Prediction method and system for chemosensitivity of ovarian cancer patient based on pathological image

The invention discloses an ovarian cancer chemosensitivity prediction method and system based on pathological image analysis. The ovarian cancer chemosensitivity prediction method mainly comprises the following steps: 1, acquiring pathological tissue slice images of ovarian cancer; 2, cutting the slice image into color blocks and screening the color blocks; 3, dividing color blocks into a training set and a verification set, and inputting the training set and the verification set into a convolutional neural network for model construction and image feature extraction; 4, predicting individual chemosensitivity scores based on image features; and 5, predicting the optimized composite model by using the data of the test set, and detecting the accuracy. The prediction system comprises an ovarian tissue slice image processing module, an image feature extraction module, a feature selection and chemotherapy sensitivity score calculation module and an intelligent ovarian cancer chemotherapy sensitivity prediction modu