Clustering of cell populations in flow cytometry data using a combination of Gaussian mixtures

作者：

Highlights：

• A density model based on the interpolation of Gaussian mixture models is presented.

• Non-negative matrix factorization is used to compress the model.

• Affine registration of Gaussian mixture models using the L2 distance is performed.

• The applicability of the method is demonstrated on flow cytometry data.

摘要

Highlights•A density model based on the interpolation of Gaussian mixture models is presented.•Non-negative matrix factorization is used to compress the model.•Affine registration of Gaussian mixture models using the L2 distance is performed.•The applicability of the method is demonstrated on flow cytometry data.

论文关键词：Gaussian mixture model,Convex non-negative matrix factorization,L2 distance,Clustering,Flow cytometry,Acute lymphblastic leukemia

论文评审过程：Received 19 October 2015, Revised 23 March 2016, Accepted 11 April 2016, Available online 29 April 2016, Version of Record 2 September 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.04.004