From August 30th to September 1st, the "2018 China (Shanghai) International Artificial Intelligence Exhibition and OFweek (2nd) International Artificial Intelligence Industry Conference" hosted by China High-tech Industry Portal OFweek and the High-Tech Association Shanghai was successfully held.
On the second day of the sub-forum-AI+ medical special session, Director of the Department of Radiology, Shuguang Hospital, Shanghai University of Traditional Chinese Medicine, Director of the Department of Imaging and Research, Director Zhan Songhua, discussed the current difficulties and challenges of medical imaging artificial intelligence. The application of technology to the difficult point of medical treatment, and proposed the development direction and recommendations for AI people.
The characteristics of imaging medicine and the status quo in China
Before the speech, Director Zhan Songhua said that artificial intelligence wanted to replace radiologists, and it is still far away.
Then, Director Zhan Songhua said that the radiology department consists of two parts, one is a physician in imaging medicine and nuclear medicine, and the other is a technician in the Department of Imaging Technology. The two are very different. Image diagnosis requires clinical knowledge and experience, and comprehensive analysis can make a correct diagnosis. This requires the radiologist to be a deeper in clinical medicine and a physician in the physician. But today, the nature of radiodiagnostic physicians is often misinterpreted and the value is underestimated. In fact, the work of radiologists goes far beyond the discovery of lesion location, qualitative lesions, diagnosis, treatment planning, treatment, and MDT as part of their responsibilities.
Director Zhan Songhua emphasized that the pulmonary nodules are only symptoms, not diagnoses. The current AI analysis of the image may be in place, but the radiological diagnosis alone depends on the qualitative judgment of the image, and the error will be very large. Therefore, when the AI ​​judges the radiological image, there will be many misreading situations.
The problem of the application of artificial intelligence in the field of imaging medicine
On the question of what artificial intelligence can do, the answer given by Director Zhan Songhua is: Detection (measurement of lesions), and Characterization (characterization). Director Zhan Songhua believes that assisting doctors to find lesions and distinguish between normal and abnormal may be the first task of AI. Although artificial intelligence is promising in detecting lesions, it is difficult to replace the doctor for diagnosis and treatment.
Here, Director Zhan Songhua gave an example of a calculator and a person. The appearance of a calculator replaced the abacus, but it was not a substitute for accounting. He conducted a vivid demonstration: finding a pulmonary nodule is a simple matter for a computer, but Radiology has more complicated and more difficult conditions, and the boundaries between normal and abnormal human beings are vague. Image diagnosis is an art of "possibility". It is impossible to solve problems by established rules and goals alone.
Pragmatic advice to promote the development of artificial intelligence
The development of artificial intelligence in the field of imaging medicine is difficult. In order to promote the better application of artificial intelligence to medical treatment, Director Zhan Songhua put forward five suggestions.
First of all, AI people need to listen to the clinical voice and understand the actual needs of doctors, instead of blindly pursuing fancy functions. Secondly, solving the false negative rate is the key to improve the certainty of AI machine judgment, thus saving doctors time and effort; It is more professional and important. It can't be stuck on a problem of nodules. There are still many other problems in medicine that need to be studied specially. To play the role of AI and help doctors effectively, make doctors' diagnosis more efficient and comprehensive. Finally, face difficulties and cannot Quick success and exaggeration.
Artificial intelligence medical development direction
Director Zhan Songhua affirmed the advantages and capabilities of artificial intelligence in detecting lesions, but he believes: "The direction of artificial intelligence medical development should not be to replace doctors but to help doctors." Because there is still a long way from analyzing the nature of the disease to the diagnosis. The distance is gone. It takes time to change the public's concept of seeing a doctor. The problem of science cannot transcend facts. Especially in medicine, it cannot be arrogant, and pragmatism and hard work are the foundation. Director Zhan Songhua believes that interaction, listening, comfort, and care are all human problems. In this respect, doctors are indispensable, and they are clinical, medical, and challenging.
Later, Director Zhan Songhua expressed his opinion on whether artificial intelligence will solve the problem of expensive medical treatment. He believes that this is a pseudo-proposition. Compared with foreign countries, it is much simpler and cheaper to see a doctor in China. The price is high and the price is high. Expensive is two completely different concepts. Don't expect AI to solve the problem of expensive medical treatment.
Finally, Director Zhan Songhua said that the development of AI+ medical treatment is the right direction, but it still needs to increase investment in scientific research. It is necessary to integrate physicians and engineers well in order to achieve success. In this era, the profession of dealing with people will only prosper, so doctors cannot be replaced by artificial intelligence.
Small Molecule Peptides Powder
Small molecule peptides powder:Bovine bone oligopeptide powder,Fish collagen peptide powder,Sunflower disk oligopeptide powder,Oat protein peptide powder,Corn oligopeptide powder,Soybean peptide powder,Whey protein peptide.
Small Molecule Peptides Powder,Collagen Peptide Powder,Small Molecule Oligopeptide Powder,Marine Collagen Peptides Protein Powder
Shaanxi Zhongyi Kangjian Biotechnology Co.,Ltd , https://www.zhongyibiotech.com