High-resolution two-photon microscopy captures fluorescent calcium indicators (GCaMP) in neurons of the visual cortex. Raw time-series imaging data contains motion artifacts and requires preprocessing before analysis.
Brain tissue moves during imaging due to breathing, heartbeat, and movement. Rigid and non-rigid registration algorithms align frames to compensate for these shifts, producing stable, artifact-free videos.
Automated detection and segmentation of individual neurons (ROIs) from the imaging field. Extract fluorescence time-series from each neuron while correcting for neuropil contamination from surrounding tissue.
Infer neural spiking activity from calcium fluorescence signals. Deconvolution algorithms account for the slow kinetics of calcium indicators (τ ≈ 0.7s for GCaMP6s) to estimate the underlying action potential firing patterns.