CMStatistics 2017: Start Registration
View Submission - CMStatistics
Title: Calcium imaging: State-of-the-art and future challenges Authors:  Jordan Rodu - University of Virginia (United States) [presenting]
Abstract: Calcium imaging is a powerful technique for observing neuronal populations of ever-increasing sizes, and it creates a unique opportunity for studying population-level dynamics in the brain. Despite the rapid advancement of the technique, many statistical challenges remain. For instance, there is substantial noise introduced in the recordings; images are obtained at a relatively low sampling rate; and there is a nonlinear relationship between the calcium fluorescence and action potentials. These and other factors make spike-timing inference difficult and susceptible to artifacts from pre-processing. We identify major areas of focus and briefly touch on the current state-of-the-art solutions. We also discuss new statistical challenges for calcium imaging studies.