On November 17, the 3rd World Internet Conference continued in Wuzhen. Baidu founder, chairman and CEO Li Yanhong gave a speech at the sub-forum to introduce new development trends in the medical industry in the era of artificial intelligence. Li Yanhong believes that Internet + medical can be divided into four levels: O2O, intelligent consultation, genetic analysis and precision medicine, and new drug research and development. Li Yanhong called for the cooperation between the Internet field and the medical and health field to benefit patients. The following is a speech record: Good morning everyone, I said in yesterday’s speech that in the era of artificial intelligence, we need to re-imagine every industry. Today I try to re-imagine the medical industry from the perspective of the Internet. Today's guests come mainly from two fields, one is the Internet field and the other is the medical health field. As I often say in the company, people in two different fields need to repair from the other end to the middle like a bridge. If the goal is the same, the bridge will pass if it is in the middle. If it is Everyone imagines that it is different. If you can't fix it in the middle, then it will fail, so we try to let each side go to the other side. Let's take a look at Baidu's artificial intelligence, what capabilities we already have, and then look at how these capabilities should be applied in the healthcare industry. The manifestation of artificial intelligence in Baidu is mainly through Baidu brain. There are four main functions of Baidu brain. One is the recognition of speech and speech. Today's speech recognition has reached 97% accuracy, which means that it is in a quiet environment. It has surpassed the hearing level of normal people; image recognition is actually very clear in the medical field. Just like the medical image that Director Li Bin once said, a doctor can only watch tens of thousands of films in his life, but for computers. It is said that watching hundreds of thousands of millions of films may be regarded as small data; the user's portrait and natural language understanding can be used in many fields in the field of intelligent consultation. For the medical industry, we have a corresponding Baidu medical brain to provide related solutions. In my opinion, or from the perspective of the Internet, Internet + medical can basically be divided into four levels. We believe that the first level is the O2O service, how to drain users offline, and distribute them to places that are suitable for dealing with user diseases. The second is a smart consultation. Just now, Academician Zhan Qimin also mentioned that, like IBM Watson's diagnosis and treatment of cancer, computers may be able to surpass human doctors in many cases. Baidu also made some attempts in Baidu doctors. The third level we think is genetic analysis and precision medicine. I will talk about it in detail later. The fourth level I think is the development of new drugs. This aspect is not enough in China at present, but I think this is an area where big data and artificial intelligence can really play a decisive role. First of all, let's look at this O2O service. Baidu's Baidu doctor now has 500,000 doctors participating in the consultation. A total of 8 million people have access to relevant medical services through the Baidu doctor platform. The second level is the intelligent consultation. I have just mentioned this example of Watson. Baidu doctor has also done some tests on the smart consultation that we recently launched. For example, I did a test at Peking University International Hospital. In the case of %, the diagnosis of Baidu doctor is consistent with the doctor's diagnosis of Peking University International Hospital, which means that its accuracy has been improved very quickly. And it may perform better in some rare cases, such as the same symptoms, 99.99% is a disease, but there may be one in ten thousand, or even one in 100,000 is another rare disease In this case, as a computer, it can assist the doctor to make some corresponding judgments. These technologies not only require machine learning for a large amount of medical knowledge, but also the ability to understand patient expressions. This is actually the direction of natural language understanding. The third level is genetic analysis and precision medicine. In fact, this is the most exciting direction for doing computer science in these years. Because we speak Moore's Law in the IT field, the only thing that fits the Moore's Law outside the IT field is the cost of gene sequencing, so we feel that there is a possibility of something revolutionary in this area. At present, the biggest problem with using genes to treat diseases is that most of the diseases caused by known genes are caused by single genes, and most of these diseases are rare diseases, and most common diseases are suspected to be caused by multiple genes. of. The combination of multiple genes, so how can you figure out which genes are caused by a combination of genes, in fact, requires a lot of calculations. Professor Zhan Qimin had a cooperative project with him during the Concord. He was to genetically sequence the esophageal cancer patients in China and try to find out which genes work together to cause cancer. Once we figure this out, the future is like genetic editing. These treatments can be used in more common diseases. Some of the current genetic editing applications are rare diseases, such as the American Spark TherapeuTIcs, which can treat a rare disease caused by a virus that causes loss of vision and blindness, but these The disease is too rare. For most diseases, we need to figure out which genes are needed together, and here we need a lot of calculations. When I talked to people in the medical industry, the medical knowledge that we think is very profound is actually very simple in their view, and vice versa. In our opinion, the calculation is very simple. In their view, this is big. Data and artificial intelligence are a bit difficult, so we hope to truly apply the computing power of hundreds of thousands of servers and the most advanced algorithms for deep learning to the medical and health fields. The fourth level is the development of new drugs. The small molecule compounds known to be able to form drugs today are probably as many as 10 and 33. This is probably because all the atoms in the universe do not add up so much. How can such an amount be combined with the disease-causing protein to treat the disease? How do the unknown molecular formulas be screened to find effective new drugs? This also requires extremely powerful computing power and most Advanced algorithms, some of the US startups, such as atomwise, are already doing this. We also think that computer science and artificial intelligence can help in this regard. As I said at the beginning, I hope that people from both sides will work together. We use our big data ability and use our inexhaustible computing power. The medical and health field hopes to collect more and more data and propose at least calculations. The increasingly difficult problem in the field, we work together to benefit hundreds of millions of patients. thank you all! The CAT6A 10Gig UTP modular jacks provide transmission performance beyond 500MHz and Alien Crosstalk requirements exceeding the industry standards set by ANSI/TIA-568-C.2 for supporting the operation of 10Gig Ethernet applications (10GBASE-T). It is designed with a shielded termination cap which minimizes additional AXT from neighboring connectors. It also features twisted pair separation posts which minimize the untwisting of pairs to allow for efficient installation and long-term peak performance. The modular jacks are also backward compatible supporting current CAT6, CAT5e, and voice network applications. CAT6A Keystone Jack,Keystone Jack CAT6A,Cat6a Jack,Keystone 6a NINGBO UONICORE ELECTRONICS CO., LTD , https://www.uonicore.com