單體素頻譜(Single Voxel Spectroscopy, SVS)為磁振頻譜(Magnetic Resonance Spectroscopy, MRS)常見之量測方式，可針對腦內預先設定之感興趣 體積(Volume of Interest, VOI)進行訊號擷取與代謝物定量分析。由於 SVS 僅 對單一 VOI 進行量測，其結果易受到 VOI 位置、大小和所涵蓋組織範圍影響。 為維持量測條件的一致性，一般研究中常採用固定感興趣體積設定策略(Fixed VOI Size Prescription Strategies, FIX)，然而 FIX 並未考量到不同受試者的腦區大 小、形狀之差異。自適應感興趣體積設定策略(Adaptive VOI Size Prescription Strategies, ADP)，可根據受試者的腦部解剖特徵調整 VOI 的大小和範圍，使其 更貼近目標腦區，然而也可能因 VOI 大小改變而影響訊號強度、頻譜品質與代 謝物定量的穩定性。
本研究比較 ADP 與 FIX 於前扣帶皮質(Anterior Cingulate Cortex, ACC)與 背外側前額葉(Dorsolateral Prefrontal Cortex, DLPFC)之 SVS 量測表現，以評 估 ADP 對頻譜品質、代謝物定量與重複量測後重現性之影響。我們使用 Osprey 對 SVS 資料進行前處理、頻譜擬合與代謝物定量分析，並以訊噪比(Signal-to- Noise Ratio, SNR)、半高全寬(Full width at half maximum, FWHM)、相對殘差 振幅(Relative Residual Amplitude, relResA)作為頻譜品質分析指標;代謝物定 量則以組織校正水參考定量(tissue-corrected water-scaled, TCW)和以總肌酸作 為內部參考之代謝物比值(ratio to tCr)作為定量指標，並以相對差異百分比描 述兩者之間的變化;最後以變異係數(Coefficient of Variation, CV)評估重複量 測下之重現性。
研究結果顯示，ADP 相較於 FIX 整體呈現較低的 SNR，顯示其訊號強度可 能受到 VOI 大小影響，然而 ADP 之 relResA 整體表現優於 FIX，說明較低的 SNR 並未同步反映擬合後的殘差增加。在代謝物定量方面，ADP 對多數代謝物 並未造成整體性的定量偏移。最後在重現性方面，多數代謝物於 ADP 與 FIX 皆 有穩定的重複量測表現，兩者之 CV 變化幅度約在 3 個百分點之內。

Single Voxel Spectroscopy (SVS) is a common measurement method in Magnetic Resonance Spectroscopy (MRS). It can perform signal extraction and metabolite quantification analysis on a pre-defined Volume of Interest (VOI) within the brain. Since SVS only measures a single VOI, its results are easily affected by the location, size, and covered tissue range of the VOI. To maintain the consistency of measurement conditions, general studies often adopt Fixed VOI Size Prescription Strategies (FIX). However, FIX does not account for differences in brain region size and shape among different subjects. Adaptive VOI Size Prescription Strategies (ADP) can adjust the size and range of the VOI according to the subject's brain anatomical features, making it fit the target brain region more closely. Nevertheless, it may also affect signal intensity, spectral quality, and the stability of metabolite quantification due to changes in VOI size.
This study compares the SVS measurement performance of ADP and FIX in the Anterior Cingulate Cortex (ACC) and Dorsolateral Prefrontal Cortex (DLPFC) to evaluate the impact of ADP on spectral quality, metabolite quantification, and scan- rescan reproducibility. We used Osprey for SVS data preprocessing, spectral fitting, and metabolite quantification analysis. Signal-to-Noise Ratio (SNR), Full Width at Half Maximum (FWHM), and Relative Residual Amplitude (relResA) were used as spectral quality metrics. For metabolite quantification, tissue-corrected water-scaled (TCW) and ratio to tCr (tCr-scaled) were adopted as quantification metrics, with relative percentage differences used to describe the variations between the two. Finally, the Coefficient of Variation (CV) was used to evaluate reproducibility under repeated measurements.
The research results indicate that, compared to FIX, ADP overall exhibits a lower SNR, demonstrating that its signal intensity might be affected by the VOI size. However, the overall performance of relResA for ADP is superior to that of FIX, indicating that a lower SNR does not simultaneously reflect an increase in residuals after fitting. In terms of metabolite quantification, ADP does not cause an overall quantification bias for most metabolites. Finally, regarding reproducibility, most metabolites show stable repeated measurement performance under both ADP and FIX, with the variation amplitude of CV between the two being within approximately 3 percentage points.

