MoCab: A Model Management System based on FHIR for Clinical Decision Support

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：57

、訪客IP：3.139.103.49

姓名

郭哲銘(Zhe-Ming Kuo) 查詢紙本館藏

畢業系所

資訊管理學系

論文名稱

(MoCab: A Model Management System based on FHIR for Clinical Decision Support)

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 (2026-7-24以後開放)

摘要(中)

在醫學領域中，有許多的研究人員開發出各種機器學習模型以支援臨床決策。然而，電子健康記錄(electronic health records, EHR)的格式在不同機構間存在差異，使得將多個模型整合到醫療資訊系統(hospital information system, HIS)中變得具有挑戰性且耗時。
為了解決這個問題，我們提出了 MoCab 框架，該框架使用 Fast Healthcare Interoperability Resources (FHIR)作為數據存儲和檢索標準，以將模型部署到多個醫療資訊系統之中。MoCab 通過標準化的配置設定和導入已保存的模型來部署模型，並基於 FHIR 標準檢索患者數據並將其傳遞給所需的模型來進行預測。基於預測結果，MoCab 透過 Clinical Decision Support (CDS) Hooks 給予醫生警示或通知，並使用基於 SMART on FHIR 架構所開發的應用程式進一步顯示患者的檢驗結果趨勢。此外，MoCab 能夠使用持續出現的新病患數據不斷訓練與優化模型，以提升已部署模型的預測效能。本文演示了三種類型的模型如何在 MoCab 中部署以支援臨床決策。我們所提出的 MoCab 框架可以增強模型在多個醫療資訊系統與電子健康記錄中的可重複利用性，並協助臨床決策的過程。

摘要(英)

Researchers have developed machine learning models to support clinical decision-making in the medical field. However, the format of electronic health records (EHRs) varies across institutions, making integrating these models into a health information system (HIS) challenging and time-consuming.
To address this issue, we proposed MoCab, which uses fast healthcare interoperability resources (FHIR) as data storage and retrieval standards for deploying models to various HISs. MoCab deploys models by configuring and importing the saved model. MoCab makes the prediction by retrieving patients′ data from the FHIR server and passing it to the requested model. If a warning or alert is needed based on the prediction results, MoCab alerts physicians through Clinical Decision Support (CDS) Hooks and displays patient laboratory result trends using Substitutable Medical Apps Reusable Technologies (SMART) on FHIR. Moreover, MoCab can fine-tune and improve the prediction performance of deployed models over time with the endless stream of new data. Three types of models are presented to demonstrate how the models can be implemented in MoCab to support decision-making. The proposed MoCab framework can enhance the reusability of models in multiple EHRs and assist in the clinical decision-making process.

關鍵字(中)

★ Fast Healthcare Interoperability Resources
★ 臨床決策支援
★ 持續訓練
★ 資訊管理系統

關鍵字(英)

★ Fast Healthcare Interoperability Resources
★ Clinical Decision Support
★ Continuous Training
★ Information Management System

論文目次

Abstract ................................................................................................................................ i
摘要 .................................................................................................................................... ii
Table of Contents ............................................................................................................... iii
List of Figure ...................................................................................................................... v
List of Table ....................................................................................................................... vi
I. Introduction ................................................................................................................ 1
II. Literature Review ....................................................................................................... 5
2.1. Integration of FHIR Standard in the Medical Field ................................................ 6
2.2. CDS Hooks and SMART on the FHIR app ............................................................. 7
2.3. Model related CDS in HIS ......................................................................................... 8
2.4. Literature review summary ........................................................................................ 9
III. Method .................................................................................................................. 10
3.1. Part 1 – Data Service Center .................................................................................... 13
3.2. Part 2 - Knowledge Model Center ........................................................................... 15
3.3. Part 3 - Endpoints ...................................................................................................... 17
3.4. Part 4 - Model Retraining Center ............................................................................ 19
3.4.1. Scheduler ...................................................................................................... 20
3.4.2. Data Retrieval Parser .................................................................................... 20
3.4.3. Data Transformation Bundler ....................................................................... 21
3.4.4. Model Trainer ............................................................................................... 23
3.4.5. Model Evaluator ........................................................................................... 23
3.4.6. Model Register ............................................................................................. 24
IV. Result .................................................................................................................... 25
4.1. Quick COVID Severity Index (qCSI) ..................................................................... 25
4.2. Pima Indians Diabetes Prediction Model ............................................................... 27
4.3. Necrotizing Soft Tissue Infections (NSTI) Model ................................................ 29
4.4. Second Primary Cancer (SPC) Model for Lung Cancer ...................................... 31
4.5. Integrations with CDS Hooks and SMART on FHIR .......................................... 33
V. Discussion ................................................................................................................. 37
VI. Conclusion ............................................................................................................ 40
Reference .......................................................................................................................... 41

