USCT声速反演成像定量评估肌肉脂肪含量的实验研究

doi:10.3877/cma.j.issn.1672-6448.2025.06.010

目的

利用基于全波形反演（FWI）的超声计算机断层扫描（USCT）新型成像技术定量评估肌肉脂肪含量。

方法

选取39例离体新鲜猪后腿模型，利用环形阵列USCT系统（512阵元，0.9 MHz）进行扫描，基于频域多尺度FWI算法重建高分辨率声速图像。在得到声速图像后，根据肌肉组织与脂肪组织之间的声速差异，对图像进行分析，计算脂肪面积百分比。以病理学分析（HE染色）测量的脂肪面积百分比为金标准进行对比验证。通过Pearson相关分析及组内相关系数（ICC）、Bland-Altman图分析评估USCT与病理结果的相关性与一致性。

结果

USCT测量的肌肉脂肪含量为（17.79±8.97）%，病理测量结果为（19.90±9.42）%。相关性分析显示，USCT与病理肌肉内脂肪含量呈显著正相关（r =0.874，P<0.001）。一致性分析显示，USCT与病理结果的ICC为0.853（95%CI：0.704～0.925，P<0.05），Bland-Altman图显示，一致性界限（LOA）为-11.22%～6.99%。USCT与病理测量的肌肉内脂肪含量存在系统性偏差（平均差值=-2.11%）。去骨后USCT成像可清晰区分皮肤、脂肪与肌肉组织，显著提升定量精度。

结论

本研究基于声速反演的USCT技术能够有效量化肌肉脂肪含量，为肌肉退行性病变的早期诊断提供非侵入性评估新途径，并为肌骨代谢性疾病的精准影像评估奠定了重要技术基础。

关键词: 超声计算机断层扫描, 肌肉骨骼系统, 肌肉脂肪含量, 全波形反演

Objective

To utilize a novel imaging technique employing ultrasound computed tomography (USCT) integrated with full waveform inversion (FWI) for the quantitative assessment of muscle fat content.

Methods

Thirty-nine ex-vivo fresh pig hindlimb models were scanned using a ring-array USCT system (512-element, 0.9 MHz). High-resolution acoustic velocity images were reconstructed using a frequency-domain multiscale FWI algorithm. Following sound speed image acquisition, fat area percentage was calculated based on the acoustic speed difference between muscle tissue and adipose tissue. Pathological analysis (hematoxylin-eosin [HE] staining) served as the gold standard for measuring fat area percentage for comparative validation. Correlation and agreement between USCT and pathological results were assessed using Pearson correlation analysis, intraclass correlation coefficient (ICC), and Bland-Altman plot analysis.

Results

The muscle fat content measured by USCT was (17.79±8.97)%, while the pathological measurement result was (19.90±9.42)%. Correlation analysis revealed a significant positive correlation between USCT and pathological intramuscular fat content (r = 0.874, P<0.001). Agreement analysis showed an ICC of 0.853 (95% confidence interval: 0.704-0.925, P<0.05) between USCT and pathological results. The Bland-Altman plot indicated that the limits of agreement (LOA) ranging from -11.22% to 6.99%. There was a systematic bias between USCT and pathological measurements of intramuscular fat content (mean difference=-2.11%). Post-deboning USCT imaging clearly differentiated skin, fat, and muscle tissues, significantly enhancing quantitative accuracy.

Conclusion

This study demonstrates that USCT technology based on sound speed inversion can effectively quantify muscle fat content. It provides a novel non-invasive approach for the early diagnosis of muscle degenerative diseases and establishes a crucial technical foundation for the precise imaging assessment of musculoskeletal metabolic disorders.

Key words: Ultrasound computed tomography, Musculoskeletal system, Muscle fat content, Full waveform inversion

图1 超声计算机断层扫描（USCT）系统及扫描示意图。图a为环形阵列USCT扫描系统；图b为USCT扫描示意图

图2 基于声速阈值的肌肉脂肪含量计算流程图注：ROI为感兴趣区

图3 肌肉脂肪含量的病理测量示意图。图a为组织病理图片（HE ×100）；图b为测量脂肪像素面积（红色为脂肪面积），计算脂肪面积百分比（%）=脂肪像素面积/组织像素面积×100%

图4 超声计算机断层扫描测量的脂肪含量与病理脂肪含量的相关性分析散点图注：USCT为超声计算机断层扫描

图5 超声计算机断层扫描测量的脂肪含量与病理脂肪含量的Bland-Altman分析图注：USCT为超声计算机断层扫描

图6 离体猪后腿及全波形反演（FWI）成像结果（带骨）。图a，c分别为受检离体猪后腿进行超声计算机断层扫描（USCT）及受检离体猪后腿大体观；图b，d分别为不同频率（0.5 MHz、1.25 MHz）FWI成像组织声速图

