請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/99348完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 陳冠元 | zh_TW |
| dc.contributor.advisor | Guan-Yuan Chen | en |
| dc.contributor.author | 陳琦韻 | zh_TW |
| dc.contributor.author | Chi-Yun Chen | en |
| dc.date.accessioned | 2025-09-09T16:06:54Z | - |
| dc.date.available | 2025-09-10 | - |
| dc.date.copyright | 2025-09-09 | - |
| dc.date.issued | 2025 | - |
| dc.date.submitted | 2025-07-01 | - |
| dc.identifier.citation | 1. UNODC. NPS Leaflet: New Psychoactive Substances. 2016; Available from: https://www.unodc.org/unodc/en/scientists/leaflet_-new-psychoactive-substances-2016.html.
2. UNODC. Online World Drug Report 2024 - Drug market patterns and trends. 2024 [cited May 2025; Available from: https://www.unodc.org/unodc/en/data-and-analysis/wdr2024-drug-market-trends.html. 3. UNODC. NPS Data Visualisations. [cited May 2025; Available from: https://www.unodc.org/LSS/Page/NPS/DataVisualisations. 4. UNODC. The challenge of New Psychoactive Substances: A technical update 2024. 2024; Available from: https://www.unodc.org/documents/scientific/The_Challenge_of_NPS_A_technical_update_2024.pdf. 5. UNODC. UNODC: The challenge of New Psychoactive Substances – A technical update 2024 is launched. [cited May 2025; Available from: https://www.unodc.org/unodc/en/scientists/the-challenge-of-new-psychoactive-substances.html. 6. Rinaldi, R., et al., The rise of new psychoactive substances and psychiatric implications: A wide-ranging, multifaceted challenge that needs far-reaching common legislative strategies. Human Psychopharmacology: Clinical and Experimental, 2020. 35(3): p. e2727. 7. EMCDDA. European Drug Report 2024: Trends and Developments. [cited May 2025; Available from: https://www.emcdda.europa.eu/publications/european-drug-report/2024_en. 8. Yeh, H.T., et al., Clinical Presentations and Predictors of In-Hospital Mortality in Illicit Drug Users in the New Psychoactive Substances (NPS) Endemic Era in Taiwan. Toxics, 2022. 10(7). 9. Mbughuni, M.M., P.J. Jannetto, and L.J. Langman, Mass Spectrometry Applications for Toxicology. Ejifcc, 2016. 27(4): p. 272-287. 10. Grebe, S.K. and R.J. Singh, LC-MS/MS in the Clinical Laboratory - Where to From Here? Clin Biochem Rev, 2011. 32(1): p. 5-31. 11. Partridge, E., et al., A Case Study Involving U-47700, Diclazepam and Flubromazepam-Application of Retrospective Analysis of HRMS Data. J Anal Toxicol, 2018. 42(9): p. 655-660. 12. Ibáñez, M., et al., Comprehensive analytical strategies based on high-resolution time-of-flight mass spectrometry to identify new psychoactive substances. TrAC Trends in Analytical Chemistry, 2014. 57: p. 107-117. 13. Aradi, S., Survey of Deep Reinforcement Learning for Motion Planning of Autonomous Vehicles. IEEE Transactions on Intelligent Transportation Systems, 2022. 23(2): p. 740-759. 14. Almalioglu, Y., et al., Deep learning-based robust positioning for all-weather autonomous driving. Nature Machine Intelligence, 2022. 4(9): p. 749-760. 15. Bachute, M.R. and J.M. Subhedar, Autonomous Driving Architectures: Insights of Machine Learning and Deep Learning Algorithms. Machine Learning with Applications, 2021. 6: p. 100164. 16. Khurana, D., et al., Natural language processing: state of the art, current trends and challenges. Multimedia Tools and Applications, 2023. 82(3): p. 3713-3744. 