Value Of The Lung Analysis Software Of 16-sct In The Solitarypulmonary Nodules

Objective:Solitary pulmonary nodules (SPN) is the hot topic of chest imaging. With the rapid development of the software and hardware of computer and wider use of multi-slice CT, multiple Computer-aided diagnosis or detection (CAD) software have appeared in the detection and diagnosis of SPN. Through the advanced lung analysis(ALA) of GE ,the detection rate of SPN,comparison of three-dimensional volume rendering (VR) and multi-planar reconstruction(MPR) image of SPN , and the change of lesion in the follow-up has been studied, the value of ALA in the detection and diagnosis of SPN has been discussed.Materials and method: 91 patients have been chest scaned by 16-slice CT and detected SPN lesions as the experimental samples, which has been divided into two groups.41 patients of group A have been detected 49 SPN lesions by reference and all of them were 2~20mm;SPN have been independent film-reading by two radiologists (R1, R2) of different experiences and were detected by used and not used ALA, then determined whether there is difference in SPN detection rate between used and not used of ALA. 50 patients of group B have been detected 50 SPN lesions by reference of ALA, and MPR and VR images of SPN were obtained and were compared in SPN shape, border and the relationship between blood vessels and pleura. 50 cases underwent follow-up review at least twice, and the information of volume before and after review,doubling time(DT) and percentage volume change(PVC) were obtained by ALAmulti.Results: In group A, the SPN detection rate of two different radiologists for not use CAD is 79.59% and 65.31%, and the SPN detection of use CAD is 95.92% and 85.71%, there was statistically significant difference between them(P<0.05).If the nodules were more than 5mm there was no significant difference and it was significantly when the nodules were less than 5mm. In group B, VR has superior display of SPN shape and acantha than MPR(P<0.05);MPR images were more clearly to spicules and blood vessels of SPN than VR(P<0.05); all the cases have been undergone follow-up review and post-processed by ALA software, DT of 17 malignant cases were 35 d~418 d ;DT of 12 benign cases with increased volume were 186 d~1212 d. There is statistically significant difference (U=8.00, P<0.05) between the increased volume of malignant and benign SPN. For the SPN with increased volume, the sensitivity to determine benign or malignant was 88.24%, and the specificity was 83.33%, with 400 days as the threshold value.Conclusion: ALA software can help radiologist to improve the SPN detection rate, and provide effective aid for the diagnosis of SPN.