Arina, Pietro;
(2025)
Perioperative Mortality and Morbidity Prediction Through Machine Learning and Artificial Intelligence: Insights from Aerobic Fitness and Physiological Data Analysis.
Doctoral thesis (Ph.D), UCL (University College London).
Preview |
Text
Arina_10209438_Thesis_sig-removed.pdf Download (5MB) | Preview |
Abstract
This thesis investigates the integration of artificial intelligence (AI) and machine learning (ML) techniques into perioperative medicine, a specialty dedicated to improving patient outcomes around the time of surgery. By addressing critical clinical challenges, this work contributes novel methodologies to advance perioperative care. The primary focus of the thesis is the development of innovative ML models using Multi-objective Symbolic Regression (MOSR), a cutting-edge technique. These models were designed to predict outcomes for patients undergoing general surgery, utilizing only preoperative data. The first model identifies patients at risk of severe postoperative morbidity, while the second predicts those with a high risk of mortality within 12 months post-surgery. These predictive capabilities provide valuable tools for early risk stratification and optimization of perioperative management. A significant contribution of this research is the demonstration of the importance of using metabolic and fitness data from cardiopulmonary exercise testing (CPET) to study patients through ML. The novelty lies in the application of ML to CPET data within the predictive models, showcasing how such data can enhance understanding of perioperative patient profiles. Additionally, the thesis highlights the use of new features, particularly metabolic ones, to gain deeper insights into perioperative risks and outcomes. Another key finding is the exploration of metabolic flexibility—the ability of a patient’s metabolism to switch between fat and glucose utilization—as a critical factor in understanding postoperative morbidity risks. This thesis demonstrates the importance of 6 metabolic flexibility as a biomarker, offering new insights into patient risk profiles and informing personalized interventions. Overall, this work underscores the potential of AI/ML in transforming perioperative medicine. By introducing novel predictive models, emphasizing metabolic flexibility, and incorporating CPET-derived features, it establishes a foundation for improved patient outcomes and more targeted, data-driven clinical decisions.
| Type: | Thesis (Doctoral) |
|---|---|
| Qualification: | Ph.D |
| Title: | Perioperative Mortality and Morbidity Prediction Through Machine Learning and Artificial Intelligence: Insights from Aerobic Fitness and Physiological Data Analysis |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2025. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Medical Sciences > Div of Medicine |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10209438 |
Archive Staff Only
 |
View Item |