[1] Grover, V. P. B., Tognarelli, J. M., Crossey, M. M. E., Cox, I. J., Taylor-Robinson, S. D., & McPhail, M. J. W. (2015). Magnetic Resonance Imaging: Principles and Techniques: Lessons for Clinicians. Journal of Clinical and Experimental Hepatology, 5(3), 246–255. doi:10.1016/j.jceh.2015.08.001

[2] Tognarelli, J. M., Dawood, M., Shariff, M. I. F., Grover, V. P. B., Crossey, M. M. E., Cox, I. J., Taylor-Robinson, S. D., & McPhail, M. J. W. (2015). Magnetic Resonance Spectroscopy: Principles and Techniques: Lessons for Clinicians. Journal of Clinical and Experimental Hepatology, 5(4), 320–328. doi:10.1016/j.jceh.2015.10.006

[3] Quadrelli, S., Mountford, C., & Ramadan, S. (2016). Hitchhiker’s Guide to Voxel Segmentation for Partial Volume Correction of in vivo Magnetic Resonance Spectroscopy. Magnetic Resonance Insights, 9, 1–8. doi:10.4137/MRI.S32903

[4] DeMayo, M. M., McGirr, A., Selby, B., MacMaster, F. P., Debert, C. T., & Harris, A. D. (2023). Consistency of frontal cortex metabolites quantified by magnetic resonance spectroscopy within overlapping small and large voxels. Scientific Reports, 13, 2246. doi:10.1038/s41598-023-29190-y

[5] Terpstra, M., Cheong, I., Lyu, T., Deelchand, D. K., Emir, U. E., Bednařík, P., Eberly, L. E., & Öz, G. (2016). Test-retest reproducibility of neurochemical profiles with short-echo, single-voxel MR spectroscopy at 3T and 7T. Magnetic Resonance in Medicine, 76(4), 1083–1091. doi:10.1002/mrm.26022

[6] Stevens, F. L., Hurley, R. A., & Taber, K. H. (2011). Anterior cingulate cortex: Unique role in cognition and emotion. The Journal of Neuropsychiatry and Clinical Neurosciences, 23(2), 121–125. doi:10.1176/jnp.23.2.jnp121

[7] Barbey, A. K., Koenigs, M., & Grafman, J. (2013). Dorsolateral prefrontal contributions to human working memory. Cortex, 49(5), 1195–1205. doi:10.1016/j.cortex.2012.05.022

[8] Fornito, A., Whittle, S., Wood, S. J., Velakoulis, D., Pantelis, C., & Yücel, M. (2006). The influence of sulcal variability on morphometry of the human anterior cingulate and paracingulate cortex. NeuroImage, 33(3), 843–854. doi:10.1016/j.neuroimage.2006.06.061

[9] Billeke, P., & Aboitiz, F. (2013). Social cognition in schizophrenia: From social stimuli processing to social engagement. Frontiers in Psychiatry, 4, 4. doi:10.3389/fpsyt.2013.00004 圖源:Brain areas that participate in social processing，Wikimedia Commons， CC BY 3.0 授權。

[10] Oeltzschner, G., Zöllner, H. J., Hui, S. C. N., Mikkelsen, M., Saleh, M. G., Tapper, S., & Edden, R. A. E. (2020). Osprey: Open-source processing, reconstruction & estimation of magnetic resonance spectroscopy data. Journal of Neuroscience Methods, 343, 108827. doi:10.1016/j.jneumeth.2020.108827

[11] Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6), 80–83. doi:10.2307/3001968

[12] Wasserstein, R. L., & Lazar, N. A. (2016). The ASA’s Statement on p-Values: Context, Process, and Purpose. The American Statistician, 70(2), 129–133. doi:10.1080/00031305.2016.1154108

[13] Armstrong, R. A. (2014). When to use the Bonferroni correction. Ophthalmic and Physiological Optics, 34(5), 502–508. doi:10.1111/opo.12131

[14] Quan, H., & Shih, W. J. (1996). Assessing reproducibility by the within-subject coefficient of variation with random effects models. Biometrics, 52(4), 1195–1203. doi:10.2307/2532835