Clinical application of deep learning-based technique for radiation dose reduction in CT-guided percutaneous transthoracic needle biopsy
10.3969/j.issn.1002-1671.2024.07.025
- VernacularTitle:基于深度学习技术在CT引导下经皮肺穿刺活检术中降低辐射剂量的临床应用
- Author:
Long XU
1
;
Yongjun JIA
;
Nan YU
;
Yong YU
;
Dong HAN
;
Guangming MA
;
Li SHEN
;
Haifeng DUAN
Author Information
1. 陕西中医药大学医学技术学院,陕西 咸阳 712046
- Keywords:
lung;
needle biopsy;
noise index;
deep learning algorithm;
radiation dose
- From:
Journal of Practical Radiology
2024;40(7):1146-1150
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the clinical application value of deep learning image reconstruction(DLIR)-based technique for radiation dose reduction with different noise index(NI)in CT-guided percutaneous transthoracic needle biopsy(PTNB).Methods Thirty-two patients undergoing PTNB were selected,and three sets of CT images with NI of 15,30,and 45 sequentially were obtained after adjusting the puncture needle using the scanning parameters of a small range(40 mm),tube voltage 100 kV,and automatic tube cur-rent modulation(ATCM).Group A was 50％weight of adaptive statistical iterative reconstruction-Veo(ASIR-V)scanned with NI 15,and group B and C were the DLIR-high reconstructed images with NI 30 and NI 45 respectively.The CT value and standard devia-tion(SD)value of paraspinal muscles,subcutaneous fat,and arterial vessels were measured at the puncture center point and its upper and lower 10 mm sclices,respectively,and the signal-to-noise ratio(SNR)and the contrast-to-noise ratio(CNR)were also calculated.Then the images were subjectively scored by two physicians,meanwhile the effective dose(ED)among the three groups was com-pared.Results According to the subjective evaluation,the image quality of group A,B,and C all met the clinical requirement for puncture.The SD and SNR of the images in group B were better than those in groups A and C.The SD and SNR of paraspinal mus-cles and sudcutaneous fat were significantly different between groups A and B,and between groups B and C(P＜0.05),however those of paraspinal muscles and subcutaneous fat between groups A and C were not significant.The differences in ED among the three groups were all statistically significant(P＜0.05).Compared with group A,the ED in groups B and C were reduced by 82.86％and 93.90％respectively,and the ED in group C was reduced by 64.44％compared with group B.Conclusion Increasing the NI combined with the DLIR technique can significantly reduce the radi-ation dose during CT-guided PTNB.
