Monday, November 25 2019
9:00am - 11:00am
Masters Presentation
Using Constrained Clustering to Partition Functional MRI Signals Spatiotemporally to Recognize Brain Pattern and BOLD Signals

Functional magnetic resonance imaging (fMRI) measures brain activity by detecting signal changes based on the blood-oxygen level dependent (BOLD) contrast. This technique frequently involves both noise and actual neuronal activation when detecting real brain activity. Thus, identifying the underlying brain signal is critical. Data-driven clustering techniques are widely used in this field. However, noise might adversely affect the cluster result when the temporal and spatial domain is large and signal noise ratio (SNR) is low. Besides, the threshold to determine cluster number remains unclear. The current research will propose a novel application which uses a two-stage constrained cluster algorithm spatial-temporally to reduce the influence of noise and to shorten the research domain by partitioning the fMRI signals into connected time windows, allowing the analysis of BOLD signals only in the active temporal domain. The method also gives a better context for clustering threshold selection.
Speaker:Aixin Zhang
Location:Anschutz Medical Campus

Download as iCalendar