Using cutting-edge AI to analyse medical data, scientists are now able to predict a person's risk of developing diseases such as Alzheimer's and heart disease up to 10 years before they are diagnosed.
The researchers used machine learning to study blood samples from more than 45,000 people.
AI tools can identify patterns of proteins in the blood that are associated with an increased risk of disease, allowing researchers to accurately predict how likely a person is to develop a disease before symptoms appear.
Experts say being able to detect early warning signs of a wide range of diseases could lead to opportunities for early intervention and prevention.
Early stage
The research team, which included researchers from the University of Edinburgh and commercial collaborators Optima Partners and Biogen, analysed blood samples from UK Biobank, a database of genetic and health information on 500,000 UK participants.
They used AI and machine learning tools to identify protein patterns in the blood that indicate the onset of common diseases such as Alzheimer's, heart disease, and type 2 diabetes.
Disease diagnosis information was obtained from participants' medical records for up to 10 years after blood sample measurement.
The team then tested whether the patterns could be used to diagnose medical conditions in blood samples from another group of people whose data had not been used to create the protein patterns.
The researchers found that protein patterns improved predictive accuracy beyond traditional risk factors such as age, sex, lifestyle, cholesterol and other commonly measured clinical variables.
While this form of analysis is not expected to be conducted anytime soon, experts say the study is a promising step forward in risk prediction.
The study is published in Nature Ageing and was funded by Wellcome.
It is heartening to know how possible it is to predict the outcome of various diseases from just a single blood sample. Being able to detect early warning signs of various diseases will create opportunities for early intervention and prevention, marking a landmark moment for the healthcare industry.
Further studies are still required to translate these findings into clinical practice, but our findings have laid a strong foundation for incorporating novel risk prediction signatures to elucidate potential pathways and mechanisms underlying disease.
This type of pattern recognition would not be possible without modern machine learning techniques and the ability to analyze data at this scale, making it possible to address some of the most pressing healthcare challenges of our time.
/University of Edinburgh Public Release. This material from the originating organization/author may be out of date and has been edited for clarity, style and length. Mirage.News takes no organizational stance or position and all views, positions and conclusions expressed here are solely those of the authors. Read the full article here.
