Tissue Section Image-Based Liver Scar Detection

Liver cirrhosis is a major cause of liver cancer. Traditionally, the diagnosis of stages of liver cirrhosis depends on doctors’ examination of large numbers of images acquired from clinical specimens, which is a relatively time-consuming and labor-intensive task. To avoid this extensive effort and possible error of human judgment, it is necessary to develop an automatic system to recognize the liver scar stages based on clinical liver tissue section images. In this study, a tissue section image-based liver scar stage (TSILSS) diagnosis system is proposed to detect liver scar stages. In this system, a local cross-thresholding method is provided to separate the scar liver tissues from normal liver tissues on a liver tissue section image. Moreover, a two-layer recognition algorithm is presented to identify the scar stage of liver tissue. Furthermore, a parameter decider genetic algorithm is proposed to determine the most suitable values of the parameters used in the TSILSS diagnosis system. The experimental results show that in segmenting scar tissues and normal tissues on liver cirrhosis images, the average precision, average recall rate, and average F-measure that the TSILSS diagnosis system obtains are greater than 94%, and the average accuracy is close to 90%. The TSILSS diagnosis system can help doctors recognize the liver tissue scar stage more efficiently.