1. NCCU Academic Hub
  2. Publications
  3. College of Science
  4. Theses
  5. Item

Publications-Theses

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 在磁共振頻譜上比較活體頻譜跟模擬頻譜的相似度
Comparison of similarity between in vivo spectra and the simulated spectra in MRS
作者 李昕縈
LEE, Xin-Ying
貢獻者 蔡尚岳
Tsai, Shang Yueh
李昕縈
LEE, Xin-Ying
關鍵詞 磁共振頻譜
FID-A 工具包
模擬活體人腦頻譜
頻譜分析工具
Magnetic resonance spectrum
FID-A toolkit
Simulated human brain spectra
Spectrum analysis tools
日期 2024
上傳時間 4-Sep-2024 15:26:20 (UTC+8)
摘要 磁共振頻譜是一種量測生物體內生化代謝訊息的非侵入式方法，目前多應用於生物醫學領域的研究，特別是針對大腦組織中的代謝物質進行分析。頻譜模擬可以幫助頻譜處理演算法開發，提供標準答案以優化演算法。此外，深度學習目前也大量應用在頻譜處理中，可實現自動特徵提取、頻譜分類和定量分析，並且頻譜模擬可提供大量訓練資料。目前有多種工具可用於模擬磁共振頻譜，以量化其中代謝物的濃度。本研究使用FID-A模擬頻譜，旨在尋找一種最適合分析頻譜相似度的指標。我們使用相關係數、內積、平均絕對百分比誤差和相互資訊這四種方法進行頻譜相似度比較。確定FID-A能夠模擬出相似活體頻譜中結果後，將模擬頻譜視為基準真相，作為量化活體頻譜的參考值，為了更接近活體頻譜，我們在基準真相中添加雜訊，並隨機產生十一萬組與基準真相代謝物濃度相似的頻譜，比較四項指標的結果。我們觀察到相關係數、內積和平均絕對百分比誤差分別在不同情境下產生較好的結果，指標將會因為雜訊、頻寬等原因影響結果。未來的研究可以根據不同情境選擇適用的指標計算頻譜相似度，同時也可以根據特定區域使用特定指標比較。
Magnetic Resonance Spectroscopy is a non-invasive method for measuring biochemical information within living organisms. It is widely applied in biomedical research, particularly in analyzing metabolites within brain tissues. Spectrum simulation aids in developing spectrum processing algorithms by providing benchmark solutions for optimization. Additionally, deep learning exhibits potential in spectrum processing, enabling automated feature extraction, spectrum classification, and quantitative analysis, with spectrum simulation providing abundant training data. Various tools are available for simulating magnetic resonance spectra to quantify the concentration of metabolites. The tool for simulating the spectrum is FID-A, which aims to identify an optimal metric for analyzing spectrum similarity. Four methods—correlation coefficient, dot product, mean absolute percentage error, and mutual information—are utilized for spectrum similarity comparison. After confirming FID-A's capability to simulate results similar to in vivo spectra, simulated spectra are treated as ground truth for quantifying in vivo spectra. To closely resemble in vivo spectra, noise is added to the ground truth, generating random spectra with metabolite concentrations similar to the ground truth. The results of four metrics are compared, revealing varying performance in different variables due to noise, spectral width, and other factors. Future research may select metrics based on specific scenarios for calculating spectrum similarity and compare them using specific metrics in designated regions.
