Citation:

Vibhor Pal, 2022. "Foundation Models for Multi-Modal Clinical Decision Support Systems", ESP Journal of Engineering & Technology Advancements 2(2): 183-191.

Abstract:

Decision support systems (DSS) are computer programs founded on artificial intelligence (AI) techniques that help in arriving at the right conclusion within a usually closed area of concern. One such DSS is called a clinical decision support system (CDSS) that can be utilized by clinicians in clinics and hospitals. This paper explores the effectiveness of different machine learning (ML) and deep learning (DL) models on Multi-Mode Clinical Decision Support Systems (CDSS) on the MIMIC-IV data set. Conventional approaches, such as the Random Forest (RF), demonstrated an accuracy of 77% whereas DL algorithms, such as LSTM or TieNet, demonstrated 82.01 and 84.8% rates, respectively. A hybrid XGBoost + Decision Tree model was suggested to improve the predictive performance, and the maximum accuracy was 92%. High precision (96%), recall (95%), and F1-score (95%) also showed that the model is robust and reliable to help make clinical decisions. These findings highlight the usefulness of hybrid ensemble methods that can be used to take advantage of multi-modal clinical data to make accurate and interpretable predictions in CDSS.

References:

[1] J. Bajwa, U. Munir, A. Nori, and B. Williams, “Artificial intelligence in healthcare: transforming the practice of medicine,” Futur. Healthc. J., vol. 8, no. 2, Jul. 2021, doi: 10.7861/fhj.2021-0095.

[2] A. Qayyum, J. Qadir, M. Bilal, and A. Al-Fuqaha, “Secure and Robust Machine Learning for Healthcare: A Survey,” IEEE Rev. Biomed. Eng., vol. 14, pp. 156–180, 2021, doi: 10.1109/RBME.2020.3013489.

[3] T. K. Mackey et al., “‘Fit-for-purpose?’ – challenges and opportunities for applications of blockchain technology in the future of healthcare,” BMC Med., vol. 17, no. 1, p. 68, Dec. 2019, doi: 10.1186/s12916-019-1296-7.

[4] A. Papa, M. Mital, P. Pisano, and M. Del Giudice, “E-health and wellbeing monitoring using smart healthcare devices: An empirical investigation,” Technol. Forecast. Soc. Change, vol. 153, p. 119226, Apr. 2020, doi: 10.1016/j.techfore.2018.02.018.

[5] M. Carlile, B. Hurt, A. Hsiao, M. Hogarth, C. A. Longhurst, and C. Dameff, “Deployment of artificial intelligence for radiographic diagnosis of COVID-19 pneumonia in the emergency department,” JACEP Open, 2020, doi: 10.1002/emp2.12297.

[6] X. Wang, Y. Peng, L. Lu, Z. Lu, and R. M. Summers, “TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-Rays,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2018. doi: 10.1109/CVPR.2018.00943.

[7] B. Jing, P. Xie, and E. P. Xing, “On the automatic generation of medical imaging reports,” in ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), 2018. doi: 10.18653/v1/p18-1240.

[8] A. Brady, R. Ó. Laoide, P. McCarthy, and R. McDermott, “Discrepancy and error in radiology: Concepts, causes and consequences,” 2012.

[9] M. Ranganathan et al., “Attenuation of ketamine-induced impairment in verbal learning and memory in healthy volunteers by the AMPA receptor potentiator PF-04958242,” Mol. Psychiatry, vol. 22, no. 11, pp. 1633–1640, Nov. 2017, doi: 10.1038/mp.2017.6.

[10] S. Gössling, “Risks, resilience, and pathways to sustainable aviation: A COVID-19 perspective,” J. Air Transp. Manag., vol. 89, p. 101933, Oct. 2020, doi: 10.1016/j.jairtraman.2020.101933.

[11] R. Rybnicek, J. Plakolm, and L. Baumgartner, “Risks in Public–Private Partnerships: A Systematic Literature Review of Risk Factors, Their Impact and Risk Mitigation Strategies,” Public Perform. Manag. Rev., 2020, doi: 10.1080/15309576.2020.1741406.

[12] S. V. B. Jardim, “The Electronic Health Record and its Contribution to Healthcare Information Systems Interoperability,” Procedia Technol., vol. 9, pp. 940–948, 2013, doi: 10.1016/j.protcy.2013.12.105.

[13] P. Siciarz, S. Alfaifi, E. Van Uytven, S. Rathod, R. Koul, and B. McCurdy, “Machine learning for dose-volume histogram based clinical decision-making support system in radiation therapy plans for brain tumors,” Clin. Transl. Radiat. Oncol., vol. 31, pp. 50–57, Nov. 2021, doi: 10.1016/j.ctro.2021.09.001.

[14] M. Qjidaa et al., “Development of a clinical decision support system for the early detection of COVID-19 using deep learning based on chest radiographic images,” in 2020 International Conference on Intelligent Systems and Computer Vision, ISCV 2020, 2020. doi: 10.1109/ISCV49265.2020.9204282.

[15] H. El Hamdaoui, S. Boujraf, N. E. H. Chaoui, and M. Maaroufi, “A Clinical support system for Prediction of Heart Disease using Machine Learning Techniques,” in 2020 International Conference on Advanced Technologies for Signal and Image Processing, ATSIP 2020, 2020. doi: 10.1109/ATSIP49331.2020.9231760.

[16] A. Yahyaoui, A. Jamil, J. Rasheed, and M. Yesiltepe, “A Decision Support System for Diabetes Prediction Using Machine Learning and Deep Learning Techniques,” in 2019 1st International Informatics and Software Engineering Conference (UBMYK), IEEE, Nov. 2019, pp. 1–4. doi: 10.1109/UBMYK48245.2019.8965556.

[17] B. Andrianto, Y. K. Suprapto, I. Pratomo, and I. Irawati, “Clinical decision support system for typhoid fever disease using classification techniques,” in Proceedings - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019, 2019. doi: 10.1109/ISITIA.2019.8937286.

[18] W. Boag, T. M. Harry Hsu, M. McDermott, G. Berner, E. Alsentzer, and P. Szolovits, “Baselines for Chest X-Ray Report Generation,” Proc. Mach. Learn. Res., vol. 116, pp. 126–140, 2019.

[19] S. P. Tembhurne, “Clinical Decision Making Using Machine Learning and ICU Data,” Helix, vol. 8, no. 5, pp. 4082–4087, 2018, doi: 10.29042/2018-4082-4087.

[20] Z. Babar, T. van Laarhoven, and E. Marchiori, “Encoder-decoder models for chest X-ray report generation perform no better than unconditioned baselines,” PLoS One, 2021, doi: 10.1371/journal.pone.0259639.

Keywords:

Automated Machine Learning (Automl), Deep Learning, Intensive Care Unit, Mortality, Multimodal, Feature Fusion.