Detecting cells using non-negative matrix factorization on calcium imaging data

• Published in: Neural networks Vol. 55; pp. 11 - 19
• Main Authors: Maruyama, Ryuichi, Maeda, Kazuma, Moroda, Hajime, Kato, Ichiro, Inoue, Masashi, Miyakawa, Hiroyoshi, Aonishi, Toru
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 01.07.2014; Elsevier
• Subjects: Algorithmics. Computability. Computer arithmetics; Algorithms; Animals; Applied sciences; Artificial intelligence; CA1 Region, Hippocampal - cytology; Calcium - analysis; Computer science; control theory; systems; Computer Simulation; Dendrite; Dendrites - chemistry; Detection, estimation, filtering, equalization, prediction; Exact sciences and technology; Independent component analysis; Information, signal and communications theory; Male; Models, Neurological; Multi-cellular calcium imaging; Neurons - chemistry; Neurons - cytology; Non-negative matrix factorization; Pattern recognition. Digital image processing. Computational geometry; Rats; Rats, Wistar; Reproducibility of Results; Semi-automatic cell detection; Signal and communications theory; Signal, noise; Telecommunications and information theory; Theoretical computing; Non-negative matrix factorization; Semi-automatic cell detection; Multi-cellular calcium imaging; Independent component analysis; Dendrite; Mental imagery; Transient response; Calcium; Image databank; Fluorescence; Matrix factorization; Transients; Local structure; Non negative matrix; Pericaryon; Bleaching; Automatic measurement; Imaging; Signal to noise ratio
• ISSN: 0893-6080; 1879-2782; 1879-2782
• DOI: 10.1016/j.neunet.2014.03.007

We propose a cell detection algorithm using non-negative matrix factorization (NMF) on Ca2+ imaging data. To apply NMF to Ca2+ imaging data, we use the bleaching line of the background fluorescence intensity as an a priori background constraint to make the NMF uniquely dissociate the background component from the image data. This constraint helps us to incorporate the effect of dye-bleaching and reduce the non-uniqueness of the solution. We demonstrate that in the case of noisy data, the NMF algorithm can detect cells more accurately than Mukamel’s independent component analysis algorithm, a state-of-art method. We then apply the NMF algorithm to Ca2+ imaging data recorded on the local activities of subcellular structures of multiple cells in a wide area. We show that our method can decompose rapid transient components corresponding to somas and dendrites of many neurons, and furthermore, that it can decompose slow transient components probably corresponding to glial cells.