A5082563118- Advanced Neuroimaging Techniques and Applications
- Advanced MRI Techniques and Applications
- Medical Image Segmentation Techniques
- Optical Coherence Tomography Applications
- Functional Brain Connectivity Studies
- Glaucoma and retinal disorders
- Retinal Imaging and Analysis
- Speech and Audio Processing
- Hearing Loss and Rehabilitation
- Microbial infections and disease research
- Ear and Head Tumors
- Digital Imaging for Blood Diseases
- Herpesvirus Infections and Treatments
- Ear Surgery and Otitis Media
- Meningioma and schwannoma management
- Neurological disorders and treatments
- Ultrasonics and Acoustic Wave Propagation
- Neural Networks and Applications
- Parkinson's Disease Mechanisms and Treatments
Massachusetts General Hospital
2024
Harvard University
2024
Vanderbilt University
2021-2023
Abstract Voxel‐based morphometry is an established technique to study focal structural brain differences in neurologic disease. More recently, texture‐based analysis methods have enabled a pattern‐based assessment of group differences, at the patch level rather than voxel level, allowing more sensitive localization between patient populations. In this study, we propose approach identify cerebellum patients with Parkinson's disease ( n = 280) and essential tremor 109). We analyzed anatomical...
In the fields of longitudinal cortical segmentation and surface-based thickness (CT) measurement, difficulty in assessing accuracy remains a substantial limitation due to inability experimental validation against ground truth. Although methods have been developed create synthetic datasets for these purposes, none provide robust mechanism measuring exact changes with approaches. This work presents registration-based technique inducing atrophy truth dataset specifically designed address this...
Synthetic data have emerged as an attractive option for developing machine-learning methods in human neuroimaging, particularly magnetic resonance imaging (MRI)-a modality where image contrast depends enormously on acquisition hardware and parameters. This retrospective paper reviews a family of recently proposed methods, based synthetic data, generalizable machine learning brain MRI analysis. Central to this framework is the concept domain randomization, which involves training neural...
Accuracy validation of cortical thickness measurement is a difficult problem due to the lack ground truth data. To address this need, many methods have been developed synthetically induce gray matter (GM) atrophy in an MRI via deformable registration, creating set images with known changes thickness. However, these often cause blurring atrophied regions, and cannot simulate realistic within deep sulci where cerebrospinal fluid (CSF) obscured or absent. In paper, we present solution using...
Magnetic resonance images (MRIs) are widely used to quantify vestibular schwannoma and the cochlea. Recently, deep learning methods have shown state-of-the-art performance for segmenting these structures. However, training segmentation models may require manual labels in target domain, which is expensive time-consuming. To overcome this problem, domain adaptation an effective way leverage information from source obtain accurate segmentations without requiring domain. In paper, we propose...
Difficulty in validating accuracy remains a substantial setback the field of surface-based cortical thickness (CT) measurement due to lack experimental validation against ground truth. Although methods have been developed create synthetic datasets for this purpose, none provide robust mechanism measuring exact changes with approaches. This work presents registration-based technique inducing atrophy longitudinal, truth dataset specifically designed CT measurements. Across entire brain, we...
Recently, deep-learning methods have achieved human-level performance on multiple sclerosis (MS) lesion segmentation. However, most established are not robust enough for practical use in the real world. They cannot generalize well to images obtained from different clinical sites, or if training and testing datasets contain MRI modalities. To address these robustness issues, bring deep neural networks closer use, we propose addition of data augmentation modality dropout during achieving...
Optical coherence tomography (OCT) is a non-invasive imaging technique widely used for ophthalmology. It can be extended to OCT angiography (OCT-A), which reveals the retinal vasculature with improved contrast. Recent deep learning algorithms produced promising vascular segmentation results; however, 3D vessel remains difficult due lack of manually annotated training data. We propose learning-based method that only supervised by self-synthesized modality named local intensity fusion (LIF)....