Machine learning assisted real-time deformability cytometry of CD34+ cells allows to identify patients with myelodysplastic syndromes

Abstract

Diagnosis of myelodysplastic syndrome (MDS) mainly relies on a manual assessment of the peripheral blood and bone marrow cell morphology. The WHO guidelines suggest a visual screening of 200 to 500 cells which inevitably turns the assessor blind to rare cell populations and leads to low reproducibility. Moreover, the human eye is not suited to detect shifts of cellular properties of entire populations. Hence, quantitative image analysis could improve the accuracy and reproducibility of MDS diagnosis. We used real-time deformability cytometry (RT-DC) to measure bone marrow biopsy samples of MDS patients and age-matched healthy individuals. RT-DC is a high-throughput (1000 cells/s) imaging flow cytometer capable of recording morphological and mechanical properties of single cells. Properties of single cells were quantified using automated image analysis, and machine learning was employed to discover morpho-mechanical patterns in thousands of individual cells that allow to distinguish healthy vs. MDS samples. We found that distribution properties of cell sizes differ between healthy and MDS, with MDS showing a narrower distribution of cell sizes. Furthermore, we found a strong correlation between the mechanical properties of cells and the number of disease-determining mutations, inaccessible with current diagnostic approaches. Hence, machine-learning assisted RT-DC could be a promising tool to automate sample analysis to assist experts during diagnosis or provide a scalable solution for MDS diagnosis to regions lacking sufficient medical experts.

Non‐Contact Microfluidic Analysis of the Stiffness of Single Large Extracellular Vesicles from IDH1‐Mutated Glioblastoma Cells

“…Such classification studies of whole cell deformation and shape have been done by others. [ 43–45 ] These studies demonstrated that analyzing multiple physical phenotype metrics together can be more accurate at classifying cancer cell lines than a single metric (e.g., stiffness) alone. Along these lines, the L‐EV stiffness measurements from non‐contact microfluidics could be an additional metric that contributes to the suite of EV characterizations used in the field to classify EVs.…”

Non-contact microfluidic analysis of the stiffness of single large extracellular vesicles from IDH1-mutated glioblastoma cells

“…Such classification studies based on whole cell deformation and shape have been done by others. [41][42][43] These studies demonstrated that analyzing multiple physical phenotype metrics together can be more accurate at classifying cancer cell lines than a single metric (e.g., stiffness) alone.…”

Optofluidic imaging meets deep learning: from merging to emerging

“…This is particularly advantageous for probing complex spatiotemporal characteristics of the biological samples, and thus a more extensive set of image read-outsenriching the information content for DL image analytics. For example, optofluidic imaging has been employed to effectively probe the mechanical properties of cells, 143,[205][206][207] which are proven biomarkers for different disease conditions such as cancer metastasis, inflammation and cardiovascular diseases. 208,209 Other active microfluidic designs are expected to be increasingly compatible with optofluidic imaging, such as on-chip impedance measurements of microparticles and cells.…”