Dr. Amy Kuceyeski’s Computational Connectomics (CoCo) Laboratory is primarily part of the Weill Cornell Medicine (WCM) Department of Radiology, but has secondary affiliations with the Department of Computational Biology on Cornell University's main campus. Her lab focuses on using quantitative methods, including machine learning, applied to neuroimaging data (mostly magnetic resonance imaging (MRI)) to understand the mysteries of the human brain. One major interest of her lab is in uncovering mechanisms of impairment and recovery after neurological injury or disease, including traumatic brain injury, multiple sclerosis (MS) and stroke. Better understanding of brain-behavior relationships will lead to the development of more accurate diagnostics, prognostics and individualized therapeutics that can boost recovery after neurological disease or injury.
Associated Lab Members
For more than a decade, Amy Kuceyeski has been interested in understanding how the human brain works to better diagnose, prognose and treat neurological disease and injury. Quantitative approaches, including machine learning, applied to data from rapidly evolving neuroimaging techniques, have the potential to enable groundbreaking discoveries about how the brain works. Amy is particularly interested in non-invasive brain stimulation and pharmacological interventions, like psychedelics, that may modulate brain activity and promote recovery from disease or injury. Amy is also the founder and co-director of the cross-campus working group Machine Learning in Medicine, which aims to bring together researchers in Cornell-Ithaca/Cornell-Tech and clinicians/researchers at Weill Cornell Medicine to address medicine's toughest problems. See the group's website.
Keith Jamison is a staff associate in the CoCo Lab. He has a B.S. in computer science and an M.S. in Biomedical Engineering from Cornell University. Through his education and training, he has developed the broad range of skills and expertise necessary to discern scientifically and clinically relevant patterns from large neuroimaging datasets. While working for the Human Connectome Project (HCP) at the University of Minnesota, he helped implement and adapt preprocessing and analysis pipelines for a large number of anatomical, functional and diffusion magnetic resonance imaging (MRI) scans. He also helped design and test new scanning protocols and modalities for some of the HCP-related studies whose data they now propose to analyze. Since joining the CoCo lab in 2017, he has built upon this expertise in neuroimaging acquisition and processing to help develop modeling approaches using this neuroimaging data to better understand the functional and structural connectivity relationship, and how connectivity relates to both healthy brain function and neurological damage or disease.
Sneha Pandya is a research associate in biomedical engineering. Over the past eight years, Sneha has worked closely with RajLab, Weill Cornell Medicine (WCM) Multiple Sclerosis Center, and Sumit Niogi’s WCM Lab, serving both radiology and neurology departments, while applying problem-solving techniques to current clinical problems in the imaging, diagnosis and treatment of major brain diseases. The CoCo Lab’s initiative to use quantitative methods and machine learning on multi-modal neuroimaging data to map brain-behavior relationships has inspired her to be part of this lab. The predominant drive of Sneha’s academic career has been to apply these techniques to current issues in neuroimaging. Sneha plans to expand her research pursuits by developing quantitative and machine learning models in understanding structural-functional relationships and predicting early onset of varying brain diseases.
Ceren Tozlu is a post-doctoral associate in the Weill Cornell Medicine (WCM) Department of Radiology, and in the Cornell University Department of Statistics and Data Science and Computational Biology. She received her M.S. and Ph.D. from the Biostatistics, Biomathematics, Bioinformatics and Health department (3B-H) of Université Claude Bernard Lyon 1 in 2014 and 2018, respectively. She gave lectures to the M.S. students of Cancer, Neuroscience, Biostatistics and Public Health at Université Claude Bernard Lyon for one year and to the medical students at École Santé des Armées (Army Health School of France) for four years. Her M.S. research project focused on application of various machine learning methods to voxel-based conventional human imaging data to predict infarction risk of 3D brain tissue in acute stroke patients. Her Ph.D. thesis focused on modeling disease evolution of stroke and multiple sclerosis (MS) patients based on cross-sectional and longitudinal clinical and imaging data plotted over five years. Her post-doctoral research focuses on (1) modeling of evolution in neurological diseases, particularly in patients with stroke and MS, using statistical and machine learning methods based on demographic, clinical, regional and pair-wise functional and structural connectivity measurements, and (2) identification of the best biomarkers of disease evolution including the particular structural and functional connections contributing to differences in patients with different neurological diseases. Her post-doctoral study aims to develop a novel and personalized model to be used by clinicians to predict individual disease evolution via an application or software, to lead to personalized treatment. She was recently awarded a postdoctoral fellowship grant from the National Multiple Sclerosis Society.