參考文獻 [1] Gujar, S. K., Maheshwari, S., Björkman-Burtscher, I., & Sundgren, P. C. (2005). Magnetic resonance spectroscopy. Journal of neuro-ophthalmology, 25(3), 217-226. [2] Lukas, L., Devos, A., Suykens, J. A., Vanhamme, L., Howe, F. A., Majós, C., ... & Van Huffel, S. (2004). Brain tumor classification based on long echo proton MRS signals. Artificial intelligence in medicine, 31(1), 73-89. [3] Dager, S. R., Corrigan, N. M., Richards, T. L., & Posse, S. (2008). Research applications of magnetic resonance spectroscopy to investigate psychiatric disorders. Topics in Magnetic Resonance Imaging, 19(2), 81-96. [4] Kreis, R., Hofmann, L., Kuhlmann, B., Boesch, C., Bossi, E., & Hüppi, P. S. (2002). Brain metabolite composition during early human brain development as measured by quantitative in vivo 1H magnetic resonance spectroscopy. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine, 48(6), 949-958. [5] Wilson, M. (2021). Adaptive baseline fitting for MR spectroscopy analysis. Magnetic Resonance in Medicine, 85(1), 13-29. [6] Wilson, M. (2021). spant: An R package for magnetic resonance spectroscopy analysis. Journal of Open Source Software, 6(67), 3646. [7] van Veenendaal, Tamar M., et al. "Glutamate quantification by PRESS or MEGA-PRESS: Validation, repeatability, and concordance." Magnetic resonance imaging 48 (2018): 107-114. [8] Soher, Brian J., et al. "GAVA: spectral simulation for in vivo MRS applications." Journal of magnetic resonance 185.2 (2007): 291-299. [9] Soher, Brian J., et al. "VeSPA: integrated applications for RF pulse design, spectral simulation and MRS data analysis." Proc Int Soc Magn Reson Med. Vol. 19. No. 19. 2011. [10] Simpson, R., Devenyi, G. A., Jezzard, P., Hennessy, T. J., & Near, J. (2017). Advanced processing and simulation of MRS data using the FID appliance (FID‐A)—an open source, MATLAB‐based toolkit. Magnetic resonance in medicine, 77(1), 23-33. [11] Landheer, Karl, Kelley M. Swanberg, and Christoph Juchem. "Magnetic resonance Spectrum simulator (MARSS), a novel software package for fast and computationally efficient basis set simulation." NMR in Biomedicine 34.5 (2021): e4129. [12] Govindaraju, V., Young, K., & Maudsley, A. A. (2000). Proton NMR chemical shifts and coupling constants for brain metabolites. NMR in Biomedicine: An International Journal Devoted to the Development and Application of Magnetic Resonance In Vivo, 13(3), 129-153.
描述 碩士
國立政治大學
應用物理研究所
110755004
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0110755004
資料類型 thesis
dc.contributor.advisor 蔡尚岳zh_TW
dc.contributor.advisor Tsai, Shang Yuehen_US
dc.contributor.author (Authors) 李昕縈zh_TW
dc.contributor.author (Authors) LEE, Xin-Yingen_US
dc.creator (作者) 李昕縈zh_TW
dc.creator (作者) LEE, Xin-Yingen_US
dc.date (日期) 2024en_US
dc.date.accessioned 4-Sep-2024 15:26:20 (UTC+8)-
dc.date.available 4-Sep-2024 15:26:20 (UTC+8)-
dc.date.issued (上傳時間) 4-Sep-2024 15:26:20 (UTC+8)-
dc.identifier (Other Identifiers) G0110755004en_US
dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/153486-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 應用物理研究所zh_TW
dc.description (描述) 110755004zh_TW
dc.description.abstract (摘要) 磁共振頻譜是一種量測生物體內生化代謝訊息的非侵入式方法，目前多應用於生物醫學領域的研究，特別是針對大腦組織中的代謝物質進行分析。頻譜模擬可以幫助頻譜處理演算法開發，提供標準答案以優化演算法。此外，深度學習目前也大量應用在頻譜處理中，可實現自動特徵提取、頻譜分類和定量分析，並且頻譜模擬可提供大量訓練資料。目前有多種工具可用於模擬磁共振頻譜，以量化其中代謝物的濃度。本研究使用FID-A模擬頻譜，旨在尋找一種最適合分析頻譜相似度的指標。我們使用相關係數、內積、平均絕對百分比誤差和相互資訊這四種方法進行頻譜相似度比較。確定FID-A能夠模擬出相似活體頻譜中結果後，將模擬頻譜視為基準真相，作為量化活體頻譜的參考值，為了更接近活體頻譜，我們在基準真相中添加雜訊，並隨機產生十一萬組與基準真相代謝物濃度相似的頻譜，比較四項指標的結果。我們觀察到相關係數、內積和平均絕對百分比誤差分別在不同情境下產生較好的結果，指標將會因為雜訊、頻寬等原因影響結果。未來的研究可以根據不同情境選擇適用的指標計算頻譜相似度，同時也可以根據特定區域使用特定指標比較。zh_TW
