top of page
BETA logo.001.png

PROJECT TEAM

Andreas Bjerrum, MD, Department of Oncology, Rigshospitalet

Mads Andersen, MD, Department of Oncology, Rigshospitalet

Charles Vesteghem, Associate Professor, Aalborg University Hospital and Aalborg University

Marianne Bertelsen, Head of Section, Centre for Financial Affairs, Capital Region of Denmark

Anders Riis, Team Leader & Data Analyst, Centre for Financial Affairs, Capital Region of Denmark

Sofie Pødenphandt Jensen, Data Analyst, Centre for Financial Affairs, Capital Region of Denmark

THE NEED

Over 16% of lung cancer patients receive systemic anticancer treatment in the final stage of life, often causing severe side effects such as fatigue, nausea, and diarrhea without therapeutic benefit. Clinicians frequently struggle to predict short-term survival accurately, resulting in counterproductive treatments that reduce quality of life and waste resources.

THE SOLUTION

Our validated AI model predicts 30-day mortality by analyzing clinical data, including comorbidities, lab results, and vital signs. By integrating into the clinical workflow, it provides oncologists with accurate, data-driven insights to help avoid potentially harmful treatments near the end of life, ensuring better patient-centered care and resource allocation.

Reducing Counterproductive Treatments - Rigshospitalet

Call 5 - 2024

500.000 DKK

Clinical Area

Oncology

Technology

Health tech

PROJECT SUMMARY

The project implements an AI model predicting 30-day mortality for lung cancer patients to guide treatment decisions. By identifying patients unlikely to benefit from systemic anticancer therapy near the end of life, the solution enhances quality of life and reduces unnecessary treatments and costs.

CLINICAL IMPACT

The AI model can prevent up to 40% of futile treatments near the end of life, avoiding unnecessary side effects, improving end-of-life care, and preserving patient dignity. For healthcare systems, it can decrease unnecessary hospitalizations, optimize resource utilization, and support evidence-based decision-making with potential for nationwide implementation.

bottom of page