17. Popel, M., et al., Transforming machine translation: a deep learning system reaches news translation quality comparable to human professionals. Nature Communications, 2020. 11(1): p. 4381. 18. Tian, Y., Artificial Intelligence Image Recognition Method Based on Convolutional Neural Network Algorithm. IEEE Access, 2020. 8: p. 125731-125744. 19. Miller, R.A., H.E. Pople, Jr., and J.D. Myers, Internist-1, an experimental computer-based diagnostic consultant for general internal medicine. N Engl J Med, 1982. 307(8): p. 468-76. 20. Esteva, A., et al., Dermatologist-level classification of skin cancer with deep neural networks. Nature, 2017. 542(7639): p. 115-118. 21. Gulshan, V., et al., Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA, 2016. 316(22): p. 2402-2410. 22. Segal, N.H., et al., Classification and subtype prediction of adult soft tissue sarcoma by functional genomics. Am J Pathol, 2003. 163(2): p. 691-700. 23. Zhao, X., et al., Predicting renal function recovery and short-term reversibility among acute kidney injury patients in the ICU: comparison of machine learning methods and conventional regression. Ren Fail, 2022. 44(1): p. 1326-1337. 24. Skinnider, M.A., et al., A deep generative model enables automated structure elucidation of novel psychoactive substances. Nature Machine Intelligence, 2021. 3(11): p. 973-984. 25. Xu, M., et al., High accuracy machine learning identification of fentanyl-relevant molecular compound classification via constituent functional group analysis. Scientific Reports, 2020. 10(1): p. 13569. 26. Koshute, P., N. Hagan, and N.J. Jameson, Machine learning model for detecting fentanyl analogs from mass spectra. Forensic Chemistry, 2022. 27: p. 100379. 27. Wong, S.L., et al., Screening unknown novel psychoactive substances using GC–MS based machine learning. Forensic Chemistry, 2023. 34: p. 100499. 28. Lee, S.Y., et al., Revealing Unknown Controlled Substances and New Psychoactive Substances Using High-Resolution LC–MS-MS Machine Learning Models and the Hybrid Similarity Search Algorithm. Journal of Analytical Toxicology, 2021. 46(7): p. 732-742. 29. Malakouti, S.M., M.B. Menhaj, and A.A. Suratgar, The usage of 10-fold cross-validation and grid search to enhance ML methods performance in solar farm power generation prediction. Cleaner Engineering and Technology, 2023. 15: p. 100664. 30. Cervantes, J., et al., A comprehensive survey on support vector machine classification: Applications, challenges and trends. Neurocomputing, 2020. 408: p. 189-215. 31. Taunk, K., et al. A Brief Review of Nearest Neighbor Algorithm for Learning and Classification. in 2019 International Conference on Intelligent Computing and Control Systems (ICCS). 2019. 32. Friedman, J.H., Stochastic gradient boosting. Computational Statistics & Data Analysis, 2002. 38(4): p. 367-378. 33. Wu, Y.-c. and J.-w. Feng, Development and Application of Artificial Neural Network. Wireless Personal Communications, 2018. 102(2): p. 1645-1656. 34. Gardner, M.W. and S.R. Dorling, Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric Environment, 1998. 32(14): p. 2627-2636. 35. UNODC. Current NPS Threats: Volume VII. July 2024; Available from: https://www.unodc.org/documents/scientific/Current_NPS_threats_VII.pdf. 36. Révész, Á., et al., Collision energies: Optimization strategies for bottom-up proteomics. Mass Spectrometry Reviews, 2023. 42(4): p. 1261-1299. 