Zijin is a third-year Ph.D. student in the Cornell University Department of Electrical and Computer Engineering. She received her bachelor’s degree in electrical engineering from Zhejiang University, China. Her research focuses on the intersection of machine learning and neuroscience. She is particularly interested in applying innovative machine learning algorithms to brain connectivity network analysis. Her project involves developing a noninvasive, spatially unconstrained and personalized method for neuromodulation, which involves creating deep neural networks for stimuli and brain activation patterns mapping. She hopes that manipulating brain connectivity networks will help alleviate symptoms or boost recovery after neurologic injury.
Emily is a third-year Ph.D. student in neuroscience at Weill Cornell Medicine. She completed her B.Sc. in neuroscience at McGill University, where she studied cortical phenotypes of autism spectrum disorder from structural brain images. In the CoCo Lab, she is studying how the brain’s functional and structural networks change after an ischemic stroke. She hopes her research will be used to better understand how brain activity changes occurring post-stroke impact patient outcomes, and to help identify potential targets for noninvasive therapies.
Lisa is a third-year student in the Weill Cornell Medicine Physiology, Biophysics and Systems Biology Graduate Program, and a recipient of a National Science Foundation Graduate Research Fellowship in bioengineering. After three years of studying economics in construction in her hometown in Russia, she immigrated to the U.S., and graduated with a B.A. in neurobiology from Hunter College in 2019. At Hunter, Lisa also conducted research in the Goldfarb lab, where she tricked brain cancer cells into expressing mutant sodium channels, and then poked them with electrodes to learn how the mutations affected voltage-dependent fast inactivation, a key property of the channel altered in disease phenotypes. Lisa’s strongest aspiration is to contribute to the development of brain-computer interfaces that improve quality of life for people with impaired physical and cognitive function (and maybe even augment natural brain capabilities of healthy people).
Parker is a Ph.D. candidate and National Science Foundation Graduate Research Fellow in computational biology at Cornell University (CU). He transferred from CU’s Department of Chemistry & Chemical Biology, where he obtained his M.S. in 2017, before spending three years as a high school and community college science teacher. Originally from South Carolina, he received his B.S. in chemistry from the University of South Carolina in 2015. He is studying the effects potent serotonergic compounds have on brain activity landscapes using network control theory. He hopes his research can be used to inform theories of consciousness and to better understand, diagnose and treat mental disorders.
The year 2021 saw several publications from the CoCo lab in high-impact journals, including Nature Communications, NeuroImage, Human Brain Mapping, Cortex, Brain and Behavior, Science Advances, Current Opinion in Neurology, European Journal of Neurology and Frontiers in Neuroscience.
Kristen Dams-O'Connor, Ph.D.
Susan A. Gauthier, D.O.
Mert Sabuncu, Ph.D.
Award or grant: National Institutes of Health (NIH)/National Institute of Neurological Disorders and Stroke (NINDS), 1R01NS102646-01A1
Traumatic Brain Injury (TBI) is a leading cause of death and long-term disability, and...
Award or grant: National Institutes of Health/National Institute of Mental Health (NIH/NIMH): 1RF1MH123232
The human brain is an unimaginably complicated system of interconnected neurons capable of complex thought, ...
Award or grant: Leon Levy Foundation Fellowship
Understanding the organization of the white matter structural connectivity (SC) network in the human brain is essential for elucidating the brain-...
Award or grant: National Institutes of Health/National Institute of Neurological Disorders and Stroke (NIH/NINNDS): R21 NS104634-01
Multiple Sclerosis (MS) damages white matter pathways that connect brain regions, i.e., the...
Motor recovery post-ischemic stroke relies on surviving brain networks'ability to compensate for damaged tissue. In rodent models, sensory and motor cortical representations remap onto intact tissue around lesion sites, but remapping to distal...
Award or grant: Cornell University MRI Facility pilot grant
Multiple sclerosis (MS) is one of the most common chronic inflammatory diseases wherein the body attacks protective tissue surrounding nerves. Women are more likely to develop MS compared to men with a female-to-male ratio of 2.8 to 1. The effect of the...
Utilizing multi-modal neuroimaging techniques including fluorescence magnetic resonance imaging (fMRI), diffusion MRI (dMRI), and positron emission tomography (PET), the lab is applying new methods in network control theory to understand the influence of...
Neuromodulation (the localized manipulation of neuronal activity) holds therapeutic promise for neurological/neuropsychiatric diseases. There is currently an unmet need for a non-invasive, safe...