sharing and circulation mechanism between data. Due to the long-term non-open and difficult standardization of medical data, on the one hand, doctors cannot obtain comprehensive data of patients' past visits for in-depth diagnosis, on the other hand, the fragmentation of data obtained by online medical companies seriously leads to so-called big data analysis can only stay on the surface. Therefore, the basis of medical treatment requires the research and analysis of the first diagnosis data and previous medical history, and how to share some data with the hospital information system within the business scope and unify the disease standard is an important issue facing online medical treatment.

(3) Online medical reimbursement is difficult, and medical insurance payment problems directly affect consumer behavior

The reason why the online medical industry is more popular than consumption and only a few patients seek medical treatment is that on the one hand, the problem of consumer trust is the problem, and on the other hand, the reimbursement problem is also prohibitive for many consumers. Yinchuan can attract a large number of online medical enterprises to carry out Internet hospital business in the local area, which is closely related to local government policies.

Future Outlook:

The online medical industry will reshuffle and return to the essence of medical treatment, and the establishment of industry standards and health record databases should be promoted as soon as possible, and medical insurance reimbursement policies in the online medical field should be introduced. At this stage, the most important contradiction in the medical industry is the imbalance between supply and demand, and the online medical industry needs to implement the model according to market demand in the future. The market demand-oriented business model will inject a steady stream of vitality into enterprises and is also the basis for foreign cooperation of medical institutions.

At present, most of the business models of the online medical industry are similar and pay more attention to registration and consulting business, and in the future, enterprises should find the market positioning, find the differences in the needs of different types of people, integrate the upstream and downstream advantages of the online medical industry to provide accurate services, and solve the real needs of customers in medical treatment. The improvement of this model is inseparable from the support of industry standard specifications. Based on unified industry norms, setting minimum standards and levels of institutions and their businesses, clarifying the responsibility systems of different institutions, and protecting patient safety and privacy rights, enterprises will have rules to follow in conducting business, which is the basis for the healthy development of the industry and the basis for customers to recognize enterprises. Realize the queryability and data structure standardization of personal health data, realize information sharing between medical institutions, and promote the investment of social capital in online diagnosis and treatment.

5.2 AI Medical Imaging

Industry pain points:

(1) The data foundation is weak, and the application scope and effect are limited.

The application of AI in the medical field is inseparable from the support of data, and there are currently a series of problems in medical data in China: First, the development level of hospital digitalization is different, the IT system construction of remote areas or grassroots hospitals is backward, the amount of data storage is small, many grassroots hospital networks are not connected, or only traditional ADSL asymmetric networks, and the basic equipment is relatively backward, resulting in obstacles in data storage, while it is difficult for doctors to enter patient data by handwriting or other offline records that are difficult to be recognized and analyzed by computers. As a result, there is a lack of channels for sharing and reusing data.

(2) There is a shortage of cross-border talents in AI and medicine, and the development process of the industry is slow.

Artificial intelligence is moving from the laboratory to the clinical application of medicine, and is in the period of technological sprint and application exploration before the big breakthrough in the industry. At this stage, high-end talents who can promote technological breakthroughs and creative applications play a vital role in the development of the industry. The ideal talent should have deep knowledge in medicine, computer science and mathematics at the same time, and have sufficient research and development capabilities.

The development of medical AI is closely related to the quantity and quality of talents. Specifically, medical AI needs talents in two aspects, namely medical talents and AI talents, and the most needed is cross-border talents who understand both medicine and AI. Application and platform developers should not only study artificial intelligence algorithms, but also have an in-depth understanding of medical expertise, and talents with a composite background of artificial intelligence + medical care are the core competitiveness of enterprises.

However, high-quality medical talents and AI talents are very scarce, and comprehensive talents with both capabilities are rare. On the one hand, doctors often have a relatively single knowledge structure, rarely have an interdisciplinary learning background, are very insufficient in data processing, engineering experience, system development, etc., and have very few doctors with AI research and development capabilities. On the other hand, AI algorithm talents themselves are few, and they generally lack experience in the engineering development and implementation of IT systems, and medical knowledge is also lacking. At present, the entire medical AI industry is facing a serious talent shortage, especially the shortage of cross-type talents with medical, computer, and mathematical research and development capabilities, which seriously hinders the development of the medical AI industry and slows down the development of the industry.

(3) The industry lacks policies and legal standards, and the division of responsibilities of relevant groups is not clear.