37. Weinmann, W., et al., Simultaneous determination of THC-COOH and THC-COOH-glucuronide in urine samples by LC/MS/MS. Forensic Science International, 2000. 113(1): p. 381-387. 38. Concheiro, M., D.M. Shakleya, and M.A. Huestis, Simultaneous quantification of buprenorphine, norbuprenorphine, buprenorphine-glucuronide and norbuprenorphine-glucuronide in human umbilical cord by liquid chromatography tandem mass spectrometry. Forensic Science International, 2009. 188(1): p. 144-151. 39. Fornal, E., Identification of substituted cathinones: 3,4-Methylenedioxy derivatives by high performance liquid chromatography–quadrupole time of flight mass spectrometry. Journal of Pharmaceutical and Biomedical Analysis, 2013. 81-82: p. 13-19. 40. Davidson, J.T., Z.J. Sasiene, and G.P. Jackson, Fragmentation pathways of odd- and even-electron N-alkylated synthetic cathinones. International Journal of Mass Spectrometry, 2020. 453: p. 116354. 41. Huang, Y., et al., A comprehensive analytical strategy based on characteristic fragments to detect synthetic cannabinoid analogs in seized products and hair samples. Talanta, 2023. 265: p. 124830. 42. Mata-Pesquera, M., et al., Exploiting the triple quadrupole mass analyzer for the open detection and tentative identification of synthetic cannabinoid receptor agonists based on common fragmentation pathways. Analytica Chimica Acta, 2024. 1329: p. 343226. 43. Fan, Y., et al., Development of a fragmentation pattern of synthetic cannabinoids based on electrospray ionization mass spectrometry in positive ion mode to screen synthetic cannabinoids in illicit products. Journal of Pharmaceutical and Biomedical Analysis, 2021. 193: p. 113723. 44. Galaon, T., et al., Simultaneous ESI-APCI(+) ionization and fragmentation pathways for nine benzodiazepines and zolpidem using single quadrupole LC-MS. Drug Testing and Analysis, 2014. 6(5): p. 439-450. 45. Kong, R., et al., Metabolic profiling of clonazolam in human liver microsomes and zebrafish models using liquid chromatography quadrupole Orbitrap mass spectrometry. Journal of Chromatography B, 2023. 1216: p. 123583. 46. Bijlsma, L., et al., Fragmentation pathways of drugs of abuse and their metabolites based on QTOF MS/MS and MSE accurate-mass spectra. Journal of Mass Spectrometry, 2011. 46(9): p. 865-875. 47. Pasin, D., et al., Characterization of hallucinogenic phenethylamines using high-resolution mass spectrometry for non-targeted screening purposes. Drug Testing and Analysis, 2017. 9(10): p. 1620-1629. 48. Matey, J.M., et al., Identification of new psychoactive substances and their metabolites using non-targeted detection with high-resolution1https://iuicp.uah.es/es/. mass spectrometry2https://cinquifor.uah.es/index-en.htm. through diagnosing fragment ions/neutral loss analysis. Talanta, 2023. 265: p. 124816. 49. Schäfer, M., et al., A fast screening method for tricyclic antidepressants and their urinary metabolites by FAB-tandem mass spectrometry. Journal of Spectroscopy, 1997. 13(3): p. 213-226. 50. Harris, D.N., et al., Fragmentation differences in the EI spectra of three synthetic cannabinoid positional isomers: JWH-250, JWH-302, and JWH-201. International Journal of Mass Spectrometry, 2014. 368: p. 23-29. 51. Pulver, B., et al., A new synthetic cathinone: 3,4-EtPV or 3,4-Pr-PipVP? An unsuccessful attempt to circumvent the German legislation on new psychoactive substances. Drug Testing and Analysis, 2023. 