米资讯| AI assistant doctor "reading film" diagnostic accuracy rate is over 95%
It is the world's first CT and MRI neuroimaging artificial intelligence-assisted diagnostic product; it is equivalent to the senior doctor's level. In the future, perhaps sitting next to the CT machine is no longer a doctor, but an artificial intelligence flashing power. Imagine, ten years later, when you walk into the hospital imaging department, the first thing you will see is a doctor or artificial intelligence? Last month, the world's first "human-machine competition" around the diagnosis of neurological diseases was launched. The National Center for Clinical Research of Neurological Diseases announced that it would prepare for the "heroes" of doctors who are committed to conquering "encephalopathy". How is the competition in a month? Recently, the reporter learned from Beijing Tiantan Hospital that more than 700 doctors from 385 hospitals and various medical institutions nationwide have signed up for the competition, and 6 of them have successfully passed the preliminaries to the finals. At the end of this month, they will compete with AI. . The participating AI players are the world's first CT and MRI neuroimaging artificial intelligence-assisted diagnostic products jointly developed by the National Center for Neurological Disease Clinical Medical Research and the Capital Medical University's Human Brain Protection High-tech Innovation Center - "BioMindTM" Day Medical wisdom. According to reports, the diagnostic accuracy of “BioMindTM†Tianzhizhi has reached more than 95%, which is equivalent to the level of a senior physician. The staff is debugging "BioMindTM" Tianzhizhi. What is the accuracy of machine diagnostics? "The nervous system is a universal link network, which is most suitable for artificial intelligence research, especially for clinical decision support in encephalopathy." Wang Yongjun is the deputy director of the National Center for Clinical Research of Nervous System Diseases and the executive vice president of Beijing Tiantan Hospital. He introduced that "BioMindTM" can achieve a diagnostic accuracy of more than 95% through deep learning of massive disease information, which is equivalent to a high-level chief physician level. This time, the recruitment of neurologists to the world will be carried out to carry out the "human-machine competition", the purpose of which is to verify the accuracy of the diagnosis of Tianzhizhi. Wang Yongjun believes that the application of Tianzhizhi has unparalleled superiority in the stages of prevention, diagnosis, prognosis and rehabilitation of neurological diseases. In the future, it will bring a subversive “technical revolution†in the field of neurological diseases. On the other hand, through the wisdom of big data, the efficient and deep learning of the human brain experience (the technology and experience of the top clinical experts), Tianzhizhi is expected to solve the "dead angle" of the disease that is difficult to solve by the "human brain". Can the diagnosis of primary hospitals be at the same level as that of large hospitals? If “BioMindTM†is reliable, what is the future application prospect? Wang Yongjun said that from a national perspective, the problem of uneven distribution of talent resources in the imaging department is prominent. Taking brain tumors as an example, Beijing Tiantan Hospital has about 10,000 operations per year, and in most primary hospitals, this number may only stay in double digits. Most patients will choose to carry the film to the top three hospitals even if they are inspected at the grassroots level. Some of them repeatedly make films in the process of “upstreamâ€, resulting in waste of resources. If the primary hospital can use AI technology to diagnose the empowerment, the primary doctors can have the same level of diagnostics as the large hospitals, and improve the diagnostic rate, which can reduce the unnecessary diagnosis and economic loss of patients, and reduce the pressure on large hospitals. In addition, in Tiantan Hospital, Tianzhizhi's ability to effectively explore the potential link between information and disease can also assist doctors to make more accurate predictions of diseases, such as predicting whether patients will have major bleeding after hematoma, which can be judged by humans. From 60% to 90%, the paramedical team provides solutions in advance for the risks that patients may experience. Can artificial intelligence be a doctor to replace a doctor? As artificial intelligence applications have gradually penetrated into more industries in the society, many people have been receiving attention because of unemployment. If the accuracy rate of artificial intelligence “reading†has reached more than 95%, does it mean that radiologists are no longer competitive in the future? “I think it is easy to replace the doctor’s person and see the doctor’s work too simple.†Gao Peiyi, director of the Neuroimaging Center of Beijing Tiantan Hospital, believes that AI has great advantages in deep learning of big data, but in actual diagnosis. Among them, radiologists still have strong irreplaceability. "In addition to imaging examinations, a qualified radiologist also needs to look at the test list, physical examination form, ask about family history, personal medical history, and understand the drugs, treatments, and reactions that patients have received. Only after the above conditions can the diagnosis be made." He said that AI may be able to replace the role of a filmmaker, but it is impossible to become a real doctor. Wang Yongjun also believes that artificial intelligence applications can free doctors from boring and repetitive tasks, thus freeing up more time for development work. In addition, AI still relies entirely on the support of data authenticity and quality in the diagnosis of neurological images. In the field of difficult and rare disease diagnosis without big data support, there is still a gap between AI and professional doctors. Tianzhizhi AI also needs "exam" The reporter learned that before the "teacher", Tianzhizhi AI had more than half a year of "learning" time. In the early 1990s, Gao Peiyi studied code programming, database and other technologies when he was studying at the University of Utah Medical School. He was involved in computer-aided diagnosis. Last year, he learned that the hospital wanted to develop an AI for neurological imaging diagnosis and also conducted an "exam" on AI. He gave AI cases of hundreds of common tumors such as schwannomas epidermoid cysts and meningioma. In less than a week, in the subsequent diagnostic tests, the AI ​​correct rate was almost 95%. The relevant person in charge of the AI ​​R&D team introduced the formal “cultivationâ€. The learning materials of AI came directly from the tens of thousands of cases related to nervous system related diseases that Tiantan Hospital received in the past ten years. The huge database became its source of knowledge. In common areas such as meningiomas, pituitary tumors, and gliomas, AI has shown excellent diagnostic levels. Half a year after "study of art", AI has been able to judge some common diseases of the nervous system. In the diagnosis of magnetic resonance imaging of some brain tumors, the accuracy rate has reached more than 90%. In the recent preparations for the "Human-Machine Wars" final, Gao Peiyi also conducted training to improve the AI ​​system's ability to fight. In more than 300 cases, it takes more than 10 hours for the sparring doctors to complete, and the AI ​​takes less than half an hour. "In addition to the speed of learning, its stability also significantly surpasses humans. I don't know how tired it is, and it is not affected by external disturbances. Unlike doctors, they are disturbed by external factors such as mood, state, time and place, thus affecting the accuracy of description. It will always remain calm and level," Gao Peiyi said. Product name
Peeling, sorting, rinsing,disinfecting, dicing, breading starch, fried.
Product name Crisp Fried Food,Fried Crispy Bugles Snacks Food,Fried Crispy Snack Food,Fried Crispy Onion Food Laiwu Manhing Vegetables Fruits Corporation , https://www.manhingfood.com
Fried Garlic,fried garlic crisp,Crispy Fried garlic Flakes, Toasted Flavored Fried Garlic, Crispy fried garlic granules
Product description
Select the high quality Fresh, mature and clean garlic.
fried shallot,fried shallot crisp,Fried Onion crispy,Crispy Fried Onion Flakes,
Crispy fried onion
Product description
Select the high quality Fresh, mature and clean wholesome bulb of the Allium cepa.Peeling, sorting, rinsing,disinfecting, dicing, breading starch, fried.