Abstract

Objective To evaluate the diagnostic value of computer-aided diagnosis with three-dimensional volumetric measurement in solid pulmonary nodules on follow-up CT. Methods Forty-six patients with 58 solid nodules were evaluated.Each patient underwent chest CT at least twice.The volume, doubling times(DTs) and volumetric growth rates(VGRs) of nodules were calculated with three-dimensional volumetric measurement software. If there were significant statistical differences in DTs between malignant and benign nodules, Nonparametric test was used to make the analysis.If there were significant statistical differences in upper DTs'values between malignant and benign nodules, X2 test was used to make the analysis.Final diagnosis of all nodule were confirmed by pathology or by response to diagnostic therapy or by clinical follow-up.Results The DTs of 13 malignant nodules were 400 days in 22 nodules, with a range from 302 days to 55 158 days (median, 2 704 days); the VGRs ranged from 0 to 24% (median, 6.5%. Twenty-one nodules that decreased in size had the VGRs ranging from -38% to -4% (median, - 16.0%), and the DTs of these nodules could not be calculated by software.There were significant statistical differences in DTs between malignant and benign nodules(P <0.05).Taking 500 days, 400 days and 300 days as the upper value for malignancies there were statistically significant differences in these values between malignant and benign nodules(P < 0.01).Their sensitivities were 100.0%, 92.3%, 76.9% respectively and specificities were 87.5%, 91.7%, 100.0% respectively.Accuracies were 91.9%, 94.6% ,91.9% respectively and positive predictive values were 8,11,0.Conclusions Taking 400 days as the upper DTs'value for malignancies has important diagnostic value. Computer-aided diagnosis with three-dimensional volumetric measurement is sensitive to the volume changes of solid pulmonary nodules on follow-up CT, and is useful in the differential diagnosis of pulmonary nodules. Key words: Coin lesion,pulmonary; Lung neoplasm; Tomography, X-ray computed; Imageprocessing, computer-assisted

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call