參考文獻

[1] E. H. Weissler et al., “The role of machine learning in clinical research: transforming the future of evidence generation,” Trials, vol. 22, no. 1. BioMed Central Ltd, Dec. 01, 2021. doi: 10.1186/s13063-021-05489-x.
[2] M. V. Maddali et al., “Validation and utility of ARDS subphenotypes identified by machine-learning models using clinical data: an observational, multicohort, retrospective analysis,” Lancet Respir Med, vol. 10, no. 4, pp. 367–377, Apr. 2022, doi: 10.1016/S2213-2600(21)00461-6.
[3] J. A. M. Sidey-Gibbons and C. J. Sidey-Gibbons, “Machine learning in medicine: a practical introduction,” BMC Med Res Methodol, vol. 19, no. 1, pp. 1–18, Mar. 2019, doi: 10.1186/S12874-019-0681-4/TABLES/5.
[4] A. Boussadi and E. Zapletal, “A Fast Healthcare Interoperability Resources (FHIR) layer implemented over i2b2,” BMC Med Inform Decis Mak, vol. 17, no. 1, pp. 1–12, Aug. 2017, doi: 10.1186/S12911-017-0513-6/FIGURES/8.
[5] G. De Moor et al., “Using electronic health records for clinical research: The case of the EHR4CR project,” J Biomed Inform, vol. 53, pp. 162–173, Feb. 2015, doi: 10.1016/J.JBI.2014.10.006.
[6] FHIR ®© HL7.org, “FHIR Overview,” 2022. https://www.hl7.org/fhir/overview.html (accessed Nov. 25, 2022).
[7] T. J. Osterman, M. Terry, and R. S. Miller, “Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative,” JCO Clin Cancer Inform, no. 4, pp. 993–1001, Nov. 2020, doi: 10.1200/cci.20.00059.
[8] H. Ulrich, A. K. Kock, P. Duhm-Harbeck, J. K. Habermann, and J. Ingenerf, “Metadata repository for improved data sharing and reuse based on HL7 FHIR,” Stud Health Technol Inform, vol. 228, pp. 162–166, 2017, doi: 10.3233/978-1-61499-678-1-162.
[9] S. El-Sappagh, F. Ali, A. Hendawi, J. H. Jang, and K. S. Kwak, “A mobile health monitoring-and-treatment system based on integration of the SSN sensor ontology and the HL7 FHIR standard,” BMC Med Inform Decis Mak, vol. 19, no. 1, pp. 1–36, May 2019, doi: 10.1186/S12911-019-0806-Z/FIGURES/11.
[10] T. F. Peng and H. W. Chiu, “Applying FHIR in Remote Consultation in Dermatology: Interoperability of Information for Digital Dermatoscope,” ACM International Conference Proceeding Series, pp. 28–31, May 2022, doi: 10.1145/3545729.3545739.
[11] K. L. Morgan et al., “Using CDS Hooks to increase SMART on FHIR app utilization: a cluster-randomized trial,” Journal of the American Medical Informatics Association, vol. 29, no. 9, pp. 1461–1470, Aug. 2022, doi: 10.1093/jamia/ocac085.
[12] J. C. Mandel, D. A. Kreda, K. D. Mandl, I. S. Kohane, and R. B. Ramoni, “SMART on FHIR: A standards-based, interoperable apps platform for electronic health records,” Journal of the American Medical Informatics Association, vol. 23, no. 5, pp. 899–908, Sep. 2016, doi: 10.1093/jamia/ocv189.
[13] “CDS Hooks,” Aug. 23, 2022. https://cds-hooks.hl7.org/ (accessed Mar. 13, 2023).
[14] M. Namita Seth Mohta and M. Kenneth D. Mandl, “Informatics Innovations:
How to Achieve ‘Medicine-Based Evidence’ | Catalyst non-issue content,” 2020. https://catalyst.nejm.org/doi/full/10.1056/CAT.20.0026 (accessed Mar.
19, 2023).
[15] K. D. Mandl et al., “Push Button Population Health: The SMART/HL7 FHIR
