Melanoma identification pictures (identify melanoma pictures German AI algorithm outperforms doctors

1. Artificial Intelligence Image Robot

Xinhua News Agency, Berlin, April 13 (Reporter Tian Ying) German researchers have recently developed an artificial intelligence algorithm that can be used to diagnose melanoma. In a diagnostic contest between 157 dermatology and artificial intelligence algorithms, the accuracy of artificial intelligence is generally better than that of other institutions, such as Germany's National Cancer Center, and researchers published research reports in the new issue of European Cancer Journal, They have developed an artificial intelligence algorithm that can analyze skin lesions for diagnosis.

2. Artificial intelligence image generation

In order to verify the algorithm, they asked 157 dermatology from 12 university hospitals in Germany to compete with artificial intelligence in diagnosis. The content was to analyze 100 skin pictures and determine whether the picture was a normal birthmark or a skin cancer - melanoma. Among the 100 pictures, 20 were melanoma and 80 were benign birthmarks.

3. High definition artificial intelligence image materials

The results showed that only 7 out of 157 doctors had more accurate judgments than artificial intelligence, 14 had comparable accuracy to artificial intelligence, and 136 did not. Researcher Johann Sven Utikal said that artificial intelligence applications will become increasingly important in the diagnosis of skin diseases, and this algorithm can serve as a beneficial supplement to the clinical diagnosis of skin cancer.

4. Artificial intelligence image creativity

However, researchers said that it is not recommended to diagnose skin lesions only through artificial intelligence algorithms. Although artificial intelligence has certain advantages in image analysis, the actual clinical diagnosis is very complex. It needs medical experts to comprehensively judge whether artificial intelligence will completely replace dermatology in 10 years by combining multiple information? The researchers provided a negative answer.

5. Artificial intelligence image materials

The head of this study, Titus Brinker, said that the relationship between the two is similar to the automatic flight system of an aircraft and the automatic flight system that can function when the pilot has good weather and a familiar range; But when encountering difficult landings, experienced pilots are very necessary