dc.description.abstract (摘要) Magnetic Resonance Spectroscopy is a non-invasive method for measuring biochemical information within living organisms. It is widely applied in biomedical research, particularly in analyzing metabolites within brain tissues. Spectrum simulation aids in developing spectrum processing algorithms by providing benchmark solutions for optimization. Additionally, deep learning exhibits potential in spectrum processing, enabling automated feature extraction, spectrum classification, and quantitative analysis, with spectrum simulation providing abundant training data. Various tools are available for simulating magnetic resonance spectra to quantify the concentration of metabolites. The tool for simulating the spectrum is FID-A, which aims to identify an optimal metric for analyzing spectrum similarity. Four methods—correlation coefficient, dot product, mean absolute percentage error, and mutual information—are utilized for spectrum similarity comparison. After confirming FID-A's capability to simulate results similar to in vivo spectra, simulated spectra are treated as ground truth for quantifying in vivo spectra. To closely resemble in vivo spectra, noise is added to the ground truth, generating random spectra with metabolite concentrations similar to the ground truth. The results of four metrics are compared, revealing varying performance in different variables due to noise, spectral width, and other factors. Future research may select metrics based on specific scenarios for calculating spectrum similarity and compare them using specific metrics in designated regions.en_US
dc.description.tableofcontents 第一章 緒論 1 1.1 磁共振頻譜 1 1.2 研究動機 3 第二章 研究方法 4 2.1 模擬頻譜 4 2.1.1 模擬代謝物頻譜 6 2.1.2 拓寬 11 2.1.3 大分子 11 2.1.4 添加雜訊 15 2.2 分析指標 17 2.2.1 相關係數 18 2.2.2 內積 18 2.2.3 平均絕對百分比誤差 18 2.2.4 相互資訊 19 第三章 研究結果 22 3.1 模擬頻譜與活體頻譜比較 22 3.2 分析指標的比較 29 第四章 結論 41 參考文獻 43zh_TW
dc.format.extent 2968101 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0110755004en_US
dc.subject (關鍵詞) 磁共振頻譜zh_TW
dc.subject (關鍵詞) FID-A 工具包zh_TW
dc.subject (關鍵詞) 模擬活體人腦頻譜zh_TW
dc.subject (關鍵詞) 頻譜分析工具zh_TW
dc.subject (關鍵詞) Magnetic resonance spectrumen_US
dc.subject (關鍵詞) FID-A toolkiten_US
dc.subject (關鍵詞) Simulated human brain spectraen_US
dc.subject (關鍵詞) Spectrum analysis toolsen_US
dc.title (題名) 在磁共振頻譜上比較活體頻譜跟模擬頻譜的相似度zh_TW
dc.title (題名) Comparison of similarity between in vivo spectra and the simulated spectra in MRSen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1] Gujar, S. K., Maheshwari, S., Björkman-Burtscher, I., & Sundgren, P. C. (2005). Magnetic resonance spectroscopy. Journal of neuro-ophthalmology, 25(3), 217-226. [2] Lukas, L., Devos, A., Suykens, J. A., Vanhamme, L., Howe, F. A., Majós, C., ... & Van Huffel, S. (2004). Brain tumor classification based on long echo proton MRS signals. Artificial intelligence in medicine, 31(1), 73-89. [3] Dager, S. R., Corrigan, N. M., Richards, T. L., & Posse, S. (2008). Research applications of magnetic resonance spectroscopy to investigate psychiatric disorders. Topics in Magnetic Resonance Imaging, 19(2), 81-96. [4] Kreis, R., Hofmann, L., Kuhlmann, B., Boesch, C., Bossi, E., & Hüppi, P. S. (2002). Brain metabolite composition during early human brain development as measured by quantitative in vivo 1H magnetic resonance spectroscopy. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine, 48(6), 949-958. [5] Wilson, M. (2021). Adaptive baseline fitting for MR spectroscopy analysis. Magnetic Resonance in Medicine, 85(1), 13-29. [6] Wilson, M. (2021). spant: An R package for magnetic resonance spectroscopy analysis. Journal of Open Source Software, 6(67), 3646. [7] van Veenendaal, Tamar M., et al. "Glutamate quantification by PRESS or MEGA-PRESS: Validation, repeatability, and concordance." Magnetic resonance imaging 48 (2018): 107-114. [8] Soher, Brian J., et al. "GAVA: spectral simulation for in vivo MRS applications." Journal of magnetic resonance 185.2 (2007): 291-299. [9] Soher, Brian J., et al. "VeSPA: integrated applications for RF pulse design, spectral simulation and MRS data analysis." Proc Int Soc Magn Reson Med. Vol. 19. No. 19. 2011. [10] Simpson, R., Devenyi, G. A., Jezzard, P., Hennessy, T. J., & Near, J. (2017). Advanced processing and simulation of MRS data using the FID appliance (FID‐A)—an open source, MATLAB‐based toolkit. Magnetic resonance in medicine, 77(1), 23-33. [11] Landheer, Karl, Kelley M. Swanberg, and Christoph Juchem. "Magnetic resonance Spectrum simulator (MARSS), a novel software package for fast and computationally efficient basis set simulation." NMR in Biomedicine 34.5 (2021): e4129. [12] Govindaraju, V., Young, K., & Maudsley, A. A. (2000). Proton NMR chemical shifts and coupling constants for brain metabolites. NMR in Biomedicine: An International Journal Devoted to the Development and Application of Magnetic Resonance In Vivo, 13(3), 129-153.zh_TW