Artificial intelligence in orthopaedic trauma surgery The hype & the hips

The use of artificial intelligence (AI) methods has made the healthcare sector curious about improving decision support tools to aid decision-making using data-driven methods, in all subsequent steps of patient care. The hype around AI triggers hope, with the rise of available electronic health record data to build large datasets, and the increase in computational power. However, many AI applications have not yet made it to clinical practice. Therefore, questions arise: ‘Why do so many promising applications have yet to be adopted in patients’ clinical care or doctors’ workflow?’ and ‘Maybe AI methods do not always lead to better outcomes than traditional methods?’. The thesis will start with an introduction to AI in healthcare, and the rise of AI applications in Orthopaedic surgery.
Subsequently, the aim of this thesis was to evaluate the hype of machine learning (ML) applications in Orthopaedic surgery, develop and (internally and externally) validate decision support tools specific to the hip fracture population (mortality prediction and postoperative delirium prediction), and propose a way forward for enabling integration of these tools into the clinical workflow.