Brain Connectivity - Pereira's Lab
Our research group brings together a dynamic team of scientists with expertise in both medicine and engineering. This unique combination of perspectives allows us to approach the challenges of early disease detection with a holistic and innovative mindset.
Our goal is to harness the power of cutting-edge technologies and established medical knowledge to develop novel methods for identifying disease markers at an early stage. By focusing on neurodegenerative disorders, we aim to revolutionise the way we diagnose and treat these debilitating conditions.
Addressing the Urgent Need for Early Detection
Early detection of diseases is crucial for improving patient outcomes, delaying disease progression, and ultimately enhancing their quality of life. However, current diagnostic methods are often limited in their ability to identify disease markers early enough to make a significant impact. Our research aims to address this gap by developing sensitive and specific methods that can detect disease-related changes before symptoms manifest themselves.
A Multimodal Approach to Early Detection
Our research encompasses a diverse range of approaches, encompassing both established medical techniques and cutting-edge innovations. We utilize neuroimaging techniques, such as magnetic resonance imaging (MRI) and positron emission tomography (PET), to assess brain structure, function and pathology. We also employ sophisticated computational methods, including graph theory and deep learning, to analyze large datasets of neuroimaging data and extract meaningful patterns.
Collaboration and Innovation
We believe that collaboration is essential for driving innovation in early disease detection. We actively collaborate with researchers from diverse fields, including neurology, neuroscience, computer science, biology and engineering. This cross-disciplinary approach fosters the exchange of ideas and expertise, leading to the development of novel and effective methods for identifying disease markers.
Advancing the Science of Early Disease Detection
Our research is not merely focused on developing new technologies; it also aims to contribute to the broader understanding of disease mechanisms. By characterising the subtle changes that precede the onset of symptoms, we hope to gain deeper insights into the underlying pathologies of neurodegenerative disorders. This knowledge can ultimately inform the development of more effective therapeutic strategies.
Join Us in Our Mission
We invite you to explore our website, https://joanabpereiralab.com, to learn more about our research and its potential impact. We are committed to sharing our methods with the scientific community and collaborating with others to advance the field of early disease detection. Together, we can revolutionise the way we diagnose and treat these debilitating conditions, improving the lives of countless individuals worldwide.
Multilayer Brain Connectivity
Neuroscientists are increasingly recognizing that the brain's connectivity is not a single process but rather a multifaceted system. Our research aims to harness this complexity by integrating information from multiple neuroimaging modalities, including diffusion tensor imaging (DTI), functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and anatomical T1-weighted images, to construct a comprehensive representation of the brain's connectome. By analyzing the interactions between these different modalities, we hope to gain a deeper understanding of the interplay between various pathophysiological processes that contribute to neurodegenerative diseases.
Dynamic Brain Connectivity
The human brain is not a rigid structure but rather an interconnected network that is constantly adapting and changing. Our research focuses on developing novel methods to capture these dynamic changes in brain connectivity. We have developed a delayed correlation-based approach that measures the temporal dependency of functional connectivity between brain regions. This method has proven to be more sensitive than traditional dynamic connectivity measures in detecting early signs of pathology in patients with neurodegenerative diseases such as Parkinson's disease and Alzheimer's disease.
The brain's connectivity patterns are not uniform but rather exhibit intricate variations across the entire brain. We are working with methods to quantify these continuous voxel-wise patterns of connectivity, which can provide valuable insights into the functional organization of the brain and its role in disease progression. By applying this approach to the precuneus, a region associated with memory and cognition, we have identified early signs of pathology in ApoE e4 carriers, individuals who carry a genetic risk factor for Alzheimer's disease. We have also used this method to characterise the complex functional connectivity patterns of the locus coeruleus, a brainstem nucleus involved in arousal, stress response, and emotional regulation. These patterns were found to be associated with age-related decline in memory, emotional regulation, anxiety, and depression.
Deep learning algorithms, inspired by the structure and function of the human brain, have emerged as powerful tools for analysing complex neuronal data. We are exploring the application of deep learning models to various aspects of neurodegenerative disease research. These models can be used to detect subtle signs of disease in neuroimaging data, identify the most predictive connectivity patterns for different outcome measures, and distinguish between different groups of patients. Our research includes a variety of deep learning architectures, including convolutional neural networks, graph neural networks, transformers, and variational autoencoders. We have also written a book on deep learning for various fields, which will be published next year by No Starch Press.
Even patients diagnosed with the same neurodegenerative disorder can exhibit distinct clinical trajectories, suggesting that there may be different pathophysiological pathways underlying their disease progression. Our research aims to identify these distinct clinical phenotypes and investigate the underlying mechanisms that drive them. By understanding how pathology spreads and how patients' clinical symptoms evolve, we can develop more personalized treatment approaches.
Proteomics and Transcriptomics
Biomarkers, such as protein levels, can serve as valuable indicators of disease progression and treatment response. Our research utilizes proteomics to identify and characterize biomarkers associated with Alzheimer's and Parkinson's diseases. These biomarkers are typically obtained from cerebrospinal fluid or blood plasma and can provide insights into the molecular mechanisms underlying neurodegeneration. We are also exploring the use of transcriptomic approaches to link imaging findings to specific genes, offering a deeper understanding of the genetic basis of these disorders.
Get in touch
If you are interested in knowing more about our research or making a donation please contact firstname.lastname@example.org.