According to the new version of the "Classification Catalogue of Medical Devices" issued by the State Food and Drug Administration (CFDA), if the diagnostic software provides diagnostic suggestions through algorithms, only has auxiliary diagnostic functions, and does not directly give diagnostic conclusions, it will declare Class II medical devices, and if the lesion site is automatically identified and clear diagnostic tips are provided, it will be managed according to Class III medical devices. The second category of devices has a clinical trial exemption catalog, and whether the diagnostic software declaration can enjoy the exemption, the CFDA has not yet made specific specifications. If each medical AI company wants to open up the road of hospital procurement, it must obtain CFDA certification. To go through the CFDA certification process, there are not only many approval links, but also long clinical trials, which require continuous resource investment, which is an obvious burden for start-up medical AI companies and affects their commercialization process. On the other hand, medical AI companies are also facing potential legal risks. There is no clear legal definition of AI, let alone a complete legal system.

When medical AI products are widely popularized, once a medical malpractice occurs, the determination of liability will be a tricky issue. For AI systems, people have high expectations, such as 100% diagnostic accuracy, and no mistakes. Even senior medical experts cannot reach the level of zero errors, and it is unreasonable to have such strict requirements for AI systems. Therefore, the development of AI medical imaging needs to give strict standards from the policy and legal levels, which not only limit the scope of the market, but also point out the direction for the development of enterprises.

Future Outlook:

Cloud + AI has become the main model, and we should focus on giving full play to and deepening the auxiliary role of artificial intelligence in the field of imaging. Cloud computing, big data and AI are in the same vein and have a close relationship. The docking of hospital information system and cloud system can realize data sharing between different departments within the hospital, between different hospitals and even between medical institutions across provinces, and a large number of real-time medical data is aggregated in the "cloud" to provide a data basis for the training of AI systems. The cloud data center undertakes the task of AI model training. Medical data will be transmitted from each hospital to the cloud in real time, greatly accelerating the progress of AI model development for new diseases and optimization of existing models. When a new disease AI model is developed, or the original disease model has a major update, only the system update needs to be performed in the cloud, and the latest medical AI service can reach all hospitals deployed at the same time.

The promotion of hierarchical diagnosis and treatment policies has promoted the popularization of third-party medical institutions such as medical alliances and regional imaging centers. With the help of the medical cloud platform, data exchange can be realized between upper and lower level hospitals, which in turn lays the foundation for referral. General diseases can be treated in primary medical institutions, and patients with serious diseases can be referred to large hospitals, and patient information can be synchronized to the hospital in a timely manner. Medical images from basic medical institutions can be uploaded to higher-level medical institutions or regional imaging centers, allowing doctors in imaging centers to read films efficiently with the assistance of AI systems.

5.3 Digital Humans

Industry pain points:

The non-standardization of information and the vague diagnosis and treatment concept of traditional medicine make the process of digital human slow. Whether it is collecting patient sign information, symptom information, or patient complaint information, it is analog information, which makes it more difficult to digitize this information. Therefore, the relationship between each custom analog information and the number is used to solve this problem. There is no standardized way for the industry to digitize clinical medical record information, which is also the dilemma of artificial intelligence facing simulated clinical information. Patients have more expectations for digital information, because the digital wave has spread in all walks of life in the past ten years. People's lives have become increasingly surrounded by numbers. However, the slow progress of digitalization by medical institutions, coupled with the agreement on diagnosis and treatment standards, the rigor of medical industry supervision also restricts the application of digital technology in the medical industry.

Understanding the human body is inherently very vague, especially for disease states, so the use of simulated information as a form of display of fuzzy states has been accepted by the world since the beginning of medicine. Western medicine has a good digital foundation because it integrates basic disciplines such as mathematics, physics, chemistry, and biology to solve problems. Traditional Chinese medicine is very different. The vague concept of TCM is far better than Western medicine, which makes the digitization of TCM more difficult. Even people are already familiar with the vague diagnosis and treatment methods of traditional Chinese medicine, and once explained digitally, it is difficult for people to accept. For example, the concept of cold and heat cannot be explained by temperature, and the concept of surface cannot be explained by physical measurement. Nevertheless, the digital collection of human body data and the collation and analysis of all human information to form a complete digital human are still expected. Online medical care, artificial intelligence medical services, drug research and development, and robot applications all have huge dependence on digital humans. It can be considered that the development of digital humans will promote the progress of the entire health industry. Therefore, it is urgent to solve the bottleneck faced by the development of digital humans.