图7 离体猪后腿及全波形反演（FWI）成像结果（无骨）。图a为受检离体猪后腿进行超声计算机断层扫描（USCT）；图b为FWI成像组织声速图

1	尤聪瑶, 杨铁毅. 肌少症诊治研究进展 [J]. 国际骨科学杂志, 2022, 43(06): 357-361.
2	Szabo TL. Diagnostic ultrasound imaging: inside out [M]. New York: Academic press, 2013.
3	Pandey PU, Quader N, Guy P, et al. Ultrasound bone segmentation: a scoping review of techniques and validation practices [J]. Ultrasound Med Biol, 2020, 46(4): 921-935.
4	Laugier P, Haïat G. Bone quantitative ultrasound [M]. Springer, 2011.
5	Foiret J, Minonzio JG, Chappard C, et al. Combined estimation of thickness and velocities using ultrasound guided waves: a pioneering study on in vitro cortical bone samples [J]. IEEE Trans Ultrason Ferroelectr Freq Control, 2014, 61(9): 1478-1488.
6	Bochud N, Vallet Q, Bala Y, et al. Genetic algorithms-based inversion of multimode guided waves for cortical bone characterization [J]. Phys Med Biol, 2016, 61(19): 6953-6974.
7	张元鸣飞, 吴同绚, 周谋望, 等. 定量超声技术在肌力评定中的应用 [J]. 中国康复医学杂志, 2018, 33(10): 1242-1245.
8	谢帅, 赵佳琦. 神经肌肉疾病骨骼肌结构与功能的定量超声研究进展 [J]. 海军军医大学学报, 2023, 44(2): 155-160.
9	Ali R, Mitcham TM, Brevett T, et al. 2-D slicewise waveform inversion of sound speed and acoustic attenuation for ring array ultrasound tomography based on a block LU solver [J]. IEEE Trans Med Imaging, 2024, 43(8): 2988-3000.
10	Yamashita MW, Larsen LH, Perez J, et al. Comparison of mammography and mammography with supplemental whole-breast US tomography for cancer detection in patients with dense breasts [J]. Radiology, 2024, 311(3): e231680.
11	Zhou C, Xu K, Ta D. Frequency-domain full-waveform inversion-based musculoskeletal ultrasound computed tomography [J]. J Acoust Soc Am, 2023, 154(1): 279-294.
12	Fincke J, Zhang X, Shin B, et al. Quantitative sound speed imaging of cortical bone and soft tissue: results from observational data sets [J]. IEEE Trans Med Imaging, 2022, 41(3): 502-514.
13	Wiskin J, Malik B, Borup D, et al. Full wave 3D inverse scattering transmission ultrasound tomography in the presence of high contrast [J]. Sci Rep, 2020, 10(1): 20166.
14	武小晴, 李玉冰, 苏畅, 等. 基于环形阵列的肌骨超声高分辨率断层成像 [J]. 仪器仪表学报, 2024, 45(6): 208-218.
15	魏聪, 陈捷, 李洪涛, 等. 超声断层成像用于关节置换术后影像学评估 [J]. 中国医学影像技术, 2023, 39(3): 412-416.
16	Plessix RE. A review of the adjoint-state method for computing the gradient of a functional with geophysical applications [J]. Geophysical Journal International, 2006, 167(2): 495-503.
17	郑浩, 季久秀, 周李生, 等. 猪肉肉色评分与色度值、大理石花纹评分及肌内脂肪含量回归模型的建立 [J]. 江西农业大学学报, 2019, 41(1): 124-131.
18	Liu W, Liu T, Zhang J, et al. Two-stage multimodal method for predicting intramuscular fat in pigs [J]. Agriculture, 2024, 14(10): 1-14.
19	Hormati A, Jovanović I, Roy O, et al. Robust ultrasound travel-time tomography using the bent ray model[C]// Proceedings of the medical imaging 2010: ultrasonic imaging, tomography, and therapy. San Diego: SPIE, 2010.
20	Li C, Duric N, Littrup P, et al. In vivo breast sound-speed imaging with ultrasound tomography [J]. Ultrasound Med Biol, 2009, 35(10): 1615-1628.
21	Huthwaite P, Simonetti F. High-resolution imaging without iteration: A fast and robust method for breast ultrasound tomography [J]. J Acoust Soc Am, 2011, 130(3): 1721-1734.
22	Wu X, Li Y, Su C, et al. Ultrasound computed tomography based on full waveform inversion with source directivity calibration [J]. Ultrasonics, 2023, 132: 107004.
23	Bernard S, Monteiller V, Komatitsch D, et al. Ultrasonic computed tomography based on full-waveform inversion for bone quantitative imaging [J]. Phys Med Biol, 2017, 62(17): 7011-7035.
24	Li Y, Shi Q, Li Y, et al. High-resolution bone microstructure imaging based on ultrasonic frequency-domain full-waveform inversion [J]. Chinese Physics B, 2021, 30(1): 014302.
25	Zhuang Z, Lin H. Fast full waveform inversion using reflection image and anatomical prior for dual-linear-array ultrasound computed tomography[C]// Proceedings of the 2024 IEEE ultrasonics, ferroelectrics, and frequency control joint symposium (UFFC-JS). Taipei: UFFC-JS, 2024.

[1]	熊雯, 罗红, 安绍宇, 吴莹, 刘芸. 胎儿骨骼系统异常与染色体异常的相关性分析[J/OL]. 中华医学超声杂志（电子版）, 2015, 12(02): 148-151.

阅读次数

全文

摘要

选择文件类型/文献管理软件名称

选择包含的内容

Quantitative assessment of muscle fat content using ultrasound computed tomography-based speed-of-sound inversion imaging: an experimental study

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

Quantitative assessment of muscle fat content using ultrasound computed tomography-based speed-of-sound inversion imaging: an experimental study