Various studies use numerous probabilistic methods to establish a cause-effect relationship between a drug and a disease. However, only a limited number of machine learning studies on establishing cause-effect relationships can be found on the internet. In this study, we explore machine learning approaches for interpreting large quantities of multivariate patient-based laboratory data for establishing cause-effect relationships for critically ill patients. We adopt principal component analysis as a primary method to capture daily patient changes after a medical intervention so that the causal relationship between the medical treatments and the outcomes can be established. Model validity and stability are evaluated using bootstrap testing. The model exhibits an acceptable significance level with a two-tailed test. Moreover, results show that the approach provides promising results in interpreting large quantities of patient data and establishing cause-effect relationships for making informed decisions for critically ill patients. If fused with other machine learning and probabilistic models, the proposed approach can provide the healthcare industry with an added tool for daily routine clinical practices. Furthermore, the approach will be able to support clinical decision-making and enable effective patient-tailored care for better health outcomes.

[1] Angus D., “Fusing Randomized Trials with Big Data: The Key to Self-Learning Health Care Systems?,” Journal of American Medical. Association, vol. 314, no. 8, pp. 767-768, 2015. doi: 10.1001/jama.2015.7762.

[2] Dai Z., Liu S., Wu J., Li M., Liu J., and Li K., “Analysis of Adult Disease Characteristics and Mortality on MIMIC-III,” Journal of PLoS One, vol. 15, no. 4, 2020. https://doi.org/10.1371/journal.pone.0232176

[3] Elbasi E. and Zreikat A., “Heart Disease Classification for Early Diagnosis based on Adaptive Hoeffding Tree Algorithm in IoMT Data,” The International Arab Journal of Information Technology, vol. 20, no. 1, pp. 38- 48, 2023. DOI:10.34028/iajit/20/1/5

[4] Feng M., Mcsparron J., Kien D., Stone D., Roberts D., Schwartzstien R., Baron A., and Celi L., “Transthoracic Echocardiography and Mortality in Sepsis: Analysis of the MIMIC-III Database,” Intensive Care Medicine, vol. 44, no. 6, pp. 884-892, 2018. doi: 10.1007/s00134-018- 5208-7

[5] Goldberger A., Amaral L., Glass L., Hausdorff J., Ivanov P., Mark R., Mietus J., Moody G., Peng C., and Stanley E., “PhysioBank, PhysioToolkit, and PhysioNet: Components of A New Research Resource for Complex Physiologic Signals,” Journal of the American Heart Association, vol. 101, no. 23, 2000. doi: 10.1161/01.cir.101.23.e215.

[6] Johnson E., Pollard T., Shen L., Lehman L., Feng M., Ghassemi M., Moody B., Szolovits P., Celi L., and Mark R., “MIMIC-III, a Freely Accessible Critical Care Database,” Journal of Nature Scientific Data, vol. 3, pp. 1-9, 2016. doi: 10.1038/sdata.2016.35.

[7] Ketchersid T., “Big Data in Nephrology: Friend or Foe?,” Blood Purification, vol. 36, no. 3-4, pp. 160-164, 2014. doi: 10.1159/000356751.

[8] Kleinberg S. and Hripcsak G., “A Review of Causal Inference for Biomedical Informatics,” Journal Biomedical Informatics, vol. 44, no. 6, pp. 1102-1112, 2011. doi: 10.1016/j.jbi.2011.07.001.

[9] Linden A. and Yarnold P., “Combining Machine Learning and Matching Techniques To Improve Causal Inference in Program Evaluation,” Journal of Evaluation in Clinical Practice, vol. 22, no. 6, pp. 864-870, 2016. DOI:10.1111/jep.12592

[10] Linting M., Os B., and Meulman J., “Statistical Significance of the Contribution of Variables to the PCA Solution: An Alternative Permutation Strategy,” Psychometrika, vol. 76, no. 3, pp. 440- 460, 2011. DOI: 10.1007/S11336-011-9216-6

[11] Monleon-Getino A. and Canela-Soler J., “Causality in Medicine and its Relationship with the Role of Statistics,” Biomedical Statistics and Informatics, vol. 2, no. 2, pp. 61-68, 2017. DOI:10.11648/j.bsi.20170202.14)

[12] Pirracchio R. Cohen M., Malenica I., Cohen J., Chambaz A., Cannesson M., Lee C., Resche- Rigon M., and Hubbard A., “Big Data and Targeted Machine Learning in Action To Assist Medical Decision in the ICU,” Anaesthesia Critical Care and Pain Medicine, vol. 38, no. 4, Establishing Cause-Effect Relationships from Medical Treatment Data in Intensive ... 701 pp. 377-384, 2019. doi: 10.1016/j.accpm.2018.09.008.

[13] Rose S. and Rizopoulos D., “Machine Learning for Causal Inference in Biostatistics,” Biostatistics, vol. 21, no. 2, pp. 336-338, 2020. https://doi.org/10.1093/biostatistics/kxz045

[14] Stern A. and Price W., “Regulatory Oversight, Causal Inference, and Safe and Effective Health Care Machine Learning,” Biostatistics, vol. 21, no. 2, pp. 363-367, 2020.

[15] Vincent J., Nielsen N., Shapiro N., Gerbasi M., Grossman A., Doroff R., Zeng F., Young P., and Russell J., “Mean Arterial Pressure and Mortality in Patients with Distributive Shock: A Retrospective Analysis of The MIMIC-III Database,” Ann. Intensive Care, vol. 8, no. 1, 2018. doi: 10.1186/s13613-018-0448-9.