Comprehensive MRI Anomaly Detection Interface for CADASIL and VCID: A Multimodal Approach
Poster No:
329
Submission Type:
Abstract Submission
Authors:
Henry Bockholt1, Bradley Baker2, Laura Eisenmenger3, Michael Geschwind4, David Liebeskind5, Lisa Krishnamurthy6, Jane Paulsen3
Institutions:
1GSU, Atlanta, GA, 2TReNDs, Atlanta, GA, 3University of Wisconsin, Madison, WI, 4UCSF, San Francisco, CA, 5UCLA, Los Angeles, CA, 6GSU, Atalnta, GA
First Author:
Co-Author(s):
Introduction:
To develop "NeuroVasc Imaging Interface (NVII)," a versatile MRI anomaly detection tool integrating multiple imaging modalities to enhance assessment of cerebrovascular abnormalities in conditions such as CADASIL and VCID.
Methods:
NVII is designed to utilize advanced machine learning algorithms for analyzing a broad spectrum of MRI data, encompassing T1, T2, FLAIR, SWI, DTI, and ASL scans. The interface is poised to detect and visualize a wide range of neurovascular anomalies including lesions, lacunes, dilated perivascular spaces, cerebral microbleeds, and infarcts. Currently in a conceptual phase, NVII aims to employ probabilistic models for precise anomaly characterization, facilitating complex imaging pattern interpretation for clinical decision-making.
Results:
This initiative addresses the pressing need for comprehensive diagnostic tools in neurovascular medicine. NVII's ability to integrate and analyze diverse MRI modalities promises to revolutionize early detection, monitoring, and management of CADASIL and VCID. Its anticipated capability to identify and categorize subtle cerebral changes has the potential to significantly improve patient care.
The figure demonstrates a visualization of the NVII interface, where the user will be able to interact with each image type, each probabilistic white matter model and rate the CADA-MRIT inventory accordingly.
The figure demonstrates a visualization of the NVII interface, where the user will be able to interact with each image type, each probabilistic white matter model and rate the CADA-MRIT inventory accordingly.
Conclusions:
NVII, the proposed MRI anomaly detection interface, could substantially advance neuroimaging diagnostics for cerebrovascular diseases. By enabling intricate brain pathology analysis, it holds promise in enhancing treatment strategies, contributing substantially to the precision medicine paradigm. Future development will focus on integrating advanced analytics to not only diagnose but also predict disease progression in patients with CADASIL and VCID.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Neuroinformatics and Data Sharing:
Workflows 2
Keywords:
Cerebral Blood Flow
Cerebrovascular Disease
Degenerative Disease
Informatics
Machine Learning
MRI
Multivariate
Neurological
Segmentation
White Matter
1|2Indicates the priority used for review
Provide references using author date format
2. Di Donato I, Bianchi S, De Stefano N, et al. Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL) as a model of small vessel disease. BMC Med. 2017;15:1-12.
3. Balakrishnan R, Hernández MdCV, Farrall AJ. Automatic segmentation of white matter hyperintensities. Computerized Medical Imaging and Graphics. 2021;88:101867.
4. Wardlaw JM, Smith EE, Biessels GJ, et al. Neuroimaging standards for small vessel disease and neurodegeneration. Lancet Neurol. 2013;12(8):822-38.
5. Hobson J. The Montreal Cognitive Assessment (MoCA). Occupational Medicine. 2015;65(9):764-5.
6. Küçükdeveci AA, Kutlay Ş, Yıldızlar D, et al. Reliability and validity of the World Health Organization Disability Assessment Schedule (WHODAS-II) in stroke. Disability and Rehabilitation. 2013;35(3):214-20.
7. Yan W, Qu G, Hu W, et al. Deep learning in neuroimaging: Promises and challenges. IEEE Signal Processing Magazine. 2022;39(2):87-98.
8. Fedorov A, Johnson J, Damaraju E, et al. Learning of brain tissue segmentation from imperfect labeling. 2017 International Joint Conference on Neural Networks (IJCNN); 2017.
9. Yan W, Qu G, Hu W, et al. Deep learning in neuroimaging: Promises and challenges. IEEE Signal Processing Magazine. 2022;39(2):87-98.
10. Alsop, DC, Detre, JA, Golay, X, et al. Recommended implementation of arterial spin-labeled perfusion MRI. Magnetic Resonance in Medicine. 2015 73(1):102–116.