Integrating Structured and Unstructured Data with BERTopic and Machine Learning: A Comprehensive Predictive Model for Mortality in ICU Heart Failure Patients
Source abstract
Abstract
Heart failure remains a leading cause of mortality worldwide, particularly within Intensive Care Unit (ICU)-patient populations. This study introduces an innovative approach to predicting ICU mortality by seamlessly integrating electronic health record (EHR) data with a BERTopic-based hybrid machine-learning methodology. The MIMIC-III database serves as the primary data source, encompassing structured and unstructured data from 6606 ICU-admitted heart-failure patients. Unstructured data are processed using BERTopic, complemented by machine-learning algorithms for prediction and performance evaluation. The results indicate that the inclusion of unstructured data significantly enhances the model’s predictive accuracy regarding patient mortality. The amalgamation of structured and unstructured data effectively identifies key variables, enhancing the precision of the predictive model. The developed model demonstrates potential in improving healthcare decision-making, elevating patient outcomes, and optimizing resource allocation within the ICU setting. The handling and application of unstructured data emphasize the utilization of clinical narrative records by healthcare professionals, elevating this research beyond the traditional structured data predictive tools. This study contributes to the ongoing discourse in critical care and predictive modeling, offering valuable insights into the potential of integrating unstructured data into healthcare analytics.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Machine Learning in Healthcare · Sepsis Diagnosis and Treatment
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Business Management, National Taipei University of Technology, Taipei 106, Taiwan
College of Management, National Taipei University of Technology, Taipei 106, Taiwan
References (36)
Show source-provided references
1. Derivation and validation of a machine learning-based risk prediction model in patients with acute heart failure10.1016/j.jjcc.2023.02.006 ↗
2. Predicting Mortality in Intensive Care Unit Patients with Heart Failure Using an Interpretable Machine Learning Model: Retrospective Cohort Study10.2196/38082 ↗
3. Mpanya, D., Celik, T., Klug, E., and Ntsinjana, H. (2023). Predicting in-hospital all-cause mortality in heart failure using machine learning. Front. Cardiovasc. Med., 9.10.3389/fcvm.2022.1032524 ↗
4. Predicting short and long-term mortality after acute ischemic stroke using EHR10.1016/j.jns.2021.117560 ↗
5. Heart Failure Diagnosis, Readmission, and Mortality Prediction Using Machine Learning and Artificial Intelligence Models10.1007/s40471-020-00259-w ↗
6. Kedia, S., and Bhushan, M. (2022, January 24–25). Prediction of mortality from heart failure using machine learning. Proceedings of the 2022 2nd International Conference on Emerging Frontiers in Electrical and Electronic Technologies (ICEFEET), Patna, India.10.1109/ICEFEET51821.2022.9848348 ↗
7. Chiu, C.-C., Wu, C.-M., Chien, T.-N., Kao, L.-J., Li, C., and Jiang, H.-L. (2022). Applying an improved stacking ensemble model to predict the mortality of ICU patients with heart failure. J. Clin. Med., 11.10.3390/jcm11216460 ↗
8. What is an intensive care unit? A report of the task force of the World Federation of Societies of Intensive and Critical Care Medicine10.1016/j.jcrc.2016.07.015 ↗
9. Romano, M. (2019). S The Role of Palliative Care in the Cardiac Intensive Care Unit. Healthcare, 7.10.3390/healthcare7010030 ↗
10. Characteristics, interventions, and longer term outcomes of COVID-19 ICU patients in Denmark—A nationwide, observational study10.1111/aas.13701 ↗
