Label‐Free and In Situ Identification of Cells via Combinational Machine Learning Models. Issue 2 (26th December 2021)

• Record Type: Journal Article
• Authors: Xue, Yun‐fan; He, Yang; Wang, Jing; Ren, Ke‐feng; Tian, Pu; Ji, Jian
• Is Part Of: Small methods. Volume 6:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-26
• Subjects: cell identification -- image processing -- label‐free -- machine learning -- microscopy; Nanotechnology -- Methodology -- Periodicals; Nanotechnology -- Periodicals; Periodicals; 620.5028
• DOI: 10.1002/smtd.202101405
• Languages: English
• ISSNs: 2366-9608

Abstract

Abstract: Cell identification and counting in living and coculture systems are crucial in cell interaction studies, but current methods primarily rely on complicated and time‐consuming staining techniques. Here, a label‐free method to precisely recognize, identify, and instantly count cells in situ in coculture systems via combinational machine learning models s presented. A convolutional neural network (CNN) model is first used to generate virtual images of cell nuclei based on unlabeled phase‐contrast images. Coordinates of all the cells are then returned according to the virtual nucleus images using two clustering algorithms. Finally, phase‐contrast images of single cells are cropped based on the coordinates and sent into another CNN model for cell‐type identification. This combinational approach is highly automatic and efficient, which requires few to no manual annotations of images in the training phase. It shows practical performance in different cell culture conditions including cell ratios, densities, and substrate materials, having great potential in real‐time cell tracking and analyzing. Abstract : Cell identification and counting, especially in living and coculture systems, are essential in diversified studies, while relying primarily on complicated and time‐consuming dyeing methods. A machine‐learning‐based method is reported for the label‐free and automatic recognition, identification, and counting of cocultured cells. This method requires few to no manual annotations of data in training and is promising in live cell analyses.