Robust Estimation of Neural Signals in Calcium Imaging

Authors: Hakan Inan, Murat A. Erdogdu, Mark Schnitzer

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the superiority of our robust estimation approach over existing methods on both simulated and real datasets.
Researcher Affiliation Collaboration Hakan Inan 1 inanh@stanford.edu Murat A. Erdogdu 2,3 erdogdu@cs.toronto.edu Mark J. Schnitzer 1,4 mschnitz@stanford.edu 1Stanford University 2Microsoft Research 3Vector Institute 4Howard Hughes Medical Institute
Pseudocode Yes Algorithm 1 Fast Solver for one-sided Huber Loss; Algorithm 2 Tractable and Robust Automated Cell Extraction
Open Source Code No The paper does not provide an explicit statement or a link indicating that the source code for the described methodology is publicly available.
Open Datasets Yes For simulated movies, we use a field of view of size 50 by 50 pixels, and produce data with 1000 time frames. [...] We put EXTRACT to test in this data regime, using an imaging dataset recorded from the dorsal CA1 region of the mouse hippocampus [17], an area known to have high cell density. [17] Y. Ziv, L. D. Burns, E. D. Cocker, E. O. Hamel, K. K. Ghosh, L. J. Kitch, A. El Gamal, and M. J. Schnitzer. Long-term dynamics of ca1 hippocampal place codes. Nature neuroscience, 16(3):264 266, 2013.
Dataset Splits No The paper describes initializations and testing conditions but does not provide specific train/validation/test dataset split information (e.g., percentages or sample counts) for reproducibility.
Hardware Specification No The paper mentions 'GPU implementation' but does not specify any particular GPU models, CPU types, or other detailed hardware specifications used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies or version numbers (e.g., Python 3.8, PyTorch 1.9) required to replicate the experiments.
Experiment Setup Yes For our fixed-point solver, we use κ = 1. [...] A range of [0.5, 1] times the standard deviation of the normally distributed noise is reasonable for κ for most practices. [...] We simulate the fluorescence traces using a Poisson process with rate 0.01 convolved with an exponential kernel with a time constant of 10 frames. [...] We initialize the algorithms with 4 different fractions of ground truth cells: X = {0.2, 0.4, 0.6, 0.8}. We carry out 20 experiments for each X, and we perform a 3 alternating estimation iterations for each algorithm. [...] We choose to use the greedy initializer of CNMF to eliminate any competitive advantage EXTRACT might have due to using its native initializer.