abstract |
The present application discloses a method for predicting CK19 of hepatocellular carcinoma based on deep learning analysis. The method includes dividing an image to be detected into several sub-images; Detect the feature map corresponding to the image; based on the feature map and the trained classification model, determine the CK19 prediction category corresponding to the image to be detected. In the present application, a segmentation network model is used to determine a feature map as the input item of the classification model, and based on the feature map, the CK19 prediction category corresponding to the image to be detected is determined. Improve the accuracy of CK19 predicted classes. |