Head and Neck Imaging
Application of Computer-aided Detection System for Pulmonary Nodules in CT Screening
Author:TENG Ya-qin, JIA Wen-xiao, WANG Yun-ling,et al
affiliation:Department of Radiology, The Second Affiliated Hospitial, Xingjiang Medical University, Xinjiang Urumqi 830063, P.R.China
PDFAbstract
Objective By comparing three reading methods of CT screening for lung nodules: CAD, double reading and the radiologist combined with CAD. To evaluation the application of computer aided testing system (CAD) in CT screening for lung nodules. Methods 60subjects diagnosed with lung nodules undergoing the rountine-dose screening CT examination were random selected. Three methods were used, Method A: using CAD system to identify the image automatically. Methods B: two radiologists reading all images , record the results and with 2 agreement as the final diagnosis; Method C: by another radiologist as the same work experience with CAD system, recorded the detection as the final diagnosis. The reference standard was determined by three raiologists. χ2 test was performed to test whether there is significant difference between sensitivities of three reading modes, and statistical significances were defined as P<0.05. Results CAD system, double reading, the radiologist with CAD method of pulmonary nodules detection sensitivity was 75.1%,73.8% and 87.3% respectively. The sensitivities of pulmonary nodule detection of the radiologist combined with CAD was significantly higher than CAD and double reading methods (P=0.001,P=0.0003).There were no significant sensitivity diffrernce between CAD and double reading method (P=0.74). Conclusion The CAD system can improve the ability of detection pulmonary nodules of the radiologist, and radiologists in combination with CAD diagnosis method is better than double reading method with same work experience.
【Keyword】X-ray Computed; Computer-aided Detection; Pulmonary Nodule
【Chart number】R322.3+5
【Document Identification Number】A
【DOI】 10.3969/j.issn.1672- 5131.2016.05.011
Chinese journal of CT and MRI
th14Volume, th 5 Issue
2016Year05Month
Related articles