Title: Contamination detection in the cultivation of leukocyte based on image sparsity evaluation

Authors: Lianghong Wu; Zhiyang Li; Liang Chen; Cili Zuo; Hongqiang Zhang

Addresses: School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China ' School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China ' School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China ' School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China ' School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, 411201, China

Abstract: The contamination in the cultivation of cells seriously affects the reliability and reproducibility of experimental results. Currently, the detection of contamination in cells relies heavily on manual observation, which is labour-intensive and time-consuming. In this paper, it proposes a sparse matrix clustering (SMC) method based on the principle of matrix sparsity to automatically detect the contamination in leukocytes. Firstly, the image segmentation and local adaptive binarisation techniques are used to eliminate the noise points and shadows. Then, a scoring map of image sparsity based on the pixel distribution of segmented images is proposed to index the pollution degree of the leukocyte. By dynamically determining the threshold for evaluating image sparsity based on the maximum distributed pixels on the scoring map, the image sparsity is used as a feature for contamination classification. Experimental results show that this method achieves an accuracy of 98.8% for detecting contamination in leukocyte culture images with fast detection speed, which can be used as an efficient cell contamination detection approach in the biomedical field.

Keywords: image sparsity evaluation; image segmentation; local adaptive binarisation; leukocyte contamination; sparse matrix clustering; contamination detection; image feature extraction; image classification.

DOI: 10.1504/IJAAC.2025.142984

International Journal of Automation and Control, 2025 Vol.19 No.1, pp.18 - 36

Received: 15 Sep 2023
Accepted: 01 Jan 2024
Published online: 02 Dec 2024 *