Using AI to Predict Hypotension during Surgery

Purpose and goal:

The objective of this project is to develop and validate a warning system for harmful blood pressure drops during surgery with the aim of creating patient-centered solutions that reduce complications, care time and risk of death after surgery. The end product is an anesthesia machine with support for our capnodynamic method connected to other monitoring equipment and equipped with AI-based algorithms that warn the healthcare staff about an imminent risk of a drop in blood pressure in the patient.

Expected results and effects:

Every year, approximately 800,000 patients are operated on in Sweden, of which a significant proportion are elderly undergoing extensive surgery. In this project, we will use artificial intelligence and a new advanced patient monitoring developed by us to reduce the number of complications after extensive surgery for that patient group. An AI-based warning system enables circulatory supportive treatments to be used preventively and thereby avoid serious complications.

Approach and implementation:

The project partners, Getinge, Karolinska and KTH, all contribute jointly to the project. The project is divided into three work packages, where we in the first work with business plans and market analysis, including analysis of customer needs and quantifications of customer value. In the second, we work with collecting data from surgeries and patients from Karolinska´s operating wards where we use equipment from Getinge that supports our Capnodynamic method.

The third work package involves development, preprocessing, training and validation of the AI models.