International Journal of
Advanced Multidisciplinary Application

Volume 1, Issue 4

• Author(s)Joao dos Santos, S. Miguel Francisco, N. Teresa Nascimento
• AffiliationUniversity of Agostinho Neto, Angola
• Page No.15-26
• Volume, Issue & YearVolume 1, Issue 4, Dec 2024
• Published On2024/12/30
• JournalInternational Journal of Advanced Multidisciplinary Application (IJAMA)
• ISSN No.3048-9350

Abstract

The growing use of electronic medical records (EMRs) offers a promising opportunity to improve trauma care through data-driven insights. However, extracting useful and actionable information from unstructured clinical text remains a significant challenge. This study addresses this issue by applying natural language processing (NLP) techniques to extract injury-related variables and classify trauma patients based on the presence of loss of consciousness (LOC). A dataset of 23,308 trauma patient EMRs, comprising both
pre-diagnosis and post-diagnosis free-text notes, was analyzed using a bilingual (English and Korean) pretrained RoBERTa model. The patients were grouped into four categories based on LOC and head trauma. To mitigate class imbalance in LOC labeling, deep learning models were trained with weighted loss functions, achieving an area under the curve (AUC) of 0.91. Local Interpretable Model-agnostic Explanations (LIME) analysis further highlighted the model's ability to identify key terms associated with head injuries and
consciousness. The results suggest that NLP can accurately identify LOC in trauma patients EMRs, with the weighted loss functions effectively addressing class imbalances. These findings pave the way for developing AI tools that can enhance trauma care and clinical decision-making

Keywords: natural language processing, text mining, deep learning, emergency departments, clinical decision support

