CRADLEConnect: Advancements in Brain Imaging: Looking into the Brain from Structure to Function

In this instalment of CRADLEConnects, join us as our guest speakers share their recent findings in brain and behaviour using state-of-the-art neuroimaging techniques.

Date: 6 May 2024, Monday

Time: 2:00pm to 4:30pm  

Venue: TR+ 30, Level 1 (The ARC) 

Click on this link or the poster below to register.

Speaker 1: Kenichi Oishi, M.D., Ph.D. - Department of Radiology and Radiological Science and the Department of Neurology at Johns Hopkins Medicine

Title: OpenMAP suite: A fast deep-learning approach to fine parcellate gray and white matter structures, sulci, and pathological lesions of the brain

Abstract:

Brain image parcellation is a fundamental process in neuroscientific and clinical research studies, enabling a detailed analysis of specific cerebral regions. Normalization-to-atlas-based methods have been employed for this task, but they face limitations due to variations in brain morphology, especially in pathological conditions. The MALF techniques improved the accuracy of the image parcellation and robustness to variations in brain morphology, but at the cost of high computational demand that requires a lengthy processing time. To overcome these limitations, we are developing a series of neural network models, OpenMAP suite. These models are designed to provide accurate and fast brain MRI parcellation that is robust to variations in machine vendors, scan parameters, and pathological changes. The OpenMAP suite includes three models: OpenMAP-T1 for T1-weighted anatomical MRI, OpenMAP-F for T2-FLAIR, and OpenMAP-D for diffusion MRI. OpenMAP-T1 integrates several U-Net models across six phases, parcellating the brain into 280 anatomical structures that cover the whole brain, including detailed gray and  white matter structures. OpenMAP-F parcellates the brain into 14 anatomical areas, segments periventricular and deep white matter hyperintensity lesions, hematoma, and intraventricular hemorrhage. OpenMAP-D parcellates the brain into 170 anatomical lesions to quantify the dMRI-derived scalar measurements. The adaptability of OpenMAP to a wide range of MRI datasets and its robustness to various scan conditions highlight its potential as a versatile tool in neuroimaging studies and clinical applications.

Biography:

Prof Kenichi Oishi is a Professor of Radiology and Neurology at Johns Hopkins Medicine. His expertise lies in neuroradiology, neurology, and neuroscience, focusing on magnetic resonance imaging and neuroinformatic technologies. Prof Oishi completed two residencies: one in internal medicine at Kobe City General Hospital in Kobe, Japan, and another in neurology at the National Center of Neurology and Psychiatry in Tokyo, Japan. Following his tenure as an Assistant Professor of Neurology at Kobe University School of Medicine, Prof Oishi joined the Division of Magnetic Resonance Research in the Department of Radiology at Johns Hopkins University School of Medicine. Prof Oishi is internationally renowned for developing electronic brain atlases, which are widely utilized in neuroimaging research. He uses these atlases to investigate brain structures involved in language, brain development, aging, and changes due to preterm birth, perinatal hypoxic-ischemic events, stroke, and neurodegenerative diseases. His methods include advanced neuroimaging techniques and comprehensive neurological, behavioral, and neuropsychological assessments to gauge brain injury severity. Prof Oishi's recent work includes analyzing real-world clinical data with artificial intelligence for precision medicine in neonatal hypoxic-ischemic encephalopathy and dementia in older adults. He also conducts in vivo and ex vivo ultra-high-resolution MRI studies to explore anatomical connectivity and changes due to neurodegeneration, along with cross-species comparative neuroanatomy to understand evolutionary changes in brain structures and functions.

Speaker 2: Florence L. Chiang, M.D., Ph.D. - Department of Radiology, Massachusetts General Hospital, Harvard Medical School

Title: “Functional Network Modeling and Microstructural Imaging in Multiple Sclerosis”

Abstract:

Multiple sclerosis is an immune-mediated condition of the central nervous system. Emerging evidence suggests that neurodegeneration is a key component of disease progression in MS. Specifically, localized patterns of gray matter atrophy have been observed in MS, which are strongly correlated with cognitive dysfunction and physical disability. Previously, we developed the Atrophy-based Functional Network (AFN) model in MS as an imaging marker of neurodegeneration in MS using a coordinate-based meta-analytically driven strategy. This targeted method leverages the large corpus of published coordinate-based literature to define an a priori model for further testing in acquired MRI data. In this seminar, I will first provide an overview of this targeted imaging marker development approach. Then, I’ll discuss implementation of the AFN model in resting-state functional MRI of MS in context of functional network disruption and clinical deterioration. Finally, I will present our results of network-based microstructural covariation in MS using high-gradient diffusion MRI. Our findings may help clarify the brain structure-function relationship in MS and support future development of quantitative non-invasive methods for sensitive monitoring of disease progression to enable prompt clinical intervention.

Biography:

Dr Florence Chiang is a neuroradiology fellow at Massachusetts General Hospital. She completed medical school, graduate school, and residency training at The University of Texas Health Science Center at San Antonio. Dr Chiang graduated from the School of Medicine with a Distinction in Research and was elected to the Alpha Omega Alpha Honor Medical Society. She then completed an integrated diagnostic radiology residency-PhD program via the American Board of Radiology Holman Research Pathway. Dr Chiang earned her PhD in Radiological Sciences from the Graduate School of Biomedical Sciences and received the Armand J. Guarino Award for Academic Excellence. She was also a recipient of the RSNA Roentgen Resident Research Award. Dr Chiang’s research is currently funded by the RSNA Research Fellow Grant and Ralph Schlaeger Fellowship Award. Her prior grant support includes the RSNA Research Resident Grant, Julio Palmaz Award for Excellence in Radiology Research, NIH/NIBIB R25, and NIH/NICHD R03. Dr Chiang has authored multiple publications and has presented nationally and internationally. Her research focuses on developing and translating computational neuroimaging methods to identify and implement imaging biomarkers of disease progression and treatment response in multiple sclerosis and other neurodegenerative diseases