Using Neuroimaging Databases and Statistical Learning to Test Biophysical Theories and Examine Neurobiological Mechanisms of Disease.

Date(s) - 01/23/2023
3:00 pm - 4:00 pm

Communicore, C1-17

Pedro A. Valdes Hernandez, Ph.D., Research Assistant Professor, College of Dentistry, University of Florida

The creation of large Human Brain Mapping Projects has opened the opportunity to explore in vivo brain mechanisms like never before. In this talk, I will present how we have used multimodal neuroimaging datasets, and statistical and machine learning to find ways to deal with the unfeasibility of some main problems present in Electrophysiology across species, to test competing biophysical theories of how the waves of activity are generated in the brain, to create clinically relevant biomarkers of epileptic foci, to characterize chronic pain-related alterations in brain anatomy and function.


Dr. Valdes-Hernandez Nuclear Physics and completed his Ph.D. in Physics at the University of Havana in 2016. He specializes in Neuroimaging, specifically, in the disciplines of Neuroinformatics, Medical Image Analysis, Biostatistics, Computational Neuroanatomy, Computational Neurosciences (e.g., modeling of brain dynamics), and the Biophysical Modelling of the Brain (specifically modeling the underlying biophysical mechanisms and signal formation of EEG/MEG and BOLD fMRI). He has worked with large multimodal neuroimaging databases of both clinical and preclinical (rats and monkeys) studies. Using these databases, he has applied statistical learning to test several biophysical theories of the brain. He has more than 16 years of experience performing MRI and EEG experiments. This includes the man in both human and rat studies, as well as the acquisition and analysis of simultaneous and standalone EEG and fMRI data. With a mindset on the betterment of the health and quality of life of society, he has dealt with a variety of pathological populations, covering a wide range of ages (e.g., children with Epilepsy and older adults with and without chronic pain). His current research agenda follows a two-pronged approach: the use of AI methods to develop biomarkers of chronic pain and to understand how the brain works.