Google has announced that its own artificial intelligence system is able to detect 26 skin diseases accurately, such as dermatologists, which are one of the most common types of diseases in the world after colds, fatigue and headaches. It is estimated that 25% of all treatments provided to patients worldwide are dedicated to skin diseases and that 37% of patients seen at the clinic have at least one skin condition.
The huge caseload and the global shortage of dermatologists have forced patients to look for general practitioners, who tend to be less specific than specialists in determining conditions.
This trend has prompted Google researchers to design an artificial intelligence system that can detect the most common skin diseases in primary care.
In a new research paper, Google explained that the system is able to detect 26 skin conditions accurately when images and metadata about the patient’s condition are presented to it, and claims to be on a par with licensed dermatologists from the United States.
Google Software Engineer Yuan Liu and Google Health Technical Director Peggy Bui wrote: We have developed a Deep Learning System (DLS) to treat the most common skin diseases in primary health care.
Google explained that dermatologists do not have a single diagnosis of skin disease, but they list possible diagnoses and systematically identify them in laboratory tests, imaging, procedures, and subsequent consultations.
The Google system follows the same approach, since it processes entries containing one or more clinical images of skin abnormalities and up to 45 types of metadata, such as age, sex, and symptoms.
Source : Google