Future Outlook:

Building a systematic digital human backup information is the foundation for the development of smart healthcare. The digital wave is driving the development of various industries, and the digitalization of the healthcare industry will be based on digital humans. It is to be expected that digital human backup will accompany everyone. With the backup of digital humans, through the storage and computing of the cloud, when a person seeks to treat diseases, digital humans will be the necessary foundation for this diagnosis and treatment. With digital human backup, Internet medical care will become a huge application place, and everyone can combine human health information and diagnosis and treatment suggestions through digital communication in the cloud to conveniently obtain diagnosis and treatment results.

Digital humans will make precision medicine and personalized diagnosis and treatment a universal application field. Digital accuracy will go far beyond analog information, making diagnosis more accurate and treatment more personalized. Digital humans are the infrastructure for the application of artificial intelligence in medicine. Artificial intelligence relies on numerical information in machine learning and even language annotation. AI-enabled medical practices will create intelligent doctors. Intelligent doctors need the entry of digital humans as a diagnosis and treatment object or assistance. Intelligent doctors will solve a series of problems for the big health industry: shortage of doctors, uneven medical skills, non-standardization of medical information, and accuracy of diagnosis and treatment. Intelligent robots need digital humans as objects of conversation. Intelligent decision-making systems require big data analysis, so digital humans will be the basic unit of big data. Therefore, building digital humans and building systematic digital human backup is a necessary element of smart healthcare.

5.4 Virtual Assistants

Industry pain points:

Semantic associations and standardized terminology affect the accuracy of virtual assistant judgments. The problem of sentence standardization is significant for both patients and doctors. Each doctor has his own medical record writing habits, each of which has slight differences in the way of expressing diseases, some abbreviated, some in English, some in the general category of diseases, and some with specific symptoms; Patients only know where discomfort or pain is about their physical condition and the disease, and sometimes the expression of the disease is not accurate or even wrong, and it is difficult to think of related lifestyle habits and some seemingly unrelated diseases so that they will ignore key information, and in fact, the same disease often produces concurrent symptoms in different parts.

Non-standardized expressions from doctors affect the accuracy of electronic medical record entry and the structuring of medical record databases, and non-standardized expressions and information loss from patients directly affect the disease judgment results. In order to reduce the occurrence of such situations, many companies develop virtual assistants that use a selective way to communicate with the application object, but from another point of view, this communication method requires a large pool of questions and has a limited range of applications.

Data acquisition is difficult and there is a lack of rational use standards. In recent years, Internet medical companies have emerged in the market, and cooperation with hospitals has also increased. In general, the hospital's external cooperation is relatively conservative, which is related to its own business nature and unit attributes, medical data is related to life safety and patient privacy, and the hospital's cooperation in data needs to ensure the data security and privacy of patients. At present, AI medical imaging companies cooperate with hospitals more, and their corresponding diseases are relatively single, while virtual assistants involve a large number of disease types and knowledge storage, and their development and application require a certain amount of time and data accumulation, the market application degree is relatively weak, the intuitive effect is not obvious, and enterprises encounter certain resistance in obtaining data. In addition, ethical issues of data use await the introduction of standards. At present, the relevant departments have not issued detailed regulations on the scope of use, reasonable use, use object, and period of use of medical data, which also makes the cooperation between hospitals and enterprises have little effect.

Future Outlook:

The development of virtual assistants needs to focus on establishing a complete knowledge map and playing a role in prediction and consultation. In the future, the application of virtual assistants in the medical field should not only play the role of chatbots, but need to solve the real needs of users based on medicine itself. The medical industry has strong professional attributes, and in order to achieve in-depth communication and assistance with users, it is necessary to establish a comprehensive medical data knowledge base, which is also the basis for the development of virtual assistants. The positioning of virtual assistants should not be too high, otherwise it is easy to cause conflicts between doctors and patients; It should not be too low, otherwise the effect will not be much and the meaning of existence will be lost.

Virtual assistants should become tools to assist doctors and patients, evaluate the basic situation of patients before the doctor's diagnosis, predict possible incurable diseases in advance and prepare, reasonably plan the time of medical treatment, and avoid the situation that a single patient in the early stage is too long and the later time is too short in a home visit period. For patients, virtual assistants need to play their role in health management, medical knowledge consultation, and initial diagnosis of symptoms, popularize health knowledge for patients, guide medical departments, and remind them of pre-treatment preparations and precautions.

