Identification of Cell Pathology by Using Stacked Auto-Encoders Combination with Rotation Forest

Cytopathology is an effective method for obtaining pathological diagnosis in clinic. Because of its safety and convenience, it has been widely used in clinical diagnosis. The classification and recognition of cell pathology images by computer can provide effective diagnostic basis for medical personnel and improve the accuracy and efficiency of diagnosis. Therefore, it is of great significance to use the computational method to identify the pathological image of the cell. In this paper, we propose a new method to identify cell pathology images based on the depth learning algorithm. First, we standardize the processing of cell pathology images, and get images that are suitable for computer algorithm; then extract the advanced features automatically by using automatic encoders; finally use rotation forest classifier to identify the extracted features. In the experiment, we use 5-fold cross-validation method to ensure the stability of the experimental results. Excellent experimental results show that our model can effectively identify cell pathology images, and provide quantitative and objective diagnostic evidence for medical staff.