CT image pulmonary nodule detection method and device, apparatus, and readable storage medium

• Main Authors: CHEN HAO, LIU QUANDE, DOU QI
• Format: Patent
• Language: Chinese; English
• Published: 14.12.2018
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a CT image pulmonary nodule detection method and device, an apparatus, and a readable storage medium. The CT image pulmonary nodule detection method comprises the following step: acquiring a CT image to be detected, performing pixel segmentation processing on the CT image through a pre-stored three-dimensional convolution neural pixel segmentation network to obtain a probability map corresponding to the CT image, and marking the probability map with a connectivity domain to obtain a candidate nodule region; predicting the candidate nodule region by a plurality of modelscorresponding to different pre-stored three-dimensional convolution neural network classifiers, obtaining each probability prediction value corresponding to the candidate nodule region, and fusing theprobability prediction values to obtain a classification result of the candidate nodule region. The invention solves the technical problem of poor accuracy of automatic detection of lung nodules based on CT images. 本发明公开了种CT图