Bulk Data Access Application Programming Interface,” NPJ Digit Med, vol. 3,
no. 1, Dec. 2020, doi: 10.1038/s41746-020-00358-4.
[16] D. Baylor et al., “Continuous Training for Production ML in the TensorFlow
Extended (TFX) Platform,” in 2019 USENIX Conference on Operational Machine Learning (OpML 19), Santa Clara, CA: USENIX Association, May 2019, pp. 51–53. [Online]. Available: https://www.usenix.org/conference/opml19/presentation/baylor
[17] I. Prapas, B. Derakhshan, A. Rezaei Mahdiraji, and V. Markl, “Continuous Training and Deployment of Deep Learning Models,” Datenbank-Spektrum, doi: 10.1007/s13222-021-00386-8.
[18] R. Hadsell, D. Rao, A. A. Rusu, and R. Pascanu, “Embracing Change: Continual Learning in Deep Neural Networks,” Trends Cogn Sci, vol. 24, no. 12, pp. 1028–1040, Dec. 2020, doi: 10.1016/J.TICS.2020.09.004.
[19] “HL7.FHIR.UV.BULKDATABulk Data Export - FHIR v4.0.1,” Nov. 09, 2021. https://build.fhir.org/ig/HL7/bulk-data/export.html (accessed May 09, 2023).
[20] Rouse Margaret, “What is Directory Services? - Definition from Techopedia,” Jun. 26, 2012. https://www.techopedia.com/definition/18887/directory-services (accessed Jul. 19, 2023).
[21] J. S. Ancker, A. Edwards, S. Nosal, D. Hauser, E. Mauer, and R. Kaushal, “Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system,” BMC Med Inform Decis Mak, vol. 17, no. 1,
pp. 1–9, Apr. 2017, doi: 10.1186/S12911-017-0430-8/FIGURES/1.
[22] M. N. Kamel Boulos, “Location-based health information services: a new
paradigm in personalised information delivery,” 2003, Accessed: Jul. 19, 2023.
[Online]. Available: www.microsoft.com/mobile/smartphone/default.asp
[23] “ImplementationGuide - FHIR v5.0.0-cibuild,” Mar. 05, 2023.
https://build.fhir.org/implementationguide.html (accessed Mar. 06, 2023).
[24] E. Chukwu, L. Garg, E. N. Obande-Ogbuinya, and V. K. Chattu,
“Standardizing Primary Healthcare Referral Datasets in Nigeria: Surveys, Form-reviews and FHIR profiling (Preprint),” JMIR Form Res, 2022, doi: 10.2196/28510.
[25] D. Seong et al., “Fast Healthcare Interoperability Resources (FHIR)-Based Quality Information Exchange for Clinical Next-Generation Sequencing Genomic Testing: Implementation Study,” J Med Internet Res, 2020, doi: 10.2196/26261.
[26] T. W. Tseng, C. F. Su, and F. Lai, “Fast Healthcare Interoperability Resources for Inpatient Deterioration Detection with Time-Series Vital Signs: Design and Implementation Study,” JMIR Med Inform, vol. 10, no. 10, Oct. 2022, doi: 10.2196/42429.
[27] M. Watkins and K. Eilbeck, “FHIR Lab Reports: using SMART on FHIR and CDS Hooks to increase the clinical utility of pharmacogenomic laboratory test results,” 2020. [Online]. Available: https://github.com/mwatkin8/FHIR-Lab- Reports-App.
[28] H. Thiess et al., “Coordinated use of Health Level 7 standards to support
clinical decision support: Case study with shared decision making and drug-drug interactions,” Int J Med Inform, vol. 162, p. 104749, Jun. 2022, doi:
10.1016/j.ijmedinf.2022.104749.
[29] H. P. Lehmann and E. H. Shortliffe, “THOMAS: building Bayesian statistical
expert systems to aid in clinical decision making,” Comput Methods Programs Biomed, vol. 35, no. 4, pp. 251–260, Aug. 1991, doi: 10.1016/0169- 2607(91)90003-C.
