"Sometimes it can be cured, it is often relieved, and it is always comforting." This famous quote is considered to succinctly explain the essence of medicine. It is engraved on the pioneer of tuberculosis, Edward Livingston Trou, who is sleeping on the shores of Lake Salanac in the United States. More doctors on the tombstone.
More than one hundred years have passed, and revolutionary changes have taken place every day in medicine. Especially in the empowerment of artificial intelligence, the technicality of medicine has been improved as never before.
Some American media said that China's steps to apply artificial intelligence to the medical field are ahead of the United States. More than 130 companies are using artificial intelligence to improve the efficiency of China's medical system, and hint that US companies should apply AI more to medical care. field.
A few days later, on February 20th, Google brain researchers announced that they can use deep learning to analyze a large number (tens of thousands of levels) of retinal images to predict the risk of cardiovascular disease and to obtain between human anatomy and disease changes. contact. It is said that this is a diagnosis and prediction method that doctors did not know before.
The advancement process of AI physician assistants to take care of human health "guardians" seems to have begun to count the seconds. In the new artificial intelligence technology that is emerging in depth learning, migration learning, etc., AI seems to be able to master the doctor's technology very quickly, and possibly More profound. According to a report by the top accounting firm PricewaterhouseCoopers, "Artificial intelligence may initially be adopted as an auxiliary measure by human doctors. Continuous interoperability will improve the accuracy of the diagnosis and treatment system. In the future, human beings will have sufficient confidence to fully authorize AI. The system performs autonomous operations."
Give the patient's body and life to the robot... Does this imagination make Dr. Trudeau sigh deeply, or is it a surprise?
Primary stage: There is still a distance from the real “AI+Medicalâ€
“The global AI report released by PricewaterhouseCoopers in October last year analyzed the impact of the development of artificial intelligence technology in all major industries, and considered that the most influential is the medical health and bio-pharmaceutical industry.†Gao Di, CEO of Beijing Institute of Infinite Artificial Intelligence Say. In fact, since 2011, the medical field has been at the forefront of AI industry applications, and International Data Corporation listed medical artificial intelligence as the field that attracts the most investment in 2016 in its Global Semi-Cognitive/Artificial Intelligence Expenditure Guide. One, this trend has not slowed down.
The popularity of investment and the prosperity of the business have made AI's application in the medical field almost refreshing people's understanding. From the very beginning of the human form, "Medical Doctor" answers questions for patients, preliminary triage, provides information on the treatment process, department location, etc. In November last year, the "Intelligent Medical Assistant" robot of the University of Science and Technology passed the examination of the clinical practitioners with a score of 456 points, and then applied to the image recognition system such as the lung nodules of the "eyes of the fire". First-line doctors reduce labor intensity and help medical resources reach remote areas.
Despite the great progress, there is still a certain distance from the real "artificial intelligence + medical". At present, many cases are not smooth. Beijing knows that Hans Uskult, chief scientist of the Institute of Infinite Artificial Intelligence and academician of the European Academy of Sciences, admits that artificial intelligence takes data as the lifeline, and it is very difficult to extract even the most basic medical information. Things.
“Extracting information from unstructured data is a very important task. Unstructured data can be either images or text.†Hans said that data from different sources is associated with specific cases for further research. "These data need to be understood within the machine, such as surgical reports, doctors will write in a personalized language, in addition to the truth, the doctor will also write down his suspected information, negative information, etc., how to extract this information, carry out What is the analysis of different categories?†Hans said that many practitioners in the medical field have established a large knowledge base, such as the Joint Medical Language System. This is a very large body of knowledge with 180,000 different concepts. It is a small one. The “knowledge mapâ€, as the infrastructure, needs to be improved together.
Hans said that in the medical field, AI has made great strides from three aspects: first, clinical research, second, pharmacological research, and third, helping doctors measure different factors and making recommendations based on data. "We call it decision support, not the decision itself." Hans stressed that because the final decision is always made by the doctor.
In addition, there is also a breakthrough in another way. Hans said that there is a patient in Berlin who can't move hands and feet. Through the scientific experiment of the brain-computer interface, the machine helps him communicate with people. With the interaction between the machine and the human brain, he You can reconnect with the world.
