On July 1st, Air Dad founder and chairman Zhang Dalei made a speech at the Medical Big Data and Medical Artificial Intelligence Summit, hosted by the Wuhan National Bio-industry Base Construction Management Office, Flint Creation, and Optics Valley Health and Smart Park. Presentation on "The Problems and Limitations of AI in the Medical Field".
Airdoc is a leading enterprise in the field of artificial intelligence in the medical field. It focuses on the application of artificial intelligence medical image recognition. It has been trying to solve the uneven distribution of medical resources in imaging through technological means. At the scene, Zhang Dalei focused on sharing practical experience in the field of intelligent imaging and the limitations of artificial intelligence in medical image recognition.
This article is based on the wonderful sharing of Mr. Zhang Dalei in the "Medical Big Data and Medical Artificial Intelligence Summit Forum". According to the guest's comments, some sensitive opinions and internal information have been deleted from the sharing.
"More Intelligent, Better Care"
"In reality, the distribution of medical resources is not average, many people are missed, and artificial intelligence can learn the experience of medical experts. It can be used at the grassroots level to assist grassroots doctors to improve their disease recognition level." Zhang Dalei described the original intention of setting up Airdoc. Sincere and sincere.
In recent years, artificial intelligence image recognition technology has developed rapidly and has surpassed humans in certain specific fields. Medical imaging has become a hot spot as one of the important paths for disease diagnosis. The medical industry involves a wide range of knowledge, and artificial intelligence can play a role in multiple links. For example, medical image recognition, biotechnology, assisted diagnosis, drug research and development, etc., the most widely used is medical image recognition.
“Overall, the whole industry is showing a thriving trend. But in fact, there are still many problems when it really landed.†Zhang Dalei briefly introduced the three conventional practices of artificial intelligence in medical image recognition and its limitations.
Three common practices of artificial intelligence in medical image recognition
Zhang Dalei introduced that there are currently three main methods of artificial intelligence in medical image recognition: classification, detection and segmentation.
Classification is the easiest. It is necessary to input a large number of samples to learn. Generally, the results of the "10,000" level training can be said to be better, but only the disease can be identified, and the medical images cannot be labeled.
Zhang Dalei's on-site example: Assume that the algorithm identifies the highly similar lesions on the image. If the classification sample size is small, it is difficult to adjust. There is a problem of insufficient sample size and unbalanced sample in this part. At the same time, the workload marked by the doctor is the smallest, and the doctor can label the weak label after classification.
Detection, that is, what is detected from the image. The amount of training sample required is generally less than the sample size of the classification, but the workload marked by the doctor has increased. For example, if you want to label the amount of radioactive sample, the doctor needs to look at each film and frame the problematic area. If there is a frame leak in the process, it will affect the detection effect.
China Antibacterial Blend Suppliers
Here you can find the related products in Antibacterial Blend, we are professional manufacturer of Antibacterial blend. We focused on international export product development, production and sales. We have improved quality control processes of Antibacterial Blend to ensure each export qualified product.
If you want to know more about the products in Antibacterial Blend, please click the Product details to view parameters, models, pictures, prices and Other information about Antibacterial Blend.
Whatever you are a group or individual, we will do our best to provide you with accurate and comprehensive message about Antibacterial Blend!
Salicylic Acid,BHA,O-hydroxybenzoic acid,Allantoin
Xi'an Gawen Biotechnology Co., Ltd , https://www.ahualynbio.com