Integration of light scattering with machine learning for label free cell detection.

Light scattering has been used for label-free cell detection. The angular light scattering patterns from the cells are unique to them based on the cell size, nucleus size, number of mitochondria, and cell surface roughness. The patterns collected from the cells can then be classified based on different image characteristics. We have also developed a machine learning (ML) method to classify these cell light scattering patterns. As a case study we have used this light scattering technique integrated with the machine learning to analyze staurosporine-treated SH-SY5Y neuroblastoma cells and compare them to non-treated control cells. Experimental results show that the ML technique can provide a classification accuracy (treated versus non-treated) of over 90%. The predicted percentage of the treated cells in a mixed solution is within 5% of the reference (ground-truth) value and the technique has the potential to be a viable method for real-time detection and diagnosis.

Physical characterization of hematopoietic stem cells using multidirectional label-free light scatterings.

An experimental setup capable of measuring simultaneous 2D scattered light angular distribution from two directions to study cell morphology without the use of bio-labels was developed. Experiments with hematopoietic stem cells (CD34+ cells) show good agreement with detailed numerical simulations of light scattering. Numerical simulations and computer models of cells are used to identify physical features of cells with the largest scattering cross sections. This allows for determination of size, geometry of the nucleus and distribution of mitochondria in hematopoietic stem cells by means of our label-free method.

Label-free and noninvasive optical detection of the distribution of nanometer-size mitochondria in single cells.

A microfluidic flow cytometric technique capable of obtaining information on nanometer-sized organelles in single cells in a label-free, noninvasive optical manner was developed. Experimental two-dimensional (2D) light scattering patterns from malignant lymphoid cells (Jurkat cell line) and normal hematopoietic stem cells (cord blood CD34+ cells) were compared with those obtained from finite-difference time-domain simulations. In the simulations, we assumed that the mitochondria were randomly distributed throughout a Jurkat cell, and aggregated in a CD34+ cell. Comparison of the experimental and simulated light scattering patterns led us to conclude that distinction from these two types of cells may be due to different mitochondrial distributions. This observation was confirmed by conventional confocal fluorescence microscopy. A method for potential cell discrimination was developed based on analysis of the 2D light scattering patterns. Potential clinical applications using mitochondria as intrinsic biological markers in single cells were discussed in terms of normal cells (CD34+ cell and lymphocytes) versus malignant cells (THP-1 and Jurkat cell lines).

Microscope-based label-free microfluidic cytometry.

A microscope-based label-free microfluidic cytometer capable of acquiring two dimensional light scatter patterns from single cells, pattern analysis of which determines cellular information such as cell size, orientation and inner nanostructure, was developed. Finite-difference time-domain numerical simulations compared favorably with experimental scatter patterns from micrometer-sized beads and cells. The device was capable of obtaining light scattering patterns from the smallest mature blood cells (platelets) and cord blood hematopoietic stem/progenitor cells (CD34 + cells) and myeloid precursor cells. The potential for evaluation of cells using this label-free microfluidic cytometric technique was discussed.