Quantitative Evaluation Method for Liver Fibrosis in Clinical Ultrasound B-Mode Image Based on Optimized Multi-Rayleigh Model

摘要

We have proposed an evaluation method for liver fibrosis using an ultrasound B-mode image. We proposed a multi-Rayleigh (MRA) distribution model to express a probability density function of ultrasound echo envelope obtained from a fibrotic liver tissue. The MRA model enabled us to estimate quantitative liver fibrosis parameters such as amount of fibrotic tissue and fibrosis progressive ratio, and to quantitatively visualize the fibrotic tissue in the B-mode image. In previous studies, we addressed several challenges to quantitatively and correctly evaluate liver fibrosis using the MRA model in clinical condition. In the present study, we integrated past achievements as the optimized MRA model and evaluated the clinical ultrasound B-mode image of fibrotic liver. Using the optimized MRA model, the change of estimated fibrotic tissue characteristics according to the liver fibrosis stage well reflected the tissue structural change caused by liver fibrosis. We concluded that the fibrotic tissue characteristics can be quantitatively and correctly estimated in clinical condition by using the optimized MRA model.