Ultrasound artificial intelligence in the assessment of liver fibrosis

Liver fibrosis is a necessary step in the progression of various chronic liver diseases to cirrhosis and hepatocellular carcinoma, so early diagnosis of liver fibrosis is of great clinical significance. Ultrasound is the first choice for diagnosing liver fibrosis and has been used as a complementary method to liver biopsy, but it is subjective and lacks uniform quantitative standards in diagnosing liver fibrosis. In recent years, artificial intelligence (AI) has been developed and integrated into ultrasound to make up for the artificial errors of traditional ultrasound, especially imaging histology and deep learning, which can extract more features reflecting disease information from a large number of ultrasound images. Among them, the AI technology based on gray.scale ultrasound and ultrasonic elastography has achieved good staging performance and also faced some challenges. Therefore, this paper reviews the application of ultrasound AI technology in staging of liver fibrosis and the objective challenges faced nowadays.