Search

Classification and sorting of cells using image-activated cell sorting (IACS) systems can bring significant insight to biomedical sciences. Incorporating deep learning algorithms into IACS enables cell classification and isolation based on complex and human-vision uninterpretable morphological features within a heterogeneous cell population. However, the limited capabilities and complicated implementation of deep learning-assisted IACS systems reported to date hinder the adoption of the systems for a wide range of biomedical research. Here, we present image-activated cell sorting by applying fast deep learning algorithms to conduct cell sorting without labeling. The overall sorting latency, including signal processing and AI inferencing, is less than 3 ms, and the training time for the deep learning model is less than 30 min with a training dataset of 20,000 images. Both values set the record for IACS with sorting by AI inference. . We demonstrated our system performance through a 2-part polystyrene beads sorting experiment with 96.6% sorting purity, and a 3-part human leukocytes sorting experiment with 89.05% sorting purity for monocytes, 92.00% sorting purity for lymphocytes, and 98.24% sorting purity for granulocytes. The above performance was achieved with simple hardware containing only 1 FPGA, 1 PC and GPU, as a result of an optimized custom CNN UNet and efficient use of computing power. The system provides a compact, sterile, low-cost, label-free, and low-latency cell sorting solution based on real-time AI inferencing and fast training of the deep learning model.

Cell type classification and isolation according to imaging and spatial characteristics, beyond traditional fluorescently labeled biomarkers, enable the development of new biological insight and establishment of connections between phenotypical, morphological, and genomic cell information in normal and diseased states. Here we demonstrate a 2D image-guided cell sorter and a 3D imaging flow cytometer using fast scanning laser excitation sources. Both systems feature a cameraless design, which reconstructs cell images from the temporal readout of photomultiplier tubes.

The microfluidic-based, label-free image-guided cell sorter offers a low-cost, high information content, and disposable solution that overcomes many limitations in conventional cell sorters. However, flow confinement for most microfluidic devices is generally only one-dimensional using sheath flow. As a result, the equilibrium distribution of cells spreads beyond the focal plane of commonly used Gaussian laser excitation beams, resulting in a large number of blurred images that hinder subsequent cell sorting based on cell image features. To address this issue, we present a Bessel–Gaussian beam image-guided cell sorter with an ultra-long depth of focus, enabling focused images of >85% of passing cells. This system features label-free sorting capabilities based on features extracted from the output temporal waveform of a photomultiplier tube (PMT) detector. For the sorting of polystyrene beads, SKNO1 leukemia cells, and Scenedesmus green algae, our results indicate a sorting purity of 97%, 97%, and 98%, respectively, showing that the temporal waveforms from the PMT outputs have strong correlations with cell image features. These correlations are also confirmed by off-line reconstructed cell images from a temporal–spatial transformation algorithm tailored to the scanning Bessel–Gaussian beam.

Classification of cell types and isolation of targeted cells according to their imaging and spatial characteristics, beyond traditional fluorescently labeled biomarkers, enable the development of new biological insight and establishment of connections between phenotypical and morphological, and genomic cell information in normal and diseased states. Here, we demonstrate a microfluidic flow cytometer cell sorter that uses fast scanning laser excitation sources and photomultiplier tubes (PMTs), coupled with real-time image processing, to image and sort cells based on user-defined spatial features. Dozens of features, including label-free characteristics such as size, shape, scattering pattern, morphology, as well as the spatial distribution of fluorescently labeled biomarkers, greatly expand the available information space in cell analysis. The use of a disposable microfluidic cartridge that is compatible with the imaging system contributes to size reduction and lowered risk of cross contamination. The so-called cameraless design that uses PMTs to replace CCD makes the system more compatible with conventional fluorescence-activated cell sorters and user friendly, while enjoying the sensitivity and benefits of PMTs. The cell image features can be directly extracted from the temporal waveforms or from the reconstructed cell images via temporal–spatial transformation.