A new study published in Nature Medicine has found that using an AI tool to support treatment decisions significantly reduces complications and readmissions in patients who have undergone colorectal cancer surgery. The tool, developed by researchers from the Center for Surgical Science at Zealand University Hospital and the University of Copenhagen, analyzes patient data to predict one-year survival and guide personalized care plans. The AI system uses information from Danish health registers and was tested on 1,000 patients from Zealand University Hospital. It categorizes patients into risk groups, enabling tailored interventions such as preoperative physical training, nutritional support, and postoperative care planning. These recommendations are inspired by the international PREHAB model, which focuses on improving patient resilience before surgery. Results show that patients supported by the AI experienced fewer serious complications and lower rates of hospital readmission—particularly among the most vulnerable individuals. This improvement suggests the tool is especially effective in high-risk cases, where outcomes are often harder to manage. Beyond clinical benefits, the study highlights significant cost savings. Researchers estimate that using the AI tool to guide treatment could reduce healthcare service needs by about DKK 18,000 per patient, freeing up resources for the health system. Ismail Gögenur, Clinical Professor at the University of Copenhagen and Senior Consultant at Zealand University Hospital, emphasized the broader implications. He noted that the findings demonstrate how Danish health data and digital tools can be used to deliver more personalized, effective care. He also pointed out that the approach could be adapted for other types of cancer surgeries, including those for lung, kidney, bladder, and ovarian cancers. The research team plans to expand the use of AI decision support across additional cancer types, with a focus on integrating clinical trials to ensure patients receive the most beneficial care. The study marks a significant step forward in using artificial intelligence not just to predict outcomes, but to actively improve patient recovery and reduce the burden on healthcare systems.