Patients who have undergone surgery for colorectal cancer experience fewer serious complications and fewer readmissions when an AI tool supports treatment decisions, according to new research. The tool has the potential to save the healthcare system significant costs.
Readmissions after surgery are demanding for patients and costly for society. A new study now shows that artificial intelligence (AI) can contribute to better courses of treatment for colorectal cancer patients - both before and after surgery. Ultimately, this could mean lower treatment costs.
A research team from the Center for Surgical Science at Zealand University Hospital and the University of Copenhagen has just published a study in the prestigious journal Nature Medicine, demonstrating how AI can be used to support the treatment of patients who have undergone colorectal cancer surgery.
The AI can, for example, help plan the patient's preoperative training and provide input on post-operative treatments and the need for home visits.
And the results are promising:
- The study showed that 32-36 % fewer patients experienced complications after colorectal cancer surgery when the AI tool was used to support treatment decisions.
- The number of readmissions also dropped by 35 %.
- Results suggest that the reduction was greatest among the most seriously ill patients.
"Complications after surgery are often a major issue for patients undergoing colorectal cancer operations. We are therefore very pleased to see how the AI tool has reduced both the number of complicated cases and the number of readmissions - especially because the trend shows that it is the most vulnerable patients that are doing better," says Ismail Gögenur, Clinical Professor at the Department of Clinical Medicine at the University of Copenhagen and Senior Consultant Zealand University Hospital.
Can potentially free up DKK 18,000 per patient
"The study shows how we can use Danish data and digital tools to improve treatment and make it more personalized. It can be applied to other types of cancer surgeries in the future. This is an important focus - especially since it is combined with clinical trials and thus reaches the patients who need it most," says Ismail Gögenur.
The results confirm previous findings, and the research team will now explore how AI as a decision support tool can tailor treatment in other cancer areas, such as lung, kidney, bladder, and ovarian cancer.
In addition to benefiting patients, the research project reveals another promising trend. The new tool also appears to save money.
"We actually see that using the AI tool to plan treatment is associated with a reduced need for healthcare services, which could free up approximately DKK 18,000 per patient for the healthcare system," says Ismail Gögenur.
Read the study "Clinical implementation of an AI-based prediction model for decision support for patients undergoing colorectal cancer surgery".
About the study
Researchers have developed an AI tool that can predict one-year survival for patients who have undergone colorectal cancer surgery.
The tool is based on data from Danish health registers and was tested on 1,000 patients from Zealand University Hospital.
Patients were divided into risk groups and received tailored treatment plans focusing on for example physical training and nutrition - drawing inspiration from the international PREHAB project.
The research is a collaboration between the Center for Surgical Science at Zealand University Hospital and the Department of Clinical Medicine at the University of Copenhagen.
