fbpx
Researchers Use Machine Learning to Aid in Early Dementia Diagnosis Researchers Use Machine Learning to Aid in Early Dementia Diagnosis
In a new study, reported in Nature Aging, researchers analyzed approximately 1,500 blood proteins from over 50,000 participants in the UK... Researchers Use Machine Learning to Aid in Early Dementia Diagnosis

In a new study, reported in Nature Aging, researchers analyzed approximately 1,500 blood proteins from over 50,000 participants in the UK Biobank, finding profiles that predict future dementia in healthy adults using machine learning algorithms.

According to the study, the four proteins—GFAP, NEFL, GDF15, and LTBP2 were identified. The study states that elevated levels strongly correlate with the future development of dementia. This discovery is significant as it represents a major step towards developing non-invasive blood tests capable of detecting Alzheimer’s disease and other dementias at a pre-symptomatic stage.

The study pinpointed four proteins associated with increased dementia risk:

  • GFAP (Glial Fibrillary Acidic Protein): A marker of astrocyte health, providing structural support to these brain cells.
  • NEFL (Neurofilament Light Chain): Reflects neuronal damage and degeneration.
  • GDF15 (Growth Differentiation Factor 15): Involved in inflammatory responses and cellular stress.
  • LTBP2 (Latent Transforming Growth Factor Beta Binding Protein 2): Plays a role in the extracellular matrix and tissue repair.

These biomarkers offer insights into the pathological processes preceding dementia symptoms, presenting an opportunity for early intervention. “Studies such as this are required if we are to intervene with disease-modifying therapies at the very earliest stage of dementia,” said Amanda Heslegrave, a neuroscientist at University College London.

So how did the study work? Well according to the authors, the methodology involved screening blood samples for levels of 1,463 proteins, utilizing machine learning to design predictive algorithms. These algorithms combined protein levels with demographic factors to predict dementia risk with about 90% accuracy for three subtypes.

This also included Alzheimer’s, more than ten years before official diagnosis. If proven to be reliable these tests could help pave the way for preventative treatment for the disease.  By developing blood tests based on these biomarkers, healthcare professionals could identify individuals at high risk for dementia well before symptoms appear.

This early detection could revolutionize treatment approaches, allowing for interventions that could delay or even prevent the onset of dementia. Currently, the study needs to be replicated and the biomarkers found need to be differentiated between diseases according to the authors.

But this is a major step forward in future treatment.

ODSC Team

ODSC Team

ODSC gathers the attendees, presenters, and companies that are shaping the present and future of data science and AI. ODSC hosts one of the largest gatherings of professional data scientists with major conferences in USA, Europe, and Asia.

1