15(1): p. 84-96. 52. Predictor Importance in STATISTICA GC&RT, Interactive Trees, and Boosted Trees. [cited May, 2024; Available from: https://docs.tibco.com/data-science/GUID-4C6F72C1-F4F8-48A9-83C7-D4C72A66A3AC.html. 53. Tamama, K., Chapter Five - Synthetic drugs of abuse, in Advances in Clinical Chemistry, G.S. Makowski, Editor. 2021, Elsevier. p. 191-214. 54. Majchrzak, M., et al., The newest cathinone derivatives as designer drugs: an analytical and toxicological review. Forensic Toxicology, 2018. 36(1): p. 33-50. 55. Davidson, J.T., et al., Fragmentation pathways of α-pyrrolidinophenone synthetic cathinones and their application to the identification of emerging synthetic cathinone derivatives. International Journal of Mass Spectrometry, 2020. 453: p. 116343. 56. Pozo, Ó.J., et al., Mass Spectrometric Evaluation of Mephedrone In Vivo Human Metabolism: Identification of Phase I and Phase II Metabolites, Including a Novel Succinyl Conjugate. Drug Metabolism and Disposition, 2015. 43(2): p. 248-257. 57. Niessen, W.M.A., Tandem mass spectrometry of organic nitro and halogen compounds: Competition between losses of molecules and of radicals. International Journal of Mass Spectrometry, 2021. 460: p. 116496. 58. Ibáñez, M., et al., Analytical strategy to investigate 3,4-methylenedioxypyrovalerone (MDPV) metabolites in consumers’ urine by high-resolution mass spectrometry. Analytical and Bioanalytical Chemistry, 2016. 408(1): p. 151-164. 59. Apirakkan, O., et al., Analytical characterization of three cathinone derivatives, 4-MPD, 4F–PHP and bk-EPDP, purchased as bulk powder from online vendors. Drug Testing and Analysis, 2018. 10(2): p. 372-378. 60. Mata-Pesquera, M., et al., Characterization of the recently detected cathinone N-cyclohexyl butylone: From structure elucidation to in silico supported pharmacological/toxicological considerations. Microchemical Journal, 2023. 190: p. 108577. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/99348 | - |
| dc.description.abstract | 近年來，新興影響精神物質（NPS）的迅速發展對公共衛生與法律監管制度造成了極大的危害。由於NPS的化學結構的多樣性與快速變化，傳統法規難以有效監管其濫用。此外，質譜分析技術依賴標準品與數據庫，對於未知或新興NPS的檢測能力受限，難以應對其結構變異所帶來的挑戰。
本研究旨在應用機器學習對NPS二級質譜圖進行分類，並探討其在質譜數據分析中的應用潛力。採用多種監督式機器學習模型，包括支援向量機（SVM）、K-近鄰（KNN）、提升樹（BT）與人工神經網路（ANN），以提升NPS分類的準確性與泛用性。 研究中使用超效能液相層析串聯質譜（UPLC-MS/MS）技術，於質荷比範圍m/z 50 - m/z 500內記錄質譜數據，並在碰撞能量10、20、40 eV條件下獲取產物離子圖譜。共收集394種包含NPS及相關藥物的質譜數據，並依其化學結構分為五大類：合成卡西酮（synthetic cathinones）、合成類大麻（synthetic cannabinoids）、苯二氮平類（benzodiazepines）、苯乙胺類（phenethylamines）及其他類（others）。 為了評估分類模型效能，本研究採用10折交叉驗證（10-fold cross-validation），並以準確率（accuracy）與F1 score作為評估指標。於驗證集資料，模型的F1 score介於0.58-1.00之間。在獨立測試集中，透過模型投票機制獲得整體準確率為87.5%（範圍62.5% - 100%）。 此外，本研究透過變數重要性分析(variable importance analysis)與質譜數據解釋，進一步解析合成卡西酮的結構特徵。在碰撞誘導解離（CID）之下，顯示N-吡咯烷（N-pyrrolidinyl）、N-烷基化（N-alkylated）、3,4-亞甲二氧基（3,4-methylenedioxy）修飾的合成卡西酮具有特定的中性丟失模式，可用於推測各類別卡西酮的碎裂途徑。 預期本研究可有效應用於NPS的分類與鑑定，並可進一步用於未知NPS的結構解析與法醫毒理學的檢測與鑑定。 | zh_TW |
| dc.description.abstract | The rapid emergence of new psychoactive substances (NPS) has posed significant challenges to public health and regulatory frameworks. Due to the structural diversity and continuous evolution of NPS, conventional legal and analytical approaches struggle to effectively control their abuse. Furthermore, mass spectrometry techniques, which rely on reference standards and established databases, often struggle to identify novel or structurally modified NPS.
