CT Large Reconstruction Matrix Combined with Reconstruction Algorithm in the Diagnosis of Pulmonary Nodules
10.3969/j.issn.1005-5185.2024.10.017
- VernacularTitle:CT大重建矩阵联合重建算法对肺结节的诊断价值
- Author:
Xu WANG
1
;
Beibei LI
;
Xiaoyu TONG
;
Anliang CHEN
;
Yujing ZHOU
;
Yijun LIU
Author Information
1. 大连医科大学附属第一医院放射科,辽宁 大连 116011
- Keywords:
Pulmonary nodules;
Tomography,X-ray computed;
Reconstruction matrix;
Iterative reconstruction algorithm;
Image quality;
Signal-to-noise ratio;
Noise
- From:
Chinese Journal of Medical Imaging
2024;32(10):1063-1068
- CountryChina
- Language:Chinese
-
Abstract:
Purpose To explore the value of CT large reconstruction matrix 1024×1024 combined with iterative reconstruction algorithm Karl in the diagnosis of pulmonary nodules.Materials and Methods A total of 500 patients who underwent chest CT examination at the First Affiliated Hospital of Dalian Medical University from October 2021 to May 2022 were prospectively collected,and the raw data of CT scans were reconstructed to divide into group A and B.Group A was reconstructed using a conventional 512×512 matrix combined with Karl 5 reconstruction;group B was reconstructed using a large 1 024×1 024 reconstruction matrix combined with different levels of Karl algorithm to obtain four subgroups,including B1(Karl 6),B2(Karl 7),B3(Karl 8)and B4(Karl 9)subgroup.The signal-to-noise ratio was calculated by measuring the CT and standard deviation values of the tracheal lumen above the arch of the aorta(tracheal area)and the avascular area of the upper lobe of the left lung(lung parenchyma).The overall image quality of the lungs in group A and B was evaluated by two physicians.The best image quality subgroup in group B was compared with the lesion display in group A.The diagnostic efficacy was analyzed based on the surgical pathology results.Results In group B,the standard deviation values of trachea and lung parenchyma gradually decreased and the signal-to-noise ratio gradually increased as the Karl grade increased compared with group A(F=675.002-2 020.903,all P<0.05).All subjective scores in group B were significantly higher than those in group A(Z=-15.361--6.465,all P<0.05),and the highest subjective scores were found in group B4.Some solid nodules(≤3 mm)and solid nodules(6.1 mm-≤3 cm)showed no statistically significant difference in clarity(Z=-2.000,-0.378,both P>0.05).Compared with group A,group B4 showed a 12%-100%improvement in nodule clarity.Only pleural depression sign showed the difference was not statistically significant(χ2=2.143,P>0.05).Taking 43 cases of surgical pathology as the gold standard,the diagnostic accuracy of group B4 was 65.12%,which was better than that of group A,which was 41.86%(χ2=4.674,P<0.05).Conclusion The combined application of the large reconstruction matrix and the Karl iteration algorithm results in superior image quality and facilitates the diagnosis of lung nodules.
