Brain is a complex system that requires huge efforts and different tools to be untangled. Over the years, it has become clear that looking at it from a single perspective, or through one lens at a time, is not enough to decode this complexity.
The aim of my research is to find - and make available - novel keys that would help shed light on the pathophysiological mechanisms of the human brain.


One of the main limitations of pharmacological fMRI is its inability to provide a molecular insight into the main effect of compounds. In fact, it relies on the strong assumption that haemodynamic changes can be considered a proxy of altered neurotransmission due to pharmacological agonism or antagonism actions, but the fMRI signal has no intrinsic selectivity to any particular receptor sites.

Supported by the receptor occupancy theory, which states that the magnitude of the drug response is a function of receptor availability and drug binding, REACT offers the possibility to enrich fMRI analysis with the molecular information about target distribution provided by PET and SPECT, a fundamental aspect in drug studies to explore the effects of pharmacological manipulation of brain networks.
See Dipasquale et al, 2019 and Dipasquale et al, 2020 to know more about the use of REACT in drug challenges.

One of the most interesting aspects of this approach is the definition of functional circuits associated to specific neurotransmitters to explore those systems that might be impaired in disease and develop innovative and efficacious targeted treatments. Overall, REACT defines the drug-specific topography of brain functional connectivity and may provide an interesting new fingerprint in the characterisation of novel compounds and potentially greater insight to the commonly observed eclectic response to treatment. Some studies have already been conducted in this direction and have showed extremely promising results.
See Cercignani et al, 2021 and Martins et al, 2021 to know more.

REACT package


Precision medicine can be defined as the “delivery of the right treatment at the right time, every time, to the right person” and aims at obtaining the maximal efficacy from therapy or prevention by accounting for individual variability in genetic and environmental factors.
Big data is a key feature to model individual differences within and across clinical cohorts and develop novel biomarkers for precision medicine. In the context of brain disorders, neuroimaging plays an important role in developing non-invasive biomarkers for an early diagnosis and treatment. Within the category of neuroimaging biomarkers, we can find functional (e.g. fMRI) or metabolic (e.g. FDG PET) markers, which can be used to assess if a patient belongs to a sub-group that will respond or not to a particular treatment.

An emerging methodology that holds the great potential to allow neuroimaging to make a step towards precision medicine is normative modelling, an approach that provides statistical inferences at the level of the individual with respect to a normative pattern. Similar to the growth charts used in paediatric medicine to map height and weight as a function of age, normative modelling generalises this notion by replacing these variables with clinically relevant variables and applying automated statistical techniques to map centiles of variation across the cohort. This strategy offers a way to quantify and characterise the degree of deviation of different individuals from the expected pattern and from one another.

Normative brain templates are becoming more and more popular for detecting structural and functional abnormalities in brain disorders at the individual level, as well as for characterising pharmacological response in imaging studies. However, they have not reached their full potential because of some technical challenges and the biological variability of the data. As a result, these tools have struggled to make an impact beyond the original application for which they have been created.

One of the projects I've been working on is the development and validation of normative PET neuroimaging atlases (PhD project of Alessio Giacomel) and neurotransmission-enriched fMRI-based normative templates to deliver individualised patient diagnosis and treatment guidance.
Having a complete and robust statistical representation of the normal population through PET and fMRI normative templates will allow to assess patient-specific alterations and predict the potential efficacy of a treatment.


The integration of MRI data with transcriptomic data of regional gene expression in the postmortem human brain (Allen Brain Atlas) has opened new avenues to explore among candidate molecular and cellular pathways underlying neuroimaging biomarkers of brain disease, treatment effects, etc.

I am involved in a project led by Dr Daniel Martins whose aim is to apply this method to advance our understanding of different neuroimaging phenotypes, such as structural, functional or molecular alterations in different clinical cohorts.