This study aims to explore the potential of machine learning algorithms in NPS classification and their application in mass spectrometric data analysis. We employed supervised machine learning models, including support vector machine (SVM), k-nearest neighbors (KNN), boosted trees (BT), and artificial neural networks (ANN), to improve the accuracy and generalization of NPS classification. Ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) was utilized to acquire mass spectra within the range of m/z 50 - 500, using collision energies of 10, 20, and 40 eV to generate product ion spectra. A total of 394 substances, including NPS and related drugs, were analyzed and categorized into five classes: synthetic cathinones, synthetic cannabinoids, benzodiazepines, phenethylamines, and others. Model performance was assessed using 10-fold cross-validation, with accuracy and F1 score serving as evaluation metrics. The models achieved F1 scores ranging from 0.58 to 1.00 on the validation dataset. In an independent testing dataset, ensemble model voting resulted in an overall accuracy of 87.5% (ranging from 62.5% to 100%). Furthermore, variable importance analysis and mass spectral interpretation were conducted to elucidate the structural characteristics of synthetic cathinones. Collision-induced dissociation revealed distinct neutral loss patterns for N-pyrrolidinyl, N-alkylated, and 3,4-methylenedioxy derivative cathinones, providing insights into potential fragmentation pathways. We expect that this research will demonstrate the efficacy of machine learning in the classification and identification of NPS, offering a robust approach for the detection and structural elucidation of unknown NPS in forensic toxicology. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2025-09-09T16:06:54Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2025-09-09T16:06:54Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 口試委員會審定書 i
致謝 ii 中文摘要 iii Abstract iv Contents vi List of Figures viii List of Tables x Chapter 1 Introduction 1 1.1 Overview of New Psychoactive Substances 1 1.2 Detection Methods of NPS 2 1.3 Applications of Machine Learning in Forensic Toxicology 4 1.4 Aim of This Study 6 Chapter 2 Material and Methods 7 2.1 Chemicals and Reagents 7 2.2 Sample Preparation 25 2.3 Instruments 26 2.4 Data Preprocessing 27 2.5 Feature Engineering 27 2.5.1 Feature Generation 27 2.5.2 Feature Selection 29 2.6 Machine Learning Algorithms 30 2.6.1 Support Vector Machine 31 2.6.2 K-Nearest Neighbors 32 2.6.3 Boosted Trees 32 2.6.4 Artificial Neural Network 33 2.6.5 Models Deployment and Voted 34 2.7 Model Evaluation 34 Chapter 3 Results and Discussion 36 3.1 Workflow of This Study 36 3.2 Dataset Construction 36 3.2.1 Characteristic of NPS and Dataset Compilation 36 3.2.2 Acquisition of Tandem Mass Spectrometry Data 40 3.3 Development of Machine Learning Model 44 3.3.1 Feature Engineering: Mass Spectral Distribution and Dataset Analysis 44 3.3.2 Performance Evaluation and Validation of the Model 56 3.3.3 Accuracy of Testing Sample Classification 62 3.3.4 Variable Importance Analysis 70 3.4 Structural Elucidation of synthetic cathinones 72 3.4.1 Features of Fragment Ion Data 73 3.4.2 Features of Neutral Loss Data 80 3.4.3 Proposed fragmentation pathway of Synthetic Cathinones 84 Chapter 4 Conclusion 101 References 102 | - |
| dc.language.iso | zh_TW | - |
| dc.subject | 二級質譜 | zh_TW |
| dc.subject | 新興影響精神物質 | zh_TW |
| dc.subject | 結構解析 | zh_TW |
| dc.subject | 超效能液相層析串聯質譜 | zh_TW |
| dc.subject | 機器學習 | zh_TW |
| dc.subject | Machine learning | en |
| dc.subject | UPLC-MS/MS | en |
| dc.subject | Structure elucidation | en |
| dc.subject | Tandem mass spectrum | en |
| dc.subject | New psychoactive substances (NPS) | en |
| dc.title | 利用機器學習及二級質譜資訊對新興影響精神物質進行分類與結構解析 | zh_TW |
| dc.title | Classification and Structure Elucidation of New Psychoactive Substances Using Machine Learning and Tandem Mass Spectrometry Data | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 113-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 翁德怡;王三源 | zh_TW |
| dc.contributor.oralexamcommittee | Te-I Weng;San-Yuan Wang | en |
| dc.subject.keyword | 新興影響精神物質,二級質譜,機器學習,超效能液相層析串聯質譜,結構解析, | zh_TW |
| dc.subject.keyword | New psychoactive substances (NPS),Tandem mass spectrum,Machine learning,UPLC-MS/MS,Structure elucidation, | en |
| dc.relation.page | 106 | - |
| dc.identifier.doi | 10.6342/NTU202501361 | - |
| dc.rights.note | 同意授權(限校園內公開) | - |
| dc.date.accepted | 2025-07-02 | - |
| dc.contributor.author-college | 醫學院 | - |
| dc.contributor.author-dept | 法醫學研究所 | - |
| dc.date.embargo-lift | 2030-06-30 | - |
| 顯示於系所單位： | 法醫學科所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-113-2.pdf 未授權公開取用 | 3.02 MB | Adobe PDF | 檢視/開啟 |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。