Medical Network September 11th After several years of development, the pattern of medical artificial intelligence industry has gradually become clear.
Recently, health statistics statistical analysis of the establishment of entrepreneurial enterprises and listed companies in the medical artificial intelligence industry found that as a tool to improve efficiency, medical AI has covered four major links in the medical industry chain. Among them, the medical link is mainly to serve patients, and provide a series of more accurate and efficient medical services for patients. The medical , medical insurance, and hospital links are more for the medical institutions and enterprises at the B end.
Building an ecosystem around serving patients
In the medical field, patients are always the most core users, and medical artificial intelligence is no exception. At present, most medical artificial enterprises in China prefer to develop products with patients as the main target, and have already successfully implemented products. Specifically, the services provided by medical artificial intelligence around patients can be divided into four parts: health management, intelligent diagnosis, intelligent treatment and intelligent rehabilitation.
In terms of health management, artificial intelligence can collect massive data and analysis results, and design individualized health management programs for individuals to identify and reduce disease risks and help people to manage health in a forward-looking manner. From the usage scenarios, the current focus is on risk identification, virtual nurses, mental health, online consultation, health interventions, and health management based on precision medicine. However, there are currently not many AI companies involved in health management in China, and the application focuses on the management and prevention of specific diseases to achieve health interventions.
In terms of intelligent diagnosis, modern medicine judges whether a person is sick or has any disease based on various biochemical and imaging examination results. Timely and accurate detection of early diseases can effectively improve the cure rate and survival rate of patients, and save patients the cost of treatment. At present, artificial intelligence is particularly mature in the application of image recognition compared to other sub-areas. Zheng Yongsheng, vice president of Yitu Medical, said in an interview with the health point that the image standardization level is relatively high, and early attempts to apply artificial intelligence. At this stage, the application of medical imaging in lung nodules, fractures, and bone age assessment is rapidly developing.
Intelligent treatment, or "artificial intelligence + auxiliary diagnosis and treatment" is the most important application scenario of artificial intelligence in the medical field. Specifically, it allows the computer to "learn" the medical knowledge of the expert doctor, simulate the doctor's thinking and diagnostic reasoning. And get a reliable treatment plan. Globally, in the application of “artificial intelligence + auxiliary diagnosis and treatmentâ€, IBM Watson is a relatively mature case in the world, and other technology giants such as Google, Microsoft and Baidu are also actively entering.
In terms of intelligent rehabilitation, robot-based artificial intelligence aids are currently the most common. Some reports predict that the compound annual growth rate of generalized rehabilitation robots in the next five years will be about 37%, of which the annual compound growth rate of exoskeleton robots is 47%, which is much higher than the average growth rate of other types of medical robots. It is estimated that by 2020, the global market for exoskeleton robots will exceed $1.8 billion.
While most start-ups are trying to lay out some of the segments, there are also some companies whose product logic is relatively complete, and they begin to lay out products that serve the entire process of patients.
In the case of Philips, this established company has begun to build an integrated solution of “health careâ€, covering its small home appliances, users' wearable devices and related medical information technology, which can radiate pre-hospital health management and disease. Screening, diagnosis and treatment of diseases in the hospital, and post-hospital disease rehabilitation and chronic disease management processes. It is understood that in Philips, 25% of scientists are conducting about 250 research projects related to AI and big data, and closely integrated with clinical scenarios and workflows, including natural language processing, big data mining and analysis, and building structured clinical Database, image recognition, image-assisted diagnosis, interventional therapy, genomics, chronic disease management, home care, cloud platform solutions, etc.
How does Philips do it in terms of intelligent health management, intelligent diagnosis, intelligent therapy, and intelligent rehabilitation? From the perspective of intelligent health management, Philips has developed a fatal fall warning product that uses artificial intelligence technology to synthesize real-time information collected by millions of patients' historical medical data and monitoring equipment to achieve predictive analysis models. A small change in user activity and pace was identified 30 days in advance to cause a fall, to track subtle changes 48 hours before cardiac arrest, and to predict cardiac arrest.
From the point of view of intelligent diagnosis, Philips' machine learning algorithm can make the error rate in the detection of lung nodules of 4mm-30mm size less than 1%, and the sensitivity (85.3%) and specificity (93.9%) reach an excellent balance. The stability of the detection algorithm is much higher than that of the radiologist, and the doubling time and percentage of growth can be calculated to support the benign and malignant risk assessment.
In terms of smart treatment, Philips has developed a personalized treatment product for liver cancer. Philips uses its own NLP technology to derive clinically relevant information from unstructured reports, correlating information from multiple reports/departments with time. Combined with machine learning, all clinically relevant information for patients with liver cancer can be presented to the doctor in chronological order. It is understood that many doctors have spent about 20 days to extract useful information from 200 unstructured reports. Zhou Zijie, chief scientist of the Philips China Research Institute, said that with the help of this technology, doctors can spend 85% less time getting clinical information from unstructured reports.
