基于深度学习图像重建的低剂量全身 CT：图像质量和病灶检测。
Low-dose whole-body CT using deep learning image reconstruction: image quality and lesion detection.
机构信息
Department of Radiology, Gifu University, 1-1 Yanagido, Gifu 501-1194, Japan.
Department of Radiology, Frontier Science for Imaging, Gifu University, 1-1 Yanagido, Gifu 501-1194, Japan.
出版信息
Br J Radiol. 2021 May 1;94(1121):20201329. doi: 10.1259/bjr.20201329. Epub 2021 Feb 22.
OBJECTIVES
To evaluate image quality and lesion detection capabilities of low-dose (LD) portal venous phase whole-body computed tomography (CT) using deep learning image reconstruction (DLIR).
METHODS
The study cohort of 59 consecutive patients (mean age, 67.2 years) who underwent whole-body LD CT and a prior standard-dose (SD) CT reconstructed with hybrid iterative reconstruction (SD-IR) within one year for surveillance of malignancy were assessed. The LD CT images were reconstructed with hybrid iterative reconstruction of 40% (LD-IR) and DLIR (LD-DLIR). The radiologists independently evaluated image quality (5-point scale) and lesion detection. Attenuation values in Hounsfield units (HU) of the liver, pancreas, spleen, abdominal aorta, and portal vein; the background noise and signal-to-noise ratio (SNR) of the liver, pancreas, and spleen were calculated. Qualitative and quantitative parameters were compared between the SD-IR, LD-IR, and LD-DLIR images. The CT dose-index volumes (CTDI) and dose-length product (DLP) were compared between SD and LD scans.
RESULTS
The image quality and lesion detection rate of the LD-DLIR was comparable to the SD-IR. The image quality was significantly better in SD-IR than in LD-IR ( < 0.017). The attenuation values of all anatomical structures were comparable between the SD-IR and LD-DLIR ( = 0.28-0.96). However, background noise was significantly lower in the LD-DLIR ( < 0.001) and resulted in improved SNRs ( < 0.001) compared to the SD-IR and LD-IR images. The mean CTDI and DLP were significantly lower in the LD (2.9 mGy and 216.2 mGy•cm) than in the SD (13.5 mGy and 1011.6 mGy•cm) ( < 0.0001).
CONCLUSION
LD CT images reconstructed with DLIR enable radiation dose reduction of >75% while maintaining image quality and lesion detection rate and superior SNR in comparison to SD-IR.
ADVANCES IN KNOWLEDGE
Deep learning image reconstruction algorithm enables around 80% reduction in radiation dose while maintaining the image quality and lesion detection compared to standard-dose whole-body CT.
目的
评估使用深度学习图像重建（DLIR）的低剂量（LD）门静脉期全身 CT 的图像质量和病灶检出能力。
方法
该研究纳入了 59 例连续患者（平均年龄，67.2 岁），他们在一年内因恶性肿瘤监测而行全身 LD CT 检查，其中 40%的 LD CT 图像使用混合迭代重建（LD-IR）和 DLIR（LD-DLIR）进行重建，1 例采用标准剂量（SD）CT 进行重建（SD-IR）。放射科医生独立评估图像质量（5 分制）和病灶检出情况。计算肝脏、胰腺、脾脏、腹主动脉和门静脉的 CT 值（HU）、背景噪声和肝脏、胰腺和脾脏的信噪比（SNR）。比较 SD-IR、LD-IR 和 LD-DLIR 图像之间的定性和定量参数。比较 SD 和 LD 扫描之间的 CT 剂量指数容积（CTDI）和剂量长度乘积（DLP）。
结果
LD-DLIR 的图像质量和病灶检出率与 SD-IR 相当。SD-IR 的图像质量明显优于 LD-IR（<0.017）。SD-IR 和 LD-DLIR 之间所有解剖结构的 CT 值均无显著差异（=0.28-0.96）。然而，LD-DLIR 的背景噪声显著降低（<0.001），与 SD-IR 和 LD-IR 图像相比，SNR 显著提高（<0.001）。LD（2.9 mGy 和 216.2 mGy•cm）的 CTDI 和 DLP 明显低于 SD（13.5 mGy 和 1011.6 mGy•cm）（<0.0001）。
结论
与 SD-IR 相比，使用 DLIR 重建的 LD CT 图像可使辐射剂量降低>75%，同时保持图像质量和病灶检出率以及更高的 SNR。
知识进展
深度学习图像重建算法可使全身 CT 标准剂量减少约 80%，同时保持图像质量和病灶检出率。
Zotero 插件