5.5 AI Health Management

Industry pain points:

The health management payment has not yet formed a clear system, and the corporate profit model needs to be improved. At present, most of the health management services provided by insurance companies to customers and the health management services of enterprises to employees stay in the aspect of physical examination, and are not common, and most of the health management services in addition are paid by individuals. The intervention of artificial intelligence in the field of health management inevitably involves the problem of payment, and if it is simply paid by individuals, its profit prospects are not optimistic for enterprises. At least in the current situation of rising domestic consumption level and weak public health awareness, it is difficult for consumers to quickly and obviously see results in health management under the toC payment model, and the market acceptance is limited. The profit model of AI health management enterprises still needs to be improved: toC or toB model is more mature? Is the focus of the toB model for the government, insurance companies or general enterprises? Are you working with people who are beyond governments, patients, hospitals, and insurance institutions? These are all important issues that need to be considered in the market-oriented development of AI health management enterprises.

Future Outlook:

In the future, AI health management should formulate personalized standards according to individual conditions and link up with related industries to achieve business upgrades. The traditional sense of physical index health level is based on big data based on a certain probability, but in fact, everyone's living habits, environmental conditions, genetic composition, genetic history, etc. are different, and it is not completely accurate to measure the health level of all people with a universal and unified standard. In the future, AI health management should be based on the genetics, genetics, lifestyle and other factors of the human body, formulate personalized health management plans and monitor and warn in real time. On the one hand, it can connect with medical institutions, respond to some temporary emergencies in time, and send patients who need emergency care to nearby hospitals as quickly as possible; At the same time, health management enterprises with AI technology as the core can carry out cross-industry cooperation to achieve rationalized data accumulation and linkage support for different intelligent application scenarios, such as strategic cooperation in smart homes, smart terminal devices and other fields, open up personal basic data and living habit data, and smart wearable device cooperation, integrate health data, and formulate more reasonable health plans.

5.6 AI drug discovery

Industry pain points:

The characteristics of artificial intelligence "black box" affect the developer's recognition of AI drug development, and performance monitoring is the focus. In the traditional sense, drug research and development needs to clarify the production mechanism of target discovery, compound screening and other processes to ensure the logical rigor of the research and development process. Drug development with the participation of artificial intelligence is difficult to explain and requires a large amount of data accumulation, which is also the key factor for artificial intelligence to have no mature products in drug research and development. Whether the future supervision and entry criteria for AI drug development will be very different from traditional artificial drug development is unknown, the basis of drug discovery lies in the accurate knowledge and understanding of diseases, whether artificial intelligence can accurately understand diseases and arrive at reasonable research and development solutions requires the support of big data and the maturity of algorithms. How to enter the market in the form and endorsement of drug research and development under new technologies, and how to ensure the safety and effectiveness of drugs are all issues worth pondering, and they are also issues that need to be confirmed before they are truly implemented.

Future Outlook:

The application of artificial intelligence in drug research and development is still in the introduction period, and in the future, small pharmaceutical companies and domestic pharmaceutical companies are expected to weaken the long-term monopoly of foreign companies, and the coordinated development of all players in the ecosystem is a key channel. If the application of AI in other fields is to reduce the burden on doctors and improve patient satisfaction, the ultimate goal of AI drug development is to improve productivity. The complexity and long-term nature of drug development also make AI companies less involved in the field and slower to progress. Compared with large foreign companies, small enterprises are more flexible, subject to less internal restraint and market attention, and are more conducive to personalized innovation.

Foreign enterprises are large-scale, cumbersome approval mechanisms, and subject to strict supervision by the market, and the transformation and upgrading process under new technologies is relatively slow. The emerging drug research and development method with artificial intelligence technology as the core has also given domestic pharmaceutical companies the opportunity to overtake on the curve. However, in order to achieve the rapid development of AI drug development, it is also necessary to cooperate within the medical ecosystem and the roles of various members of emerging information technology companies, such as sharing data, establishing scientific knowledge graphs and medical record databases, and data types such as genomics data, bioomics data, proteomics data, etc. Or technology companies and pharmaceutical companies jointly develop drug development models, there are many unknowns and challenges under new technologies, and the formation of cooperative alliances can achieve a win-win situation faster.

Source: Misawa Research Institute, infringement must be deleted!