UC researchers use artificial intelligence to reveal effects of chronic disease on brain ageing

A team of researchers from the University of Coimbra (UC) has demonstrated the impact that certain chronic diseases associated with cognitive decline, such as Alzheimer's disease, type 2 diabetes, and schizophrenia, can have on brain ageing. The scientists used artificial intelligence techniques and several local and global databases to estimate the Brain Age Gap (BrainAGE)—the difference between predicted brain age and chronological age. This method provides a new way to measure the impact of these chronic diseases, which directly or indirectly affect the brain. In patients with Alzheimer's disease, the brain age gap can be as much as nine years higher than the patient's actual age.

The study was recently published in the journal Brain Communications, with Maria Fátima Dias as first author. Dr Dias is a researcher at the Coimbra Institute for Biomedical Imaging and Translational Research (CIBIT) of the UC Institute of Nuclear Sciences Applied to Health (ICNAS), and the Centre for Informatics and Systems of the University of Coimbra (CISUC). The study was carried out under the supervision of professors and researchers Miguel Castelo-Branco (director of CIBIT and professor at the UC Faculty of Medicine), and Paulo de Carvalho (director of the CIUSC Clinical Informatics Laboratory and professor at the UC Faculty of Science and Technology). The research uses the new concept of brain age gap estimation - the difference between a person's chronological age and their estimated brain age (determined using artificial intelligence models that analyse magnetic resonance images of the brain) - to demonstrate the impact of specific diseases on brain ageing.

"Estimated brain age is the 'biological age' of the brain, predicted by models that analyse brain images. Comparing it with 'chronological age' (the actual age of a person, measured in years) allows us to determine whether the brain has aged more or less quickly than expected. A positive brain age gap indicates accelerated brain ageing, while a negative value indicates a biologically younger brain with slower ageing," explains Miguel Castelo-Branco, lead author of the article.

The study used different artificial intelligence models to create maps that help interpret which brain regions contribute most to estimating biological age. Metrics were also established to determine the average impact of each of the diseases studied (all of which are associated with risk factors for cognitive decline) on brain ageing. "In the case of schizophrenia, brain ageing is about 2 years, in type 2 diabetes it is 5 years and in Alzheimer's disease it reaches 9 years," explains the CIBIT researcher and director.

These findings could pave the way for new approaches to diagnosing the cognitive decline associated with these diseases. "In practice, this measure could be used as a useful biomarker for the early diagnosis of neurodegenerative diseases," concludes Miguel Castelo-Branco.

The study involved researchers from the Faculty of Medicine at the University of Coimbra (UC), the Centre for Biomedical Imaging and Translational Research, the Institute of Nuclear Sciences Applied to Health (ICNAS), the Centre for Informatics and Systems at UC, and the Intelligent Systems Associated Laboratory.

The paper, authored by Maria Fátima Dias, João Valente Duarte, Paulo de Carvalho, and Miguel Castelo-Branco, is available at https://academic.oup.com/braincomms/article/7/2/fcaf109/8069058.