An electronic microfluidic platform for on-chip apoptosis quantification using annexin v-based PS externalization detection

Abstract

An exemplary system and method are disclosed for a microfluidic platform for on-chip apoptosis quantification using an impedance-based instrumented quantification chamber that can provide a direct measure of subpopulation with apoptosis cells within a cell population as the cells flow through the instrumented quantification chamber in the analysis. The exemplary system and method can accurately and quickly detect apoptotic markers of a cell in a stream of cells without disturbing cellular conditions to provide near real-time measure of apoptotic cell subpopulation and with minimal wet-lab processing.

Description

AN ELECTRONIC MICROFLUIDIC PLATFORM FOR ON-CHIP APOPTOSIS QUANTIFICATION USING ANNEXIN V-BASED PS EXTERNALIZATION DETECTIONRelated Application

[0001] This U.S. application claims priority to, and the benefit of, U.S. Provisional Patent Application No. 63 / 602,187, filed November 22, 2023, entitled “An Electronic Microfluidic Platform for On-Chip Apoptosis Quantification using Annexin V-Based PS Extemalization Detection-GT NEXT,” which is incorporated by reference herein in its entirety.Background

[0002] Programmed cell death - apoptosis - is a process in multicellular organisms with physiological and pathological events that define the initial stage of cell death. Apoptosis immunoassays and labels are commercially available reagents that bind to apoptotic cells for staining cell populations that are apoptotic and are used, for example, via flow cytometry in the study of pathogens and in the development of therapeutics.

[0003] Flow cytometry is often used to detect and measure physical and chemical characteristics of a cell population (or particles) by suspending cells (or particles) in a fluid and injecting the fluid into a flow cytometer instrument, typically having a camera / microscope, that can provide cell count, cell sort, cell characterization, biomarker information, among others, based on immunoassays labels applied to the cell population. Staining cells and measuring them via flow cytometry can be labor intensive requiring specialized laboratory technicians to perform the wet lab work, e.g., cell staining, and run the equipment.

[0004] There is a benefit to improving the system and method for cell quantification.Summary

[0005] An exemplary system and method are disclosed for a microfluidic platform for on-chip apoptosis quantification using an impedance-based instrumented quantification chamber that can provide a direct measure of the subpopulation with apoptosis cells within a cell population as the cells flow through the instrumented quantification chamber in the analysis. The exemplary system and method can accurately and quickly detect apoptotic markers of a cell in a stream of cells without disturbing cellular conditions to provide near real-time measure of apoptotic cell subpopulation and with minimal wet-lab processing. Insome embodiments, the binding agent is Annexin V (e.g., biotinylated annexin- V) or another in-line apoptosis or immunoassay configured, as an apoptosis marker, to bind to an externalized phosphatidylserine (PS) or other phospholipid that translocate itself from an inner to an outer leaflet of the plasma membrane due to apoptosis.

[0006] The instrumented quantification chamber includes (i) structures in the chamber functionalized with a coating comprising a biotin-binding reagent (e.g., NeutrAvidin) configured to bind the binding reagent and bounded apoptotic cells in the instrumented quantification chamber and (ii) instrumentation (e.g., embedded electrodes) at the entrance and exist of the chamber to provide cell count of apoptotic cells or non-apoptotic cells in a given cell population under analysis.

[0007] In another aspect, the exemplary system and method may be employed with a machine learning algorithm trained to recognize patterns and distinguish between the different signal characteristics of live and dead cells using training data acquired from the exemplary microfluidic platform and impedance-based instrumented quantification method. The algorithm can classify via capture-chamber-less electrode-based measurements the cells as either live or dead based on the patterns it has learned during the training phase acquired with use of the capture chamber. Impedance-based quantification may include impedance spectroscopy.

