According to a report by Agence France-Presse on May 29, researchers said on the same day that a study comparing humans with machines to find better and faster diagnostic methods has found that computer diagnostics for skin cancer are accurate. The rate is higher than human dermatologists.
A team from Germany, the United States, and France trained an artificial intelligence system to view more than 100,000 images of skin lesions, allowing them to distinguish between images of malignant skin lesions and benign skin lesions.
This machine (a deep neural network) conducts test competitions with 58 dermatologists from 17 countries to identify images of malignant melanoma and benign moles. Of these dermatologists, only more than half have had medical experience for more than 5 years and are experts, 19% have 2 to 5 years experience, and 29% are still beginners. time.
The research team wrote in a paper published in the Annals of Oncology that most dermatologists do not have the ability to recognize skin cancer as they do in artificial intelligence systems. On average, dermatologists accurately detected 86.6% of skin cancers from the images provided, and the accuracy of the artificial intelligence system was 95%.
The study's first author, Holger at the University of Heidelberg, said in a statement that the number of melanomas that were missing from the AI ​​system diagnosis was very low, which means that it is more sensitive and accurate than dermatologists in identifying skin cancer. In addition, the probability of misdiagnosis is also lower, which reduces some unnecessary surgery.
The team said that artificial intelligence may be a useful tool for faster and easier diagnosis of skin cancer and can help doctors perform surgery on patients before the surgery begins. They added that there are approximately 232,000 new cases of melanoma in the world each year, causing 55,500 deaths.
However, machines are not likely to completely replace human doctors; they can only serve as an aid. Images of melanomas in certain parts of the body, such as fingers, toes, and scalp, are difficult to extract from the machine, and artificial intelligence may also be difficult to identify "atypical" lesions or lesions that the patient himself does not know. (Li Meimei Tian Ruizhe)
living room lounge chairs,Living Room Chairs,living room half recliner chairs
Guangzhou LoPhiDa Co.Ltd , https://www.guangdongwidinlsa.com