A5027591073- Functional Brain Connectivity Studies
- Medical Image Segmentation Techniques
- Cell Image Analysis Techniques
- Advanced Neural Network Applications
- Advanced Neuroimaging Techniques and Applications
- Machine Learning in Materials Science
- Neural Networks and Applications
- EEG and Brain-Computer Interfaces
- Machine Learning and Data Classification
- Machine Learning and Algorithms
- Optical Imaging and Spectroscopy Techniques
- COVID-19 epidemiological studies
- Advanced Bandit Algorithms Research
- COVID-19 diagnosis using AI
- Advanced MRI Techniques and Applications
- Data-Driven Disease Surveillance
Georgia Institute of Technology
2020-2021
Georgia Tech Research Institute
2021
Neural microarchitecture is heterogeneous, varying both across and within brain regions. The consistent identification of regions interest one the most critical aspects in examining neurocircuitry, as these structures serve vital landmarks with which to map pathways. Access continuous, three-dimensional volumes that span multiple areas not only provides richer context for identifying such landmarks, but also enables a deeper probing microstructures within. Here, we describe X-ray...
Abstract Neural cytoarchitecture is heterogeneous, varying both across and within brain regions. The consistent identification of regions interest one the most critical aspects in examining neurocircuitry, as these structures serve vital landmarks with which to map pathways. Access continuous, three-dimensional volumes that span multiple areas not only provides richer context for identifying such landmarks, but also enables a deeper probing microstructures within. Here, we describe X-ray...
Methods for resolving the brain's microstructure are rapidly improving, allowing us to image large brain volumes at high resolutions. As a result, interrogation of samples spanning multiple diversified regions is becoming increasingly common. Understanding these often requires multi-scale processing: segmentation detailed and large-scale modelling macrostructure. Current mapping algorithms analyze data only single scale, optimization each scale occurs independently, potentially limiting...
Most existing black-box optimization methods assume that all variables in the system being optimized have equal cost and can change freely at each iteration. However, many real world systems, inputs are passed through a sequence of different operations or modules, making earlier stages processing more costly to update. Such structure imposes on switching early parts data pipeline. In this work, we propose new algorithm for switch cost-aware called Lazy Modular Bayesian Optimization (LaMBO)....