Improving the Evaluation Accuracies of Histopathologic Grade and Ki-67 Immunohistochemistry Expression of Breast Carcinoma Using Computer Image Processing (II)

摘要

In clinical pathology screening, scoring based on Immunohistochemistry staining ER, PR, Her-2 and Ki-67 are important indicators for evaluations of recurrence risk of breast carcinoma in addition to histological pathology of grading such as scoring based on tubule formation, Pleomorphism, and mitotic count. In this research, we extend the previous research results developed using a series of image processing operations and classification techniques so that an automatic evaluation of IHC4 and primary grading of breast carcinoma can be developed successfully. The propose system not only can improve the problems caused by staining variations at different labs or environments, but also evaluates indicators efficiently discussed previously. Therefore, the proposed system can provide physicians important references in diagnosis, treatments, and prognosis analyses.