Refined pulmonary nodule detection method, system and device and medium

Abstract

The invention discloses a refined pulmonary nodule detection method, system and device and a medium. The refined pulmonary nodule detection method comprises the following steps: obtaining a lung CT image, and carrying out preprocessing and RESIZE processing on the lung CT image; positioning the positions of candidate pulmonary nodules in the lung CT image after RESIZE processing by using a pulmonary nodule screening network to obtain a first candidate pulmonary nodule set P; according to candidate pulmonary nodules in the first candidate pulmonary nodule set P, obtaining an original candidate pulmonary nodule set SP from the preprocessed lung CT image according to the same position; extracting original candidate pulmonary nodules from the original candidate pulmonary nodule set SP by using a pulmonary nodule detection network to obtain a second candidate pulmonary nodule set N; and detecting the original candidate pulmonary nodules in the second candidate pulmonary nodule set N by using the pulmonary nodule false positive reduction network, and outputting a detection result of true positive pulmonary nodules. According to the refined pulmonary nodule detection method, the accuracy of the original lung data is high, and the phase quality and the resolution of the CT image of the lung are high.

Description

technical field [0001] The invention relates to the technical field of computer-aided diagnosis of medical imaging, in particular to a refined pulmonary nodule detection method, system, device and computer-readable storage medium. Background technique [0002] Lung cancer is the cancer with the highest morbidity and mortality in the world. Lung cancer exists in the form of pulmonary nodules in the early stage. Early diagnosis and treatment of pulmonary nodules can improve the survival rate of patients. Computed tomography (CT) is currently the most commonly used image for pulmonary nodule detection. At present, the existing computer-aided detection system for pulmonary nodules mainly consists of two stages: candidate nodule detection and false positive screening. The goal of the candidate nodule detection stage is to detect all suspected nodules in the CT image as much as possible to improve the detection sensitivity of pulmonary nodules; the goal of the false positive scr...

AI Technical Summary

Problems solved by technology

[0003] However, due to the different body sizes of patients undergoing computed tomography scans, the obtained CT images show great differences in the shape, size, and texture of the lungs, and there may be more noise in the CT images The motion artifacts caused by the movement make the data accuracy, phase quality and resolution of CT images low, which affects the screening and detection of pulmonary nodules in the later stage

Images