January 2025 - Lorenz Kapral

English subtitles on Youtube available: click on the gear icon and select “Untertitel” – “English”

Dipl.-Ing. Lorenz Kapral

MedUni Wien RESEARCHER OF THE MONTH January 2025

Intraoperative hypotension is linked to postoperative complications and should be prevented. We developed a temporal fusion transformer (TFT) model to forecast mean arterial pressure (MAP) 7 minutes ahead using only low-resolution data (every 15 s) on patient demographics, vital signs, medications, and ventilation. Trained on 73,009 patients and tested on internal (n=8113) and external (n=5065) cohorts, the model achieved a mean absolute MAP prediction error of 4 mmHg internally and 7 mmHg externally. For binary prediction of hypotension (MAP <65 mmHg), the model’s discrimination was excellent, achieving AUROCs of 0.933 internally and 0.919 externally.
 

Selected Literature

1. Kapral, L., Dibiasi, C., Jeremic, N., Bartos, S., Behrens, S., Bilir, A., Heitzinger, C., & Kimberger, O. (2024). Development and external validation of temporal fusion transformer models for continuous intraoperative blood pressure forecasting. eClinicalMedicine. https://doi.org/10.1016/j.eclinm.2024.102797

2. Dibiasi, C., Agibetov, A., Kapral, L., Zeiner, S., & Kimberger, O. (2023). Predicting intraoperative hypothermia burden during non-cardiac surgery: A retrospective study comparing regression to six machine learning algorithms. Journal of Clinical Medicine, 12(13), 4434. https://doi.org/10.3390/jcm12134434

3. Bologheanu, R., Kapral, L., Laxar, D., Maleczek, M., Dibiasi, C., Zeiner, S., Agibetov, A., Ercole, A., Thoral, P., Elbers, P., et al. (2023). Development of a reinforcement learning algorithm to optimize corticosteroid therapy in critically ill patients with sepsis. Journal of Clinical Medicine, 12(4), 1513. https://doi.org/10.3390/jcm12041513

4. Maleczek, M., Laxar, D., Kapral, L., Kuhrn, M., Abulesz, Y.-T., Dibiasi, C., & Kimberger, O. (2024). A comparison of five algorithmic methods and machine learning pattern recognition for artifact detection in electronic records of five different vital signs: A retrospective analysis. Anesthesiology. https://doi.org/10.1097/aln.0000000000004971

5. Hriberšek, M., Eibensteiner, F., Kapral, L., Teufel, A., Nawaz, F. A., Cenanovic, M., Siva Sai, C., Devkota, H. P., De, R., Singla, R. K., et al. (2023). "Loved ones are not ‘visitors' in a patient's life"—The importance of including loved ones in the patient’s hospital stay: An international Twitter study of #HospitalsTalkToLovedOnes in times of COVID-19. Frontiers in Public Health, 11, 1100280. https://doi.org/10.3389/fpubh.2023.1100280

6. Lintschinger, J. M., Laxar, D., Kapral, L., Ulbing, S., Glock, T., Behrens, S., Frimmel, C., Renner, R., Klaus, D. A., Willschke, H., et al. (2024). A retrospective analysis of the need for on-site emergency physician presence and mission characteristics of a rural ground-based emergency medical service. BMC Emergency Medicine. https://doi.org/10.1186/s12873-024-01062-2

7. Kapral, L., Zawisky, M., & Abele, H. (2020). Neutron radiography and tomography of the drying process of screed samples. Journal of Imaging, 6(11), 118. https://doi.org/10.3390/jimaging6110118