Researchers at Aalto University in Finland have developed a new artificial intelligence (AI) tool that can more accurately predict an individual's risk of developing common diseases, including cardiovascular disease, diabetes, or liver disease, thereby providing a new approach for medical health risk assessment.
According to a recent press release from Aalto University, this AI tool, named “survivalFM,” uses machine learning technology to comprehensively consider the complex relationships among multiple risk factors, offering more accurate and personalized risk assessments than traditional predictive models.
The researchers explained that traditional models often analyze risk factors individually, while the new tool can simultaneously analyze the interactions between factors such as age, cholesterol levels, and lifestyle, taking into account their impact on long-term health, thus making predictions closer to real-life situations.
They have already tested the tool using data from the UK Biobank. The UK Biobank contains medical records, lifestyle data, genetic information, and more from about 500,000 UK volunteers. After training, the tool can predict the risk of developing 10 common diseases within the next 10 years. Test results showed that this tool outperformed traditional models in most cases.
The researchers stated that the tool is also interpretable: medical and research personnel can not only receive high-risk alerts but also see which risk factors jointly influenced the assessment results. The related research paper was recently published in the British academic journal Nature Communications.