With the rapid development of modern image technology, quantitative analysis of images has been widely applied in the field of clinical diagnosis and pathological analysis. Especially in microscopic, hematoxylin and eosin (H&E) staining histology and Pap smear microscopy, cell nuclei segmentation is an important part of image analysis. Nucleus segmentation helps to observe the stress response of each cell in the sample to drug treatment, so that drug detection is more effective (www.kaggle.com/c/data-science-bowl-2018). However, due to the irregular shape of the nucleus and the unevenness of the nucleus pixels (see figure 1), it is still a challenge to accurately segment nuclei.The traditional watershed-based segmentation method is one of the most widely used methods in image analysis. However, the single watershed method is prone to over-segmentation. Therefore, many scholars have proposed some improved methods to overcome its limitations. A method of watershed and rule merger is proposed for segmenting 3D prostate histopathological images (Adiga et al 2001). This method combines helices of the cell with tiny pieces of the parent cell to reduce the number of overly-divided cells, and shows more than 90% segmentation accuracy. A watershed algorithm based on scale spatial filtering is proposed to segment white cells (Jiang et al 2003). It first extracts nuclei by scale spatial filtering, and then extracts cytoplasm by watershed clustering in hue, saturation and value (HSV) space. Finally, the whole white cell region is obtained by morphological

No abstract

Background and Objectives: The diagnosis of cancer is concerned, and the prediction of cell carcinoma is of great importance for the treatment. Materials and Methods: First, we obtain a series of slices of tumor cell pathology in clinical data, with being followed training sets and test sets gained by adding data model. Then, we design a convolutional neural network training and prediction model. After that, we optimize parameters for training and prediction model, combining experience. Results: In experiment, the accuracy of the model predicting for cell carcinoma is 87.38%. Conclusions: This study provides a reference that predicts the extent of cell carcinoma progression by using deep learning model.