In terms of intelligent rehabilitation, the “Cerebral and Cerebrovascular Family Care and Rehabilitation Program†developed by Philips in cooperation with the First Hospital of Peking University in 2017, through the information system of connected families, professional medical care institutions and hospitals, tracked the postoperative rehabilitation of patients and improved Postoperative patient's own disease rehabilitation management ability. Professor Huo Yong, the chief physician of cardiovascular medicine at Peking University First Hospital, once said, “This management system is very effective. From our macro scientific data, if these patients with cardiovascular and cerebrovascular diseases can be effectively managed after discharge, they can add extra Reduce the incidence of cardiovascular and cerebrovascular events by 30%-40%."
Extend to the industry upstream and downstream
While building a product ecosystem with patients as the core, artificial intelligence is also extending to the upstream and downstream of the medical industry, covering processes such as medicine, hospital management, and medical insurance control fees.
In the field of medicine, artificial intelligence can be mainly applied to the following fields, such as: artificial intelligence applied to the structure-activity relationship analysis of compounds, artificial intelligence applied to the prediction of crystal structure of small molecule drugs, and information recruitment of volunteers. In general, the application of artificial intelligence in the field of medicine can improve the efficiency of pharmaceutical companies in the development of new drugs. Take China's local company Jingtai Technology, a company that is a computationally driven drug-based solid-phase R&D company that provides pharmaceutical crystal design services to innovative pharmaceutical companies worldwide. It was established in September 2015 and in 2015. In December, Tencent and Renren Company received tens of millions of RMB A round of financing. It is understood that Jingtai Technology is committed to achieving highly accurate drug solid phase screening and design through computational physics, quantum chemistry and powerful intelligent algorithms in the cloud, which greatly shortens the time for drug design, solid phase screening and drug formulation development. The patent declaration and protection of enterprises play a key role. It mainly provides drug crystal form prediction and crystal patent protection services to help pharmaceutical companies improve R&D efficiency and reduce drug quality risks and patent risks.
Hospital management refers to the management science of hospitals. It uses modern management theories and methods to plan, organize, coordinate and control people, finances, materials, information, time and other resources according to the objective laws of hospital work. Make full use of the hospital's existing resources to maximize medical utility. In the aspect of hospital management, artificial intelligence can also play a certain role. The traditional hospital management methods mostly rely on labor, and the medical staff spends time and effort, which also causes waste of medical resources. Artificial intelligence can replace part of the administrative work of some medical staff through machine learning, such as medical consultation, user survey, data collection and so on. It is also possible to provide hospital administrators with certain decision support through big data analysis. At present, the most widely used artificial intelligence in hospital management is intelligent guidance and triage. In recent years, with the combination of intelligent robot technology and medical care, intelligent guide robots have become a new landscape of hospitals. They perform semantic analysis through the patient's voice input, and then give the hospital's triage and guidance recommendations, saving manpower and facilitating patients. More advanced diagnostic robots can also collect vital signs from patients through sensors to give more accurate advice.
In addition to playing a certain role in medical research and development and hospital management, artificial intelligence also has a layout in terms of medical insurance control fees. Medical insurance monitoring has gradually moved toward an intelligent era. In addition to the experience of regulatory methods and tools, in the form of supervision, developed countries began to use more information technology to supervise the entire process of the use of medical insurance funds.
The medical insurance smart supervision has broad prospects, and domestic enterprises involved in this business field are also favored by the capital market. Chengdu Dianlian Yikang Technology Co., Ltd. was established in 2015. It focuses on the use of big data to provide intelligent audit, policy making decision-making, medical behavior supervision and other services for local social and social bureaus, health planning committees, medical institutions and commercial insurance companies. Medical insurance third party service provider. In November 2016, Digital Alliance Yikang completed a 10 million-level A round of financing, the investor is Tianshili Holding Group Co., Ltd.
According to Zhang Yanlong, CEO of Digital Alliance Yikang, the business of Digital Alliance Yikang is mainly divided into four aspects: the first is for the government, the second is for commercial insurance companies, the third is for medical institutions, and the fourth is for pharmaceutical companies. Among them, the government's business is mainly in the field of medical insurance, through the medical insurance intelligent audit system, big data supervision platform to achieve medical insurance violation control fees, the use of big data model to monitor the hospital's fraud protection behavior, do DRG system reform. For the cooperation with the government, Zhang Yanlong said frankly that the biggest challenge is that the business model is difficult to run through. "Whether it is to help the human society department to do DRG analysis, or the payment method reform data support, it can only be charged through policy bidding and procurement, it is difficult to establish a true business model."
In the field of medical insurance control fees, Digital Alliance will also face many competitors, such as Haihong Holdings, which is mainly engaged in PBM mode, many traditional HIS manufacturers such as Neusoft Medical, Donghua Medical, and Weining Software, as well as Ping An Insurance. insurance giant, as well as health insurance information and communication, medical and many other beans start-up companies. (Text / Zheng Qi)
Interventional Accessories,Introducer Sheath,Introducer Sheath Kit,arterial sheath introducer