[0008] In some aspects, the techniques described herein relate to a method including: treating a pre-treated cell population (e.g., pre-treated cells with a therapeutic, reagent, or condition to invoke programmed cell death) with a binding reagent (e.g., Annexin V) configured to bind to apoptotic cells in the cell population to form a treated cell population; flowing the treated cell population through a microfluidic device including, including through, a functionalized chamber configured with a plurality of functionalized surfaces each coated to bind with the binding reagent and combined cells bound by binding reagents to maintain apoptotic subpopulation in the chamber; and measuring, via a first electrode located at an inlet of the functionalized chamber and a second electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, to evaluate a pre-treatment applied to the pre-treated cell population.

[0009] In some aspects, the techniques described herein relate to a method further including: mixing the cell population and the binding reagent in a mixing space (e.g., continuous flow mixing chamber, e.g., serpentine or looped shaped mixing chamber).

[0010] In some aspects, the techniques described herein relate to a method, wherein the functionalized chamber includes protrusion structures (e.g., pillars) configured with a portion of the functionalized surfaces.

[0011] In some aspects, the techniques described herein relate to a method, wherein the protrusion structures includes (i) a first subset of protrusion structures located at a first region in the functionalized chamber, the first subset of protrusion structures having a first density arrangement, and (ii) a second subset of protrusion structures located at a second region in the functionalized chamber downstream to the first region, the second subset of protrusion structures having a second density arrangement, wherein the second density arrangement is higher the first density arrangement.

[0012] In some aspects, the techniques described herein relate to a method, wherein the binding reagent includes biotinylated Annexin V and an Annexin V binding buffer.

[0013] In some aspects, the techniques described herein relate to a method, wherein the binding reagent is an immunoassay probe (e.g., protein such as antibody, antigen, etc.).

[0014] In some aspects, the techniques described herein relate to a method, including: analyzing (e.g., using a mass balance equation) the set of measurement signals to determine at least one of (i) the number of cells entering the inlet to the functionalized chamber, (ii) the number of cells exiting at the outlet of the functionalized chamber, (iii) the number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, and (iv) a combination thereof.

[0015] In some aspects, the techniques described herein relate to a method, including: flowing the treated cell population through a second functionalized chamber located downstream to the functionalized chamber, wherein the second functionalized chamber is configured with a second plurality of functionalized surfaces each coated to bind a second binding reagent (e.g., antibodies, aptamers, DNA probes, engineered proteins, etc.) different from the binding reagent; and measuring, via a third electrode located at an inlet of the second functionalized chamber and a fourth electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the second functionalized chamber and a number of cells exiting at the outlet of the second functionalized chamber to determine a number of cells having correspondence to the second binding reagent.

[0016] In some aspects, the techniques described herein relate to a method, including: flowing the treated cell population through a third functionalized chamber located in parallel to the functionalized chamber, wherein the third functionalized chamber is configured with a third plurality of functionalized surfaces each coated to bind a third binding reagent (e.g., antibodies, aptamers, DNA probes, engineered proteins, etc.) different from the binding reagent; and measuring, via a fifth electrode located at an inlet of the third functionalized chamber and a sixth electrode located at an outlet of the third functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the third functionalized chamber and a number of cells exiting at the outlet of the third functionalized chamber to determine a number of cells having correspondence to the third binding reagent.

[0017] In some aspects, the techniques described herein relate to a microfluidic device including: a substrate having a device inlet configured to receive a plurality of pretreated cells; a functionalized chamber formed in the substrate, the functionalized chamber being in fluidic communication with the device inlet, the functionalized chamber configured with a plurality of functionalized surfaces each coated to bind (i) an apoptotic binding reagent and (ii) the apoptotic binding reagent when bounded to apoptotic cells of the pre-treated cells, to maintain apoptotic subpopulation in the functionalized chamber; and a set of electrodes positioned in the substrate, wherein the set of electrodes includes (i) a first electrode located at an inlet of the functionalized chamber and (ii) a second electrode located at an outlet of the functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population.

[0018] In some aspects, the techniques described herein relate to a microfluidic device including: a mixing channel defined in the substrate between the device inlet and the functionalized chamber, the mixing channel having at least a first channel portion having a first flow direction, a second channel portion having a second flow direction, and a third channel portion having a third flow direction.

