Fig. 5: Traditional image processing and machine learning-based image processing techniques in cell sorting.

a Traditional threshold segmentation approach for edge detection and identification of target cells. Reproduced with permission¹⁰⁵. Copyright 2011, The Royal Society of Chemistry. b An intensity-phase-based beam-shaping algorithm can be utilized to identify individual metal nanoparticles based on pixel intensity, enabling effective sorting. Reproduced with permission⁶⁵. Copyright 2020, American Chemical Society. c Support Vector Machines (SVM) are utilized for the classification of human chromosomes¹⁰⁸. Reproduced with permission¹⁰⁸. Copyright 2017, Elsevier Ltd. d Multilayer Perceptrons (MLP) can effectively differentiate between healthy and diseased cells by analysing their mechanical properties. Reproduced with permission¹⁰⁹. Copyright 2023, Springer Nature. e K-Nearest Neighbor (KNN) can be used to identify epileptic components. Reproduced with permission¹¹⁰. Copyright 2022, Wiley Periodicals LLC. f Random Forest can accurately classify between breast cancer and normal single cells. Reproduced with permission¹¹². Copyright 2020, Elsevier B.V