| 钟梓杰,沈智威,林黛英.优化SPECIAL 1H-MR波谱并结合自制LCModel软件模拟代谢物光谱数据集定量分析离体猪肝葡萄糖[J].中国介入影像与治疗学,2025,22(4):273-278 |
| 优化SPECIAL 1H-MR波谱并结合自制LCModel软件模拟代谢物光谱数据集定量分析离体猪肝葡萄糖 |
| Optimization of SPECIAL 1H-MR spectroscopy and combination with self-made simulated metabolite spectral data set of LCModel software for quantitative analysis of pig liver glucose in vitro |
| 投稿时间：2025-02-08 修订日期：2025-03-26 |
| DOI：10.13929/j.issn.1672-8475.2025.04.009 |
| 中文关键词: 肝脏 葡萄糖代谢 磁共振波谱学 SPECIAL序列 |
| 英文关键词:liver glucose metabolism magnetic resonance spectroscopy SPECIAL sequence |
| 基金项目: |
|
| 摘要点击次数: 38 |
| 全文下载次数: 14 |
| 中文摘要: |
| 目的 优化SPECIAL1H-MR波谱（MRS）扫描参数，并观察以之结合自制LCModel软件模拟代谢物光谱数据集（B集）定量分析离体猪肝葡萄糖（Glc）的可行性。方法 利用生理盐水将鱼肝油、Glc及胆碱配置成Glc浓度不同（10、20、30、40、50 mmol/L）而鱼肝油（0.125 mg/ml）及胆碱（100 mmol/L）浓度固定的混合溶液，以试管模型模拟肝脏代谢物；每种Glc浓度各设5个、共配置25个试管模型。以不同参数SPECIAL序列扫描试管模型，获得1H-MRS图像。利用LCModel软件分别基于自带代谢物基础集（A集）及B集分析试管模型SPECIAL序列数据，生成SPECIAL 1H-MRS图像并获得其信噪比（SNR）及标准差百分比（SD%）、Glc信号强度；根据SNR及SD%评估SPECIAL1H-MRS图像质量，以选择最优扫描参数，分析以最优参数结合B集所获试管模型Glc信号强度与Glc浓度的相关性。以最优SPECIAL序列扫描离体猪肝（n=5），行更新脂肪抑制（FS）-SPECIAL序列扫描，基于B集获得相应1H-MRS图像并评估其质量。结果 SPECIAL序列最优扫描参数为TR 3 500 ms、TE 4.42 ms、TM 20.00 ms、重复次数（averages）为256，以之结合B集所获试管模型SPECIAL 1H-MRS的SNR达40.5±1.1，SD%为（13.5±1.0）%，谱线更清晰，基线更平稳，Glc峰分辨率更高。以1H-MRS所获试管模型Glc信号强度与Glc浓度呈正相关（r=0.997，P＜0.001）。离体猪肝SPECIAL 1H-MRS的SNR为24.0±2.7，SD%为（13.5±1.1）%，而FS-SPECIAL 1H-MRS质量较高，SNR为29.5±2.3，SD%为（4.0±0.8）%，可见亚甲基峰被抑制而Glc峰分辨率更高。结论 利用参数优化SPECIAL 1H-MRS结合自制LCModel软件模拟数据集可准确定量分析离体猪肝Glc。 |
| 英文摘要: |
| Objective To optimize scanning parameters of SPECIAL ¹H-MR spectroscopy （MRS）， and to observe the feasibility of combining with self-made simulated metabolite spectral data set （B set） of LCModel software for quantitative analysis of pig liver glucose （Glc） in vitro. Methods Metabolite mixture of cod liver oil， Glc and choline with different concentrations of Glc （10， 20， 30， 40， 50 mmol/L） and the fixed concentration of cod liver oil （0.125 mg/ml） and choline （100 mmol/L） were prepared with saline to simulate liver metabolism phantoms. There were 5 tube models for each Glc concentration， with 25 tube models configured. SPECIAL sequence was used to scan tube models with different parameters， and 1H-MRS images were obtained. The acquired SPECIAL data of tube models were analyzed using LCModel software and built-in metabolite basic set （A set） and B set，respectively. SPECIAL 1H-MRS images were generated， and signal-to-noise ratio （SNR）， standard deviation% （SD%） and Glc signal intensity were obtained. The quality of SPECIAL 1H-MRS images were evaluated according to SNR and SD%， and the optimal scanning parameters were selected. The correlation of Glc signal intensity of phantoms obtained by combining optimal parameters with B set and Glc concentration were analyzed. The optimal SPECIAL sequence was used to scan pig liver in vitro （n=5）. Then updated fat suppression （FS）-SPECIAL sequence scanning were performed， the corresponding 1H-MRS images were obtained based on B set， and the quality was observed. Results The optimal scanning parameters of SPECIAL sequence included TR 3 500 ms， TE 4.42 ms， TM 20.00 ms， and the number of repetitions （averages） was 256. SNR of phantoms SPECIAL 1H-MRS acquired with the optimal scanning parameters and B set was 40.5±1.1 and SD% was （13.5±1.0）%， with clearer spectral lines， smoother baselines and higher Glc peak resolution. There was positive correlation between Glc signal intensity obtained with 1H-MRS and Glc concentration of phantoms （r=0.997， P＜0.001）. SNR of SPECIAL 1H-MRS of pig liver in vitro was 24.0±2.7 and SD% was （13.5±1.1）%， while SNR of FS-SPECIAL 1H-MRS was 29.5±2.3 and SD% was （4.0±0.8）%， the methylene peak was suppressed and the resolution of Glc peak was higher. Conclusion SPECIAL 1H-MRS with optimized parameters combining with self-made simulation data set of LCModel software could be used for accurately quantitative analysis of pig liver Glc in vitro. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
