关键词: 胃肿瘤, 体层摄影术, X线计算机, 双能量CT, 图像处理, 计算机辅助

Abstract:

Objective To explore the value of the Monoenergetic+ algorithm in dual-energy CT （DECT） for optimizing the display of gastric cancer lesions. Methods A total of 48 gastric cancer patients who underwent abdominal DECT contrast-enhanced examinations at Nantong First People's Hospital， Jiangsu Province from October 2023 to December 2024 were enrolled as study subjects. Using the Syngovia post-processing workstation and the Monoenergetic+ algorithm， a total of 12 groups of arterial phase and venous phase images （40-90 keV） were obtained. The 100 kVp and 140 kVp images of the arterial phase and venous phase were fused with a coefficient of 0.5 to obtain 2 groups of 120 kVp fused images. The cross-sectional images with a layer thickness of 2 mm and a layer spacing of 2 mm were reconstructed for subjective and objective assessment， and for comparing the subjective and objective score differences of each energy level. Results In terms of subjective scoring， for the arterial phase， the image scores at 120 kVp and 40， 50， 60， 70， 80， and 90 keV were 4 （4， 4）， 5 （5， 5）， 4 （4， 5）， 3 （3， 4）， 3 （3， 3）， 2 （2， 3）， and 2 （2， 2） points， respectively. For the venous phase， the image scores of the above 7 groups were 4 （4， 5）， 5 （5， 5）， 4 （4， 5）， 3 （3， 4）， 3 （3， 3）， 2 （2， 3）， and 2 （2， 2） points， respectively， with statistically significant differences （H=259.67， P<0.001； H=258.37， P<0.001）. Further pairwise comparison results showed that， in both arterial and venous phases， the scores of 40 keV images were significantly higher than those of the concurrent 120 kVp fused images and other monoenergetic images （all P<0.05）. Regarding objective scoring， the signal-to-noise ratios （SNR） of the above 7 groups of images in the arterial phase were 8.23±3.16， 5.00±1.91， 4.74±1.65， 4.65±1.65， 4.58±1.62， 4.37±1.74， and 4.38±1.53， respectively， and the contrast-to-noise ratios （CNR） were 2.81 （1.45， 3.91）， 3.46 （2.81， 4.93）， 2.86 （2.01， 3.83）， 2.28 （1.59， 3.36）， 1.80 （1.14， 2.74）， 1.32 （0.50， 2.30）， and 0.84 （0.04， 1.48）， respectively， with statistically significant differences （F=70.99， P<0.001； H=234.76， P<0.001）. Pairwise comparison results showed that the SNR of the 120 kVp fused images in the arterial phase was significantly higher than that of each monoenergetic image （all P<0.05）； whereas the CNR of the 40 keV images in the arterial phase was significantly higher than that of other monoenergetic images and the 120 kVp fused images （all P<0.05）. The SNR of the above 7 groups of images in the venous phase were 9.27±2.81， 6.17±2.20， 5.81±2.13， 5.57±1.91， 5.47±1.74， 5.40±1.76， and 5.38±1.59， respectively， and the CNR were 3.48 （2.45， 4.61）， 4.23 （3.44， 5.59）， 3.53 （2.89， 4.59）， 2.97 （2.33， 3.92）， 2.38 （1.59， 3.39）， 1.87 （1.18， 2.65）， and 1.30 （0.70， 2.23）， respectively， with statistically significant differences （F=82.34， P<0.001； H=256.45， P<0.001）. Pairwise comparison results showed that the SNR of the 120 kVp fused images in the venous phase was significantly higher than that of each monoenergetic image （all P<0.05）， and the SNR of the 40 keV images in the venous phase was significantly higher than that of other monoenergetic images （all P<0.05）； whereas the CNR of the 40 keV monoenergetic image in the venous phase was significantly higher than that of other monoenergetic images and the 120 kVp fused images （all P<0.05）. Conclusions The 40 keV monoenergetic image obtained by the DECT Monoenergetic+ algorithm can significantly optimize the display of gastric cancer lesions and is helpful for the identification and observation of gastric cancer lesions.

Key words: Stomach neoplasms, Tomography, X-ray computed, Dual-energy CT, Image processing, computer-assisted