Identification of non-activated lymphocytes using three-dimensional refractive index tomography and machine learning

Abstract

Identification of lymphocyte cell types are crucial for understanding their pathophysiological roles in human diseases. Current methods for discriminating lymphocyte cell types primarily rely on labelling techniques with magnetic beads or fluorescence agents, which take time and have costs for sample preparation and may also have a potential risk of altering cellular functions. Here, we present the identification of non-activated lymphocyte cell types at the single-cell level using refractive index (RI) tomography and machine learning. From the measurements of three-dimensional RI maps of individual lymphocytes, the morphological and biochemical properties of the cells are quantitatively retrieved. To construct cell type classification models, various statistical classification algorithms are compared, and the k-NN (k = 4) algorithm was selected. The algorithm combines multiple quantitative characteristics of the lymphocyte to construct the cell type classifiers. After optimizing the feature sets via cross-validation, the trained classifiers enable identification of three lymphocyte cell types (B, CD4+ T, and CD8+ T cells) with high sensitivity and specificity. The present method, which combines RI tomography and machine learning for the first time to our knowledge, could be a versatile tool for investigating the pathophysiological roles of lymphocytes in various diseases including cancers, autoimmune diseases, and virus infections.

Evaluation of holographic imaging cytometer holomonitor M4® motility applications

“…As a label-free imaging technology that is non-cytotoxic even with prolonged exposure, QPI is a noninvasive method for single cell analysis of adherent cell behavior. More recent reports have used machine learning based analyses of features derived from QPI images to distinguish metastatic potential, drug-response, program activation, and modes of cellular migration (17)(18)(19)(20)(21)(22). (8)).…”

Label‐free bioanalysis of Leishmania infantum using refractive index tomography with partially coherent illumination

“…In the reconstructed 3D RI contrast it is observed that part of the cell's boundary at the top and bottom is missing, which makes difficult to estimate the cell's volume. This vanishing effect is due to the well‐known missing‐cone problem which results from the limited NA of the condenser and objective lenses . Note that the missing‐cone problem affects to all the ODT modalities and can be avoided by performing multiple object rotations as reported in Ref.…”

“…, which in turns also provides isotropic resolution. Nevertheless, it has been proposed iterative and machine learning methods to mitigate the boundary degradation allowing better estimation of the cell's volume in presence of the missing‐cone problem . Here, we have considered the analysis of the average DMC instead of the dry mass because it avoids integrating the DMC over the cell's volume, thus preventing such a side effect of the missing‐cone problem in the analysis.…”

Machine learning‐based in‐line holographic sensing of unstained malaria‐infected red blood cells

“…However, exogenous labeling agents were used in the diagnosis. Furthermore, quantitative phase‐contrast imaging (QPI) techniques combined with machine learning algorithms have been utilized to recognize types of cells or classify the states of biosamples, including bacteria , cancer cells , sperm cells , lymphocytes , macrophage activation , microorganisms , microobjects and RBCs . Since QPI techniques provide valuable phase information related with 3D morphology and biophysical properties of samples, iRBCs could be distinguished from healthy RBCs (hRBCs) with a relatively high accuracy (>91%) .…”