Prediction of Complications After Major Gastrointestinal Surgery With Machine Learning and Point of Care Ultrasound

The primary aim of this study is to develop a machine learning framework to predict major complications after major gastro-intestinal surgery. Secondary aims include combining this framework with point of care ultrasound to determine the best initial resuscitative strategy; and to determine which ultrasound parameters are best predictive of fluid intolerance. Furthermore if the renin angiotensin aldosterone system is more active after liver resection. Study design: Single centre observational cohort study Study population: Adult patients undergoing elective major gastrointestinal surgery Primary study parameters/outcome of the study: The main study endpoint is a machine learning framework based on the hemodynamic profile to predict major complications,especially cardiovascular/pulmonary instability, including, sepsis and septic shock. Data from the ClearSight will be used to collect non-invasive arterial pressure waveforms. point of care ultrasound of heart, lungs and abdominal veins, and clinical data from the electronic medical record will be collected Secondary study parameters/outcome of the study (if applicable): point of care ultrasound of heart, lungs and abdominal veins, and clinical data from the electronic medical record will be collected. Ina subgroup of 40 patients RAAS levels and portal blood samples will be analysed.