METHOD AND DEVICE FOR DETECTING PULMONARY NODULE IN CT IMAGE, AND READABLE STORAGE MEDIUM

摘要

A method and device for detecting a pulmonary nodule in a computer tomography (CT) image, and a readable storage medium; the method for detecting a pulmonary nodule in a CT image comprises: obtaining an electronic CT image to be detected, performing pixel division processing on the CT image by means of a pre-stored three-dimensional convolutional neural pixel division network to obtain a probability map corresponding to the CT image, and performing connected-component labeling on the probability map to obtain a candidate nodule region (S10); predicting the candidate nodule region by means of a plurality of models corresponding to different pre-stored three-dimensional convolutional neural network classifiers to obtain respective probability prediction values corresponding to the candidate nodule region, and performing fusion processing on said respective probability prediction values to obtain a classification result of the candidate nodule region (S20). Thus, the present method improves the accuracy of the automated detection of pulmonary nodules in CT images.