Key to development: data accumulation is the prerequisite for completing tasks
"Our hospital has a special medical data statistics department, which has accumulated a large number of patients' basic information and their disease diagnosis and treatment data. The data statistics department plans to better support the treatment of patients through the analysis of these data." Peking University Cancer Hospital Ji Xinqiang said.
Represented by Peking University Cancer Hospital, many hospitals have established data centers to accumulate disease information for mining analysis. In Xishan, Beijing, the Chinese Academy of Medical Sciences Fuwai Hospital built a national-level, shareable and extended health medical big data and biological sample library platform based on professional private clouds. Through in-depth mining and use of data samples, the aim is to help doctors accurately treat and Look for new drug targets that indicate functional gene loci.
Data accumulation is a prerequisite for AI to complete its mission. In the Google brain's task of predicting the risk of cardiovascular disease from retinal images supplemented by various factors such as age, gender, smoking history, blood pressure, etc., the system used 130,000 retinal images for training.
Discovering the details that are not perceptible by traditional methods, Chinese scientists are also conducting cutting-edge research. Tian Feng, a researcher at the Institute of Software, Chinese Academy of Sciences, said that his team is developing various medical-grade wearable devices with the support of national key research and development programs. "In the traditional diagnosis of Parkinson's disease, doctors will let patients connect to paper. The spiral is used to judge the condition. The pen with the sensor can detect the user's pen pressure change, pen-like information, etc., which we did not collect before. We found that these are also related to the early signs of Parkinson's disease."
"The acquisition of hand posture is being piloted in the intelligent clinic to help doctors diagnose diseases in the nervous system." Tian Feng said that the project participating unit Concord Hospital is undergoing pilot application, and further research is still being followed.
The ultimate form: AI doctor may not really come
Although Da Vinci surgical robots have used multiple minimally invasive methods to perform multiple complex surgeries, Tian Xiangyang, a leading scholar who has studied public health for many years, believes that patients cannot be handed over to robots. In the book "Handbook of Communication between Doctors and Patients", he wrote: "Medical is benevolence, medicine is benevolent."
"Medical is first of all humanistic, not technical." Tian Xiangyang is told by a story. During the Second World War, there was a Jewish doctor in the Nazi concentration camp. He saw a Jewish compatriot who had just been beaten, shouting and yelling because of pain. However, because there was no rescue device and he was in a hurry, he subconsciously took the other side into his arms. At this time, the miracle appeared. The patient suddenly stopped shouting and screaming, and suddenly he calmed down and his face showed a serene expression. As if he didn't hurt, as if he was seriously injured in the body.
This story is in the beginning of the article, the technical cure is only the "sometimes" of medicine. Tian Xiangyang introduced that among the health-related factors, technical medical services accounted for less than 10%, as well as human biological factors, social and physical environmental factors, and psychological behavior factors. Modern evidence-based medicine provides important guiding ideology for solving human health problems, but inquiring medicine is not perfect. For example, the evidence of effectiveness obtained through AI technology is 99%, which is for 1% of patients. 100% pain and misfortune.
"People are the most sophisticated, complex and sensitive living things in the world. The same diseases are different in different individuals. A disease differs greatly at different stages of the same individual. Medicine is human science, not mechanics. Physics, not biology, cytology, and disease science, is much more complicated than any other discipline. People's problems must be solved by people, and relying solely on technology is not feasible."
Tian Xiangyang expressed concern that excessive reliance on artificial intelligence technology will turn the technology that is soaked in humanistic warmth into a cold-blooded technology, turning a warm hospital into a human body repair shop, which is contrary to the "initial heart" of medicine, and may eventually lead to the alienation of medicine. .
In the medical field, the application of artificial intelligence has broad prospects, such as disease diagnosis, cause inference, treatment plan selection, precision surgery, etc., but only if it is controlled by doctors, any artificial intelligence technology can only be human brain and human hands. Extension is a tool to help doctors solve patient health problems. It can be seen that the last step of AI diagnosis and treatment is the most distant step. (Technology Daily)
China Extract Powder For Use As Dietary Supplement Extract Powder, Extract Powder Manufacturer
Shaanxi Kang New Pharmaceutical co., Ltd. , https://www.anabolicsteriod.com