Sepsis Prevention in Surgery: Informing the Clinical Process through Artificial Intelligence


Research is needed to better predict who might be at most risk of severe infections after surgery. This research aims to develop a tool that can predict the likelihood of infection and subsequent sepsis in surgery patients, and support clinical decision making.


Microbes, such as bacteria, viruses, and fungi, can cause a life-threatening infectious condition known as sepsis. In Europe, sepsis causes death in around 38% of patients who develop the condition. Patients with sepsis rapidly deteriorate within a few hours and need intensive treatment.  Sepsis can be particularly dangerous after surgery.  Even though operations are carried out in sterile conditions, some people develop infections. We are still not good at predicting who will be most affected and how we can prevent these infections.


Protecting individuals from getting an infection is the best solution. This is especially important for individuals with underlying risk factors for infection.

It’s possible to predict the likelihood of contracting a particular disease based on the individual’s particular risk factors.

Artificial Intelligence (AI) uses computer systems to perform complex tasks that would normally require human intelligence. Using AI tools, we can predict this likelihood, which in turn can guide clinicians’ decision-making and thereby help put strategies in place to protect patients from developing sepsis.

This project will involve the development of an AI tool that can guide clinicians in predicting the likelihood of infection and subsequent sepsis in patients undergoing surgery to remove their gallbladder, in particular. This tool will be prepared using pseudonymised data from patient records. For the tool to be implemented in routine clinical care, it will first need to be developed, then validated and tested.


The researchers have worked with patients who have had surgery and members of the public.  This included discussions with members of the public using VOICE (, which is a large organisation from patients and members of the public from UK and internationally, that provides insights and feedback on clinical research through different activities.


This project was supported unanimously by the PIONEER Data Trust Committee after requesting more public and patient involvement in the project.

The study is being run by Dr Nehal Hassan, The University of Newcastle.

2 Responses

  1. As a sepsis Survivor this is wonderful news. Do you think, at this stage, that there could be a wider use for this tool say with cancer patients who are at high risk of developing an infection?
    Regards David

  2. Dear David,

    Many thanks for your thoughtful comment and interest in PIONEER projects.

    It might be that this tool could be used to help different patient groups such as those with cancer who are at risk of infections and sepsis, but this will require further testing in these specific groups, to make sure the tools would be accurate and helpful. We have passed your comment on to the study team, and they are grateful for this suggestion.

    Kind regards

    Ben Crosby

Leave a Reply

Your email address will not be published. Required fields are marked *