European data platform
The FWF is co-funding the large-scale project ArtiPro run by ERA PerMed, a European partnership for personalized medicine. With the help of AI, new factors are being identified to improve the treatment of depression.
Why me? Depression in the brain scan
Sitting deep inside the brain, the amygdala is responsible for regulating emotions. A well-known hypothesis about depression posits that this area of the brain is hyper-reactive in people who are susceptible to suffer from the disorder. Those affected react particularly strongly to emotional stimuli from their environment. But it is not quite that simple.
“The problem is that you can't differentiate between people with different susceptibility to illness just based on this one factor alone. To date , such knowledge has no practical consequences,” explains Roberto Viviani, Professor of Clinical Psychology at the University of Innsbruck. Viviani is an expert in brain scans from functional magnetic resonance imaging (fMRI), which show which areas of the brain are active at what point and, for instance, react to external stimuli.
A major European project
In collaboration with an international team of researchers from Germany, Italy, Croatia, Norway and Israel, Viviani has compiled a comprehensive database on depression, ranging from genetic information to imaging data. The Austrian Science Fund FWF is funding the Austrian participation in the large-scale project “Artificial intelligence for personalized medicine in depression (ArtiPro)” of ERA PerMed, a European partnership for the promotion of personalized medicine. With the help of artificial intelligence (AI), the researchers now aim to vet the complex data for new factors that can improve predictions as to susceptibility to the affliction as well as the response to treatment.
International efforts
Data-based medicine can offer people with depressions more precise diagnoses and individually tailored therapies.
Taxing work with huge amounts of data
While fMRI technology has been established for over 30 years, the medical significance of the abundance of data it furnishes is scantily understood , as Viviani emphasizes. “I wonder whether our evaluations or use of these data have been too simplistic in the past. That's why I'm investigating what kind of signals or information can be found in the imaging,” says the psychiatrist. Despite AI support, this is a very time-consuming undertaking due to the small-scale image data – but, as initial results show, it is a worthwhile effort.
When Viviani’s team analyzed the fMRI images, factors came into focus that were traditionally regarded as “physiological noise” or disruptive factors. The researchers explored what useful information can be extracted from the signals that document brain activity. They compared images from the entire brain with the thin area of the cranial bone, which contains no neurons but does have blood vessels. Superimposing these results showed that a significant proportion of the fluctuating activity of the brain actually correlates with blood vessel signals.
“The results indicate that there are other important signals besides the cortex that are not originated by neurons,” explains Viviani. He suspects they are caused by a mechanism of the autonomic nervous system that simultaneously affects blood flow to the brain and cranial bones. Further studies will be devoted to elucidating how these differences affect the processing of emotional stimuli, which are also perceived differently in depression. In a next step, the team seeks to analyze the images of people while actively engaged in solving tasks or while absorbing stimuli instead of those of people while resting, as hitherto.
International cooperation is the only option
In the context of the joint project, Viviani is also involved in studies on individual differences in response to psychotropic drugs, particularly antidepressants. “With partners from Germany, we have developed statistical models to describe the influence of genetic variants in important enzymes on drug metabolism,” notes Viviani.
Viviani considers international cooperation to be an essential requirement for compiling such comprehensive data platforms. “We need healthcare data, especially in the era of AI,” Viviani says with conviction, even if national data protection regulations sometimes severely restrict the exchange of sensitive information. Data-based personalized medicine, whose advancement is fostered by ERA PerMed, could offer people with depressions more precise diagnoses and individually tailored therapies – all across Europe.
Personal details
Roberto Viviani is Associate Professor in Clinical Psychology at the University of Innsbruck and head of a research group at the University of Ulm, Germany. He studied psychiatry and neurology at the University of Cambridge, UK. His research interests include imaging in the neurosciences and the statistical analysis of image data. The Austrian Science Fund FWF is providing roughly EUR 287,000 in funding to the project “Artificial intelligence for personalized medicine in depression (ArtiPro)” run by ERA PerMed, a European partnership for personalized medicine.
Publications
Huber D., Rabl L., Orsini C., Labek K., Viviani R.: The fMRI global signal and its association with the signal from cranial bone, in: NeuroImage 2024
Viviani R., Berres J., Stingl J.C.: Phenotypic Models of Drug-Drug-Gene Interactions Mediated by Cytochrome Drug-Metabolizing Enzymes, in: Clinical Pharmacology and Therapeutics 2024