[30] N. Peiffer-Smadja et al., “Narrative review Machine learning for clinical decision support in infectious diseases: a narrative review of current applications,” 2019, doi: 10.1016/j.cmi.2019.09.009.
[31] S. K. Lakshmanaprabu, S. N. Mohanty, S. R. S., S. Krishnamoorthy, J. Uthayakumar, and K. Shankar, “Online clinical decision support system using optimal deep neural networks,” Appl Soft Comput, vol. 81, p. 105487, Aug. 2019, doi: 10.1016/J.ASOC.2019.105487.
[32] G. E. Soto and J. A. Spertus, “EPOCH® and ePRISM®: A web-based translational framework for bridging outcomes research and clinical Practice,” Comput Cardiol, vol. 34, pp. 205–208, 2007, doi: 10.1109/CIC.2007.4745457.
[33] J. Gruendner et al., “Ketos: Clinical decision support and machine learning as a service – A training and deployment platform based on Docker, OMOP-CDM, and FHIR Web Services,” PLoS One, vol. 14, no. 10, Oct. 2019, doi: 10.1371/journal.pone.0223010.
[34] “The Heat is On: US Caught FHIR in 2019 - Health IT Buzz Health IT Buzz,” Jul. 29, 2019. https://www.healthit.gov/buzz-blog/health-it/the-heat-is-on-us- caught-fhir-in-2019 (accessed Jun. 16, 2023).
[35] Smile CDR, “HAPI FHIR - The Open Source FHIR API for Java,” 2022.
https://hapifhir.io/ (accessed Nov. 30, 2022).
[36] “fhir.resources · PyPI,” Jan. 01, 2023. https://pypi.org/project/fhir.resources/ (accessed Jun. 16, 2023).
[37] “GitHub - nazrulworld/fhir.resources: FHIR Resources https://www.hl7.org/fhir/resourcelist.html,” Jan. 01, 2023. https://github.com/nazrulworld/fhir.resources (accessed Jun. 16, 2023).
[38] K. Li-Chun, T. Tzu-Yun, and L. Chung-Yueh, “Taiwan Cancer Registry Short Form Implementation Guide, 0.1.0 - ci-build,” Feb. 03, 2023. https://mitw.dicom.org.tw/IG/TWCR/ImplementationGuide/fhir.TWCRtest (accessed Jun. 14, 2023).
[39] K. F. Chen et al., “Development and validation of a parsimonious and pragmatic CHARM score to predict mortality in patients with suspected sepsis,” American Journal of Emergency Medicine, vol. 35, no. 4, pp. 640–646, Apr. 2017, doi: 10.1016/j.ajem.2016.10.075.
[40] “HL7.TERMINOLOGYActEncounterCode - FHIR v4.0.1,” Feb. 23, 2022. https://terminology.hl7.org/3.1.0/ValueSet-v3-ActEncounterCode.html (accessed Mar. 14, 2023).
[41] D. Kreuzberger, N. Kühl, and S. Hirschl, “Machine Learning Operations (MLOps): Overview, Definition, and Architecture”.
[42] M. Herschel et al., Advances in Database Technology - EDBT 2019 22nd International Conference on Extending Database Technology, Lisbon, Portugal, March 26-29, 2019, proceedings. 2019.
[43] J.-F. Hong and Y.-J. Tseng, “Performance vs. Privacy: Evaluating the Performance of Predicting Second Primary Cancer in Lung Cancer Survivors with Privacy-preserving Approaches,” in 2022 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), 2022, pp. 1–4. doi:
10.1109/BHI56158.2022.9926935.
[44] “Observation - FHIR v4.0.1,” Nov. 01, 2019.
http://hl7.org/fhir/R4/observation-examples.html (accessed Jun. 27, 2023).
[45] “StructureDefinition: VitalSigns - FHIR v4.0.1,” Nov. 01, 2019.
https://hl7.org/fhir/R4/vitalsigns.html (accessed Jun. 14, 2023).
