Not long ago, Baidu announced a major reorganization of the medical business unit, adjustment and optimization of the organizational structure of the medical business, concentration of superior resources, and the focus of the medical business in the field of artificial intelligence. This move aroused the attention of the industry as soon as Baidu’s actions in the medical field caused various discussions. As everyone knows, there is not much to say here. Then, the point of convergence between artificial intelligence technology and medical business is where? Can artificial intelligence at this stage have a significant impact on the medical industry? Let’s analyze it briefly.
There are many drawbacks in the traditional medical industry and models.
As a special industry, due to the uniqueness and closeness of its own system, it is inevitable that there will be problems of one kind or another. As everyone knows, it is difficult and expensive to see a doctor. Become the object of widespread complaints. The people’s living standards have been gradually improved, and the demand for medical resources has also increased. As a result, the contradiction between health service demand and medical and health resources has become increasingly prominent.
Concentration of medical resources and going to big hospitals for minor illnesses imposes a huge burden on the hospitals. The function of large hospitals should be to treat critically ill and difficult patients, but they treat a large number of common and frequently-occurring patients, which not only makes it difficult and expensive to see a doctor, but also wastes a lot of precious resources; the abnormal development of “medicine to feed the doctor” It also brings unnecessary economic burdens to patients. It is commonplace to prescribe hundreds of thousands of dollars for small problems. The hidden rules of rebates have caused some doctors not to consider the best diagnosis and treatment plan for patients, but the most expensive plan; the distribution of public medical resources between regions is uneven. For example, most of the top three hospitals are distributed in provincial capital cities, and advanced medical equipment and high-quality medical resources are also concentrated in these hospitals, which naturally leads to a large number of The influx of patients into these hospitals makes it difficult to guarantee the results of treatment. From another point of view, traditional medical methods have certain errors in the diagnosis of diseases, and the diagnosis and treatment of certain intractable diseases is even darker.
Where does AI technology have advantages?
It can be seen that medical treatment is a “historical legacy” problem, which is the case in my country, and similar problems exist in many developing and even developed countries. As for artificial intelligence technology, its innate advantages in the field of big data and computing speed may bring amazing progress to the medical industry. In the simplest stage of laboratory analysis, artificial intelligence equipment is now available to perform this operation. Although sample collection such as blood collection, feces collection, puncture, etc. still need to be performed manually, the subsequent steps can be done by artificial intelligence technology. Steps such as sample classification, centrifugation, pushing, staining, and dicing are much more efficient than human operations. Even identification can be determined by analyzing and comparing sample data with big data.
The forecast data of the statistical agency IDC shows that by 2020, the amount of medical data will reach 40 trillion GB, and the speed of data generation and sharing will increase rapidly, of which more than 80% of the data is unstructured data. IDC believes that artificial intelligence technology will be widely used in the medical field in the future, especially in subdivision medical scenarios such as auxiliary diagnosis, drug research, medical imaging, and genetic science. From the current point of view, IBM’s “Watson” should be the world’s leading medical artificial intelligence system. It combines the data integration, analysis and judgment capabilities of artificial intelligence with the diagnosis and treatment experience of human doctors to provide auxiliary medical processing logic.
The era of new medical technology relying more on artificial intelligence and traditional Chinese medicine “seeing, hearing, and asking” is long gone. Today’s medical technology is more rigorous, rigorous, and meticulous, which happens to be what artificial intelligence technology is good at. In terms of difficult and miscellaneous diseases, artificial intelligence technology can integrate global cases into a huge database, and it only takes a few milliseconds to retrieve and retrieve key data; and technologies based on neural networks, computer vision, deep learning, and speech recognition The artificial intelligence system can also provide early warning and diagnosis of diseases such as Alzheimer’s disease and schizophrenia.
Winterlight’s machine learning software analyzes the speeches of patients with Alzheimer’s disease and healthy people, and finds the differences in speech speed, intonation and grammar from the corpus, and finds out the rules. Ordinary people can use this software to test and know how high their risk of Alzheimer’s disease or other cognitive impairments is in the future. This technology can help people predict depression, stroke, aphasia, autism, hyperactivity, etc. Know the obstacles, and then prevent or receive treatment early; Boston Biomedical’s BERG artificial intelligence system compares and analyzes samples collected from cancer patients and healthy people, trying to find the key to “prescribe the right medicine” among the 14 trillion data nodes Nodes, and such a large amount of data nodes cannot rely on human doctors to analyze them at all. It can be seen that tasks that are difficult to be completed by human doctors due to the huge amount of data and rare cases are being discovered and solved little by little by artificial intelligence technology. Although it will take a long time for artificial intelligence to form a complete diagnosis and treatment capability, however, it has affected the working mode of the medical industry, making new drug research and development, pathological diagnosis and other work more efficient; similarly, future new medical technologies will also rely more on artificial intelligence.
Big data system provides tailor-made medical services for the population.
Compared with the high price of hiring family doctors, artificial intelligence technology can be tailored through many details such as people’s working environment, working hours, routines, dietary preferences, patient history, etc. A set of medical services would be suitable for every individual, including fitness, health care and so on. Obtain people’s data through smart bracelets, smart heart rhythm belts, smart underwear and other peripheral devices, and upload them to the cloud server, and then formulate a set of reference medical service rules through the system. Similar things have been piloted in European and American countries. Presumably it is only a matter of time before it is fully rolled out. As for the cutting-edge science of gene sequencing, people in the industry generally believe that gene sequencing will be popularized by all people in the future. Adding genes, exercise, diet, sensors, etc., can perform effective health prediction based on in-depth analysis of big data.
“Artificial intelligence + medical treatment” is not a gimmick, but the future.
Generally speaking, there are seven major opportunities for artificial intelligence in the medical field:
First, provide clinical assistance diagnosis and other medical services for early screening, diagnosis, and rehabilitation, surgery risk and other assessment scenarios;
The second is the informatization of medical institutions, which helps medical institutions to improve their operational efficiency through data analysis;
The third is to identify medical images to help doctors read the patient’s images faster and more accurately;
The fourth is to help medical institutions’ big data visualization and Increase the value of data;
The fifth is to solve the problem of long drug development cycle and high cost in the field of drug research and development;
The sixth is health management service, which monitors users’ personal health data through means including wearable devices, and predicts and controls disease risks;
The seventh, in the field of gene sequencing, deep learning is used to analyze genetic data and promote precision medicine.