References

1. [1] A. Rajkomar et al., “Scalable and accurate deep learning for electronic health records,” JAMA, vol. 322, no. 19, pp. 1837–1845, 2019.
2. [2] A. E. Johnson et al., “MIMIC-III, a freely accessible critical care database,” Scientific Data, vol. 3, p. 160035, 2016.
3. [3] S. Vashishth et al., “Natural language processing for electronic health records: A review,” Journal of Healthcare Engineering, vol. 2021, 2021.
4. [4] Y. Wang et al., “A comprehensive review of deep learning for healthcare applications,” IEEE Access, vol. 8, pp. 17310–17327, 2020.
5. [5] Y. Zhang et al., “Deep learning for healthcare applications,” International Journal of Medical Informatics, vol. 114, pp. 96–108, 2018.
6. [6] S. Wang et al., “The state of the art in electronic health record-based machine learning applications,” Journal of Healthcare Informatics Research, vol. 3, no. 2, pp. 121–141, 2019.
7. [7] S. Liang et al., “Automated detection of loss of consciousness using electronic health records: A deep learning approach,” Journal of Medical Internet Research, vol. 22, no. 10, p. e17956, 2020.
8. [8] A. Rajkomar et al., “Artificial intelligence in health care: Anticipating challenges to ethics, privacy, and governance,” JAMA, vol. 323, no. 11, pp. 1052–1053, 2020.
9. [9] L. Sweeney et al., “Natural language processing for health data: A review of challenges and opportunities,” Journal of Biomedical Informatics, vol. 92, p. 103158, 2019.
10. [10] J. Zeng et al., “Impact of natural language processing in medical document classification: A systematic review,” Artificial Intelligence in Medicine, vol. 114, p. 101024, 2021.
11. [11] Y. Kim et al., “Text mining for medical research applications,” Journal of Healthcare Engineering, vol. 2018, p. 4301046, 2018.
12. [12] F. Wang et al., “Data mining techniques for identifying medical conditions in electronic health records,” Journal of Clinical Informatics, vol. 5, no. 3, pp. 255–270, 2020.
13. [13] K. Cho et al., “Natural language processing-based prediction of head trauma in emergency department patients,” Journal of Medical Systems, vol. 44, pp. 1–9, 2020.
14. [14] S. Liu et al., “Deep learning-based classification of traumatic brain injury using medical records,” Frontiers in Neuroscience, vol. 13, p. 1020, 2019.
15. [15] L. Zhang et al., “Application of natural language processing in medical text mining,” Journal of Health Informatics, vol. 29, no. 2, pp. 89–94, 2019.
16. [16] Y. Hu et al., “Deep learning in the diagnosis of traumatic brain injury: A review,” Neuroinformatics, vol. 15, pp. 253–265, 2017.
17. [17] Z. Zhi et al., “Data-driven methods for medical document classification using NLP,” Healthcare Informatics Research, vol. 25, no. 4, pp. 263–272, 2019.
18. [18] S. Yadav et al., “Improving health outcomes with natural language processing,” Journal of Artificial Intelligence in Medicine, vol. 110, pp. 1–12, 2021.
19. [19] M. Chen et al., “Machine learning techniques in healthcare,” IEEE Transactions on Biomedical Engineering, vol. 65, no. 8, pp. 1829–1839, 2018.
20. [20] S. Prakash et al., “Applications of deep learning for medical data classification,” Journal of Digital Health, vol. 2, pp. 1–12, 2020.
21. [21] A. Smith et al., “Clinical decision support with artificial intelligence in emergency medicine,” Journal of Clinical Medicine, vol. 9, no. 4, p. 1278, 2020.
22. [22] E. Choi et al., “Learning to predict clinical events from electronic health records using deep learning,” Journal of Machine Learning Research, vol. 17, pp. 1–19, 2016.
23. [23] M. Buda et al., “A review of machine learning algorithms in clinical decision support systems,” Journal of Medical Systems, vol. 42, no. 6, pp. 1–16, 2018.
24. [24] S. Marafino et al., “Automated machine learning models in clinical care: Impact on patient outcomes,” Journal of Clinical Decision Support, vol. 35, no. 2, pp. 75–82, 2020.
25. [25] Q. Nguyen et al., “A comparative study of NLP-based methods for medical text classification,” BMC Medical Informatics and Decision Making, vol. 21, no. 1, p. 95, 2021.
26. [26] S. Lee et al., “Deep learning applications in medical diagnostics: A review,” Computational Biology and Chemistry, vol. 78, pp. 90–102, 2019.
27. [27] X. Tong et al., “The role of NLP in automated clinical text mining,” Journal of Medical Informatics, vol. 12, no. 6, pp. 245–255, 2018.
28. [28] X. Xu et al., “NLP techniques for clinical text mining: An overview,” Artificial Intelligence in Medicine, vol. 113, p. 102034, 2020.
29. [29] M. Wang et al., “Optimizing NLP for patient data classification in EMRs,” Journal of Healthcare Engineering, vol. 2021, 2021.
30. [30] H. Kim et al., “Natural language processing for medical text analysis in emergency medicine,” Journal of Emergency Medicine, vol. 58, no. 3, pp. 341–347, 2020.
31. [31] Z. Liu et al., “Trauma care support system based on NLP,” IEEE Access, vol. 8, pp. 102234–102242, 2020.
32. [32] S. Gehrmann et al., “RoBERTa: A robustly optimized BERT pretraining approach,” arXiv:1907.11692, 2018.
33. [33] A. Vaswani et al., “Attention is all you need,” in Proc. NeurIPS, vol. 30, pp. 5998–6008, 2017.
34. [34] H. Shin et al., “Deep learning for medical image analysis: A survey,” IEEE Transactions on Medical Imaging, vol. 39, no. 4, pp. 1004–1025, 2020.
35. [35] T. Brown et al., “Language models are few-shot learners,” arXiv:2005.14165, 2020.
36. [36] Y. Choi et al., “Predictive modeling in healthcare using natural language processing,” Journal of Healthcare Data Analytics, vol. 4, no. 2, pp. 125–139, 2019.
37. [37] B. Yu et al., “Application of deep learning to health data: A survey,” Journal of Artificial Intelligence Research, vol. 64, pp. 1–15, 2018.
38. [38] W. Zhang et al., “NLP in health information systems: A systematic review,” International Journal of Medical Informatics, vol. 134, p. 104079, 2020.
39. [39] A. Sarker et al., “Using NLP for medical event extraction,” in Proc. ACL, pp. 257–268, 2020.
40. [40] R. Sharma et al., “Natural language processing for health risk assessments in electronic health records,” Journal of Healthcare Engineering, vol. 2020, 2020.
41. [41] A. Rios et al., “Integrating NLP for improving patient outcomes in trauma care,” Journal of Clinical Data Science, vol. 2, no. 3, pp. 159–172, 2021.
42. [42] L. Zhang et al., “Trauma care decision support using NLP and deep learning,” Journal of Trauma and Acute Care Surgery, vol. 91, no. 4, pp. 675–682, 2021.
43. [43] A. Gupta et al., “Predictive analytics in healthcare using machine learning and natural language processing,” Journal of Health Informatics, vol. 13, no. 2, pp. 45–56, 2021.
44. [44] X. Gao et al., “Evaluating the effectiveness of deep learning for clinical text classification in the ED,” Computers in Biology and Medicine, vol. 122, p. 103804, 2020.
45. [45] S. Kim et al., “Exploring NLP-based medical documentation for improving trauma care,” Journal of Clinical AI, vol. 3, no. 1, pp. 12–18, 2021.
46. [46] S. De P et al., “Predictive modeling for trauma using EMRs: An NLP approach,” Journal of Digital Health, vol. 10, pp. 245–255, 2019.
47. [47] S. Tran et al., “Deep learning techniques for automatic medical diagnosis from electronic health records,” Journal of Medical Systems, vol. 42, no. 6, pp. 126–136, 2018.
48. [48] H. Cho et al., “Natural language processing for health informatics: Challenges and opportunities,” Journal of Healthcare Engineering, vol. 2017, 2017.

Explore Our Related Journals

Looking for the right journal for your next manuscript? Explore our international peer-reviewed journals covering engineering, management, computer science, artificial intelligence and multidisciplinary research.