11. Kim, J.Y., Yee, J., Park, T.I., Shin, S.Y., Ha, M.H., and Gwak, H.S. (2021). Risk Scoring System of Mortality and Prediction Model of Hospital Stay for Critically Ill Patients Receiving Parenteral Nutrition. Healthcare, 9.10.3390/healthcare9070853 ↗
12. Zhang, D.D., Yin, C.C., Zeng, J.C., Yuan, X.H., and Zhang, P. (2020). Combining structured and unstructured data for predictive models: A deep learning approach. BMC Med. Inform. Decis. Mak., 20.10.1186/s12911-020-01297-6 ↗
13. Natural Language Processing Accurately Differentiates Cancer Symptom Information in Electronic Health Record Narratives10.1200/CCI.23.00235 ↗
14. Noaeen, M., Amini, S., Bhasker, S., Ghezelsefli, Z., Ahmed, A., Jafarinezhad, O., and Abad, Z.S.H. (2023, January 24–27). Unlocking the power of EHRs: Harnessing unstructured data for machine learning-based outcome predictions. Proceedings of the 2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Sydney, Australia.10.1109/EMBC40787.2023.10340232 ↗
15. Machine learning in clinical decision making10.1016/j.medj.2021.04.006 ↗
16. Assessing Electronic Health Record (EHR) Use during a Major EHR Transition: An Innovative Mixed Methods Approach10.1007/s11606-023-08318-w ↗
17. Egger, R., and Yu, J.N. (2022). A Topic Modeling Comparison between LDA, NMF, Top2Vec, and BERTopic to Demystify Twitter Posts. Front. Sociol., 7.10.3389/fsoc.2022.886498 ↗
18. Improving Arabic Cognitive Distortion Classification in Twitter using BERTopic
19. Latent Dirichlet allocation
20. An enhanced guided LDA model augmented with BERT based semantic strength for aspect term extraction in sentiment analysis10.1016/j.knosys.2022.108668 ↗
21. MIMIC-III, a freely accessible critical care database10.1038/sdata.2016.35 ↗
22. Association between base excess and mortality in patients with congestive heart failure10.1002/ehf2.12939 ↗
23. Blagus, R., and Lusa, L. (2015). Joint use of over-and under-sampling techniques and cross-validation for the development and assessment of prediction models. BMC Bioinform., 16.10.1186/s12859-015-0784-9 ↗
24. Devlin, J., Chang, M.W., Lee, K., Toutanova, K., and Assoc Computat, L. (2019, January 2–7). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the Conference of the North-American-Chapter of the Association-for-Computational-Linguistics—Human Language Technologies (NAACL-HLT), Minneapolis, MN, USA.
25. Pushpakumar, R., Prabu, R., Priscilla, M., Renisha, P., Prabu, R.T., and Muthuraman, U. (2022, January 22–24). A Novel Approach to Identify Dynamic Deficiency in Cell using Gaussian NB Classifier. Proceedings of the 2022 7th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India.10.1109/ICCES54183.2022.9835813 ↗
26. Accuracy/diversity and ensemble MLP classifier design10.1109/TNN.2006.875979 ↗
27. Multi-class adaboost10.4310/SII.2009.v2.n3.a8 ↗
28. Bagging predictors10.1007/BF00058655 ↗
29. Stochastic gradient boosting10.1016/S0167-9473(01)00065-2 ↗
30. Lightgbm: A highly efficient gradient boosting decision tree
31. Chen, T., and Guestrin, C. (2016, January 13–17). Xgboost: A scalable tree boosting system. Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.10.1145/2939672.2939785 ↗
32. Abuzayed, A., and Al-Khalifa, H. (2021, January 4–5). BERT for Arabic Topic Modeling: An Experimental Study on BERTopic Technique. Proceedings of the 5th Conference on AI in Computational Linguistics (ACLing), Electr Network, Online.
33. Rating analysis and BERTopic modeling of consumer versus regulated mHealth app reviews in Germany10.1038/s41746-023-00862-3 ↗
34. ISeeU2: Visually interpretable mortality prediction inside the ICU using deep learning and free-text medical notes10.1016/j.eswa.2022.117190 ↗
35. The prediction of in-hospital mortality in chronic kidney disease patients with coronary artery disease using machine learning models10.1186/s40001-023-00995-x ↗
36. Machine learning to predict in-hospital mortality risk among heterogenous STEMI patients with diabetes10.1093/eurheartj/ehab849.176 ↗