Classification and Recognition of Ovarian Cells Based on Two-Dimensional Light Scattering Technology.

Ovarian cancer is a very insidious malignant tumor. In order to detect ovarian cancer cells early, the classification and recognition of ovarian cancer cells is mainly studied by two-dimensional light scattering technology. Firstly, a single-cell two-dimensional light scattering pattern acquisition...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Medical Systems Vol. 43; no. 5
Autores principales: Chen, Qi, Zhang, Jianling
Formato: equations & formulas pictorial tables/charts Journal Article
Publicado: Springer Nature May2019
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:Ovarian cancer is a very insidious malignant tumor. In order to detect ovarian cancer cells early, the classification and recognition of ovarian cancer cells is mainly studied by two-dimensional light scattering technology. Firstly, a single-cell two-dimensional light scattering pattern acquisition platform based on single-mode optical fiber illumination is designed to collect a certain number of two-dimensional light scattering patterns of ovarian cancer cells and normal ovarian cells. Then, the HOG (Histogram of Oriented Gradient) algorithm is used to extract shaving anisotropy feature of two-dimensional light scattering pattern. The results show that the accuracy of classification and identification of ovarian cancer cells by two-dimensional light scattering technology is 90.81%, which suggests that the specificity of cancer cells and normal cells can be characterized by two-dimensional light scattering technology.