Study Of The Automatic Detection Model Of Pulmonary Nodules Based On CT Images

Lung cancer is one of the fastest growing malignant tumors with incidence rate and mortality rate.Timely detection of lung cancer and treatment are of great importance to human health protection.Most of the traditional pulmonary nodule detection methods are based on the combination of professional medical knowledge and the characteristics of CT images to extract features,and then select useful features to train in the classifier,which is not only cumbersome,but also poor generalization ability of the trained model,and can not play a very good auxiliary role.In recent years,computer vision based on deep learning algorithms has developed rapidly,especially in the detection of pulmonary nodules.In this paper,a Faster RCNN algorithm based on deep learning is proposed to detect pulmonary nodules,which solves the problem that traditional methods can not automatically detect pulmonary nodules.The main contents of this paper are as follows:(1)using LabelImg tool to label lung cancer datasets in the format of LIDC-IDRI and DICOM in Affiliated Hospital of Ningxia Medical University respectively;(2)using Faster RCNN algorithm to detect lung nodules,aiming at the problem that Faster RCNN cannot detect automatically,this paper proposes an improved Faster RCNN algorithm,combining Faster RCNN with 3D spatial features,realizes automatic detection and improves the accuracy of pulmonary nodule detection;(3)based on Faster RCNN algorithm,a pulmonary nodule auxiliary detection system is designed by python.The improved Faster RCNN algorithm is based on Faster RCNN by adding dual path and u-net network structure,so that the improved algorithm proposed in this paper can directly process medical data in DICOM format without conversion to standard JPG image,and can capture rich 3D image features in DICOM format.Experiments show that the detection accuracy of the improved Faster RCNN algorithm reaches 96.6 percent,which is 1.70 percent higher than that of the original Faster RCNN algorithm.The improved Faster RCNN not only simplifies the complicated lung nodule detection steps,but also improves the detection accuracy.This study provides a theoretical basis for the accurate detection of lung cancer and has a good application value for reducing the burden on doctors.