基于生成对抗网络的上腹部 40keV 虚拟单能量图像生成及图像质量评价

Generation of virtual monoenergetic images at 40 keV of the upper abdomen and image quality evaluation based on generative adversarial networks.

机构信息

Department of Radiology, ZhongShan Hospital of Xiamen University, School of Medicine, Xiamen University, Hubinnan Road, Siming District, Xiamen, Fujian, 361004, China.

Clinical Science, Philips Healthcare, Shanghai, China.

出版信息

BMC Med Imaging. 2024 Jun 18;24(1):151. doi: 10.1186/s12880-024-01331-3.

DOI:10.1186/s12880-024-01331-3

PMID:38890572

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11184875/

Abstract

BACKGROUND

Abdominal CT scans are vital for diagnosing abdominal diseases but have limitations in tissue analysis and soft tissue detection. Dual-energy CT (DECT) can improve these issues by offering low keV virtual monoenergetic images (VMI), enhancing lesion detection and tissue characterization. However, its cost limits widespread use.

PURPOSE

To develop a model that converts conventional images (CI) into generative virtual monoenergetic images at 40 keV (Gen-VMI) of the upper abdomen CT scan.

METHODS

Totally 444 patients who underwent upper abdominal spectral contrast-enhanced CT were enrolled and assigned to the training and validation datasets (7:3). Then, 40-keV portal-vein virtual monoenergetic (VMI) and CI, generated from spectral CT scans, served as target and source images. These images were employed to build and train a CI-VMI model. Indexes such as Mean Absolute Error (MAE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity (SSIM) were utilized to determine the best generator mode. An additional 198 cases were divided into three test groups, including Group 1 (58 cases with visible abnormalities), Group 2 (40 cases with hepatocellular carcinoma [HCC]) and Group 3 (100 cases from a publicly available HCC dataset). Both subjective and objective evaluations were performed. Comparisons, correlation analyses and Bland-Altman plot analyses were performed.

RESULTS

The 192nd iteration produced the best generator mode (lower MAE and highest PSNR and SSIM). In the Test groups (1 and 2), both VMI and Gen-VMI significantly improved CT values, as well as SNR and CNR, for all organs compared to CI. Significant positive correlations for objective indexes were found between Gen-VMI and VMI in various organs and lesions. Bland-Altman analysis showed that the differences between both imaging types mostly fell within the 95% confidence interval. Pearson's and Spearman's correlation coefficients for objective scores between Gen-VMI and VMI in Groups 1 and 2 ranged from 0.645 to 0.980. In Group 3, Gen-VMI yielded significantly higher CT values for HCC (220.5HU vs. 109.1HU) and liver (220.0HU vs. 112.8HU) compared to CI (p < 0.01). The CNR for HCC/liver was also significantly higher in Gen-VMI (2.0 vs. 1.2) than in CI (p < 0.01). Additionally, Gen-VMI was subjectively evaluated to have a higher image quality compared to CI.

CONCLUSION

CI-VMI model can generate Gen-VMI from conventional CT scan, closely resembling VMI.

摘要

背景

腹部 CT 扫描对诊断腹部疾病至关重要，但在组织分析和软组织检测方面存在局限性。双能 CT（DECT）可以通过提供低 keV 虚拟单能量图像（VMI）来改善这些问题，从而提高病变检测和组织特征化。然而，其成本限制了其广泛应用。

目的

开发一种将常规图像（CI）转换为上腹部 CT 扫描的 40keV 生成性虚拟单能量图像（Gen-VMI）的模型。

方法

共纳入 444 例行上腹部能谱增强 CT 的患者，并将其分为训练集和验证集（7:3）。然后，40keV 门静脉虚拟单能量（VMI）和能谱 CT 生成的 CI 作为目标和源图像。使用这些图像构建和训练 CI-VMI 模型。采用平均绝对误差（MAE）、峰值信噪比（PSNR）和结构相似性（SSIM）等指标来确定最佳生成器模式。另外 198 例患者分为三组，包括第 1 组（58 例有明显异常）、第 2 组（40 例肝细胞癌 [HCC]）和第 3 组（100 例来自公共 HCC 数据集）。进行主观和客观评估。进行比较、相关性分析和 Bland-Altman 图分析。

结果

第 192 次迭代产生了最佳的生成器模式（更低的 MAE 和更高的 PSNR 和 SSIM）。在测试组（1 和 2）中，与 CI 相比，所有器官的 VMI 和 Gen-VMI 均显著提高了 CT 值以及 SNR 和 CNR。各种器官和病变的客观指标之间均存在显著的正相关。Bland-Altman 分析表明，两种成像类型之间的差异主要落在 95%置信区间内。第 1 组和第 2 组中，Gen-VMI 和 VMI 之间的客观评分的 Pearson 和 Spearman 相关系数在 0.645 至 0.980 之间。在第 3 组中，Gen-VMI 与 CI 相比，HCC（220.5HU 比 109.1HU）和肝脏（220.0HU 比 112.8HU）的 CT 值显著升高（p < 0.01）。HCC/肝脏的 CNR 也在 Gen-VMI（2.0）中显著高于 CI（1.2）（p < 0.01）。此外，与 CI 相比，Gen-VMI 的主观评估图像质量更高。