[46] A. D. Haimovich et al., “Development and Validation of the Quick COVID-19
Severity Index: A Prognostic Tool for Early Clinical Decompensation,” Ann Emerg Med, vol. 76, no. 4, pp. 442–453, Oct. 2020, doi: 10.1016/j.annemergmed.2020.07.022.
[47] “Diabetes,” Apr. 05, 2023. https://www.who.int/news-room/fact- sheets/detail/diabetes (accessed Jun. 27, 2023).
[48] “Pima Indians Diabetes Database - dataset by data-society | data.world,” 2016. https://data.world/data-society/pima-indians-diabetes-database (accessed Nov. 25, 2022).
[49] S. Larabi-Marie-Sainte, L. Aburahmah, R. Almohaini, and T. Saba, “Current techniques for diabetes prediction: Review and case study,” Applied Sciences (Switzerland), vol. 9, no. 21. MDPI AG, Nov. 01, 2019. doi: 10.3390/app9214604.
[50] Y. Gupta et al., “Risk factors for necrotizing fasciitis and its outcome at A tertiary care centre,” J. Ayub Med. Coll. Abbottabad, vol. 28, no. 4, pp. 680– 682, Oct. 2016.
[51] C.-Y. Wu and C.-H. Chen, “應用機器學習技術分析肢體壞死性軟組織感染風險,” 2022. https://hdl.handle.net/11296/853rff (accessed Mar. 19, 2023).
[52] “Definition of second primary cancer - NCI Dictionary of Cancer Terms - NCI.” https://www.cancer.gov/publications/dictionaries/cancer- terms/def/second-primary-cancer (accessed Jun. 15, 2023).
[53] J. V Aredo et al., “Tobacco Smoking and Risk of Second Primary Lung Cancer,” 2021, doi: 10.1016/j.jtho.2021.02.024.
[54] K. Der Lee et al., “Increased risk for second primary malignancies in women with breast cancer diagnosed at young age: a population-based study in Taiwan,” Cancer Epidemiol Biomarkers Prev, vol. 17, no. 10, pp. 2647–2655, Oct. 2008, doi: 10.1158/1055-9965.EPI-08-0109.
[55] “Worldwide cancer data | World Cancer Research Fund International,” Mar. 23, 2022. https://www.wcrf.org/cancer-trends/worldwide-cancer-data/ (accessed Jun. 15, 2023).
[56] “What Is Lung Cancer? | CDC,” Oct. 25, 2022. https://www.cdc.gov/cancer/lung/basic_info/what-is-lung-cancer.htm (accessed Jun. 15, 2023).
[57] “HL7.FHIR.US.MCODEHome - FHIR v4.0.1,” Mar. 27, 2023. https://build.fhir.org/ig/HL7/fhir-mCODE-ig/index.html (accessed Jun. 15, 2023).
[58] “Gazelle 癌登表單測試 - HackMD,” Oct. 27, 2021. https://hackmd.io/@cancer2fhir/BJSmqeOrK (accessed Jun. 15, 2023).
[59] “GitHub - smart-on-fhir/smart-dev-sandbox: Docker based sandbox for smart apps,” Feb. 23, 2023. https://github.com/smart-on-fhir/smart-dev-sandbox (accessed Jun. 14, 2023).
[60] R. G. Hauser, D. B. Quine, A. Ryder, and S. Campbell, “Unit conversions between LOINC codes,” J Am Med Inform Assoc, vol. 25, no. 2, p. 192, Feb. 2018, doi: 10.1093/JAMIA/OCX056.
[61] R. M. Al-Tam et al., “A Hybrid Workflow of Residual Convolutional Transformer Encoder for Breast Cancer Classification Using Digital X-ray Mammograms,” Biomedicines 2022, Vol. 10, Page 2971, vol. 10, no. 11, p. 2971, Nov. 2022, doi: 10.3390/BIOMEDICINES10112971.
[62] J. H. Lee, G. W. Lee, G. Bong, H. J. Yoo, and H. K. Kim, “Deep-Learning- Based Detection of Infants with Autism Spectrum Disorder Using Auto- Encoder Feature Representation,” Sensors 2020, Vol. 20, Page 6762, vol. 20, no. 23, p. 6762, Nov. 2020, doi: 10.3390/S20236762.

指導教授

許智誠曾意儒(Chih-Cheng Hsu Yi-Ju Tseng)

審核日期

2023-7-25

推文