[0019] In some aspects, the techniques described herein relate to a microfluidic device, wherein the first channel portion, the second channel portion, and the third channel portion forms a serpentine or meandering channel structure.

[0020] In some aspects, the techniques described herein relate to a microfluidic device, wherein the functionalized chamber includes protrusion structures (e.g., pillars) configured with a portion of the functionalized surfaces.

[0021] In some aspects, the techniques described herein relate to a microfluidic device, wherein the protrusion structures includes (i) a first subset of protrusion structures located at a first region in the functionalized chamber, the first subset of protrusion structures having a first density arrangement, and (ii) a second subset of protrusion structures located at a second region in the functionalized chamber downstream to the first region, the second subset of protrusion structures having a second density arrangement, wherein the second density arrangement is higher the first density arrangement.

[0022] In some aspects, the techniques described herein relate to a microfluidic device, wherein the first electrode and the second electrode each include an interdigitated arrangement of three metal structures (e.g., electrode pair and common electrode).

[0023] In some aspects, the techniques described herein relate to a microfluidic device, wherein the device includes a filter for sample purification.

[0024] In some aspects, the techniques described herein relate to a microfluidic device including: a second functionalized chamber formed in the substrate, the second functionalized chamber being in fluidic downstream communication with the functionalized chamber, the second functionalized chamber configured with a second plurality of functionalized surfaces each coated to bind a second binding reagent to maintain a second cell subpopulation associated with the second binding reagent in the functionalized chamber; and a second set of electrodes positioned in the substrate, wherein the second set of electrodes includes (i) a third electrode located at an inlet of the second functionalized chamber and (ii) a fourth electrode located at an outlet of the second functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the second functionalized chamber and a number of cells exiting at the outlet of the second functionalized chamber to determine a number of cells associated with the second binding reagent.

[0025] In some aspects, the techniques described herein relate to a microfluidic device including: a third functionalized chamber formed in the substrate, the third functionalized chamber being in fluidic parallel communication with the functionalized chamber, the third functionalized chamber configured with a third plurality of functionalized surfaces each coated to bind a third binding reagent to maintain a third cell subpopulation associated with the third binding reagent in the third functionalized chamber; and a third setof electrodes positioned in the substrate, wherein the third set of electrodes includes (i) a fifth electrode located at an inlet of the third functionalized chamber and (ii) a sixth electrode located at an outlet of the third functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the third functionalized chamber and a number of cells exiting at the outlet of the third functionalized chamber to determine a number of cells associated with the third binding reagent.

[0026] In some aspects, the techniques described herein relate to a microfluidic device, wherein the set of electrodes are coupled to an instrument for impedance spectroscopy. In some aspects, the techniques described herein relate to a method, including: providing a pre-treated cell population; flowing the treated cell population through a microfluidic device comprising, including through, a functionalized chamber configured with a plurality of functionalized surfaces (e.g., immobilized peptide ligands, protein scaffolds, or antibodies) each coated to bind to the pre-treated cell; and measuring, via a first electrode located at an inlet of the functionalized chamber and a second electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, to evaluate a pre-treatment applied to the pre-treated cell population.Brief Description of Drawings

[0027] Fig. 1 shows a microfluidic platform system for on-chip apoptosis quantification using an impedance-based instrumented quantification chamber that can provide a direct measure of subpopulation with apoptosis cells within a cell population as the cells flow through the instrumented quantification chamber in the analysis, in accordance with an illustrative embodiment.

[0028] Fig. 2 shows an example method of operation for the microfluidic platform of Fig. 1, in accordance with an illustrative embodiment.

[0029] Figs. 3A - 3C each shows an example microfluidic platform employing a microfluidic layer coupled with an electrical sensor network, wherein the microfluidic layer accommodates inlets (e.g., cell sample inlet, solution inlet), a mixer, and a cell capture chamber in accordance with an illustrative embodiment. The exemplary microfluidic platform can further incorporate impedance spectroscopy, machine learning (ML) model, andimmobilization method. Fig. 33 shows the exemplary platform having an electrical sensor network configured with impedance spectroscopy. Fig. 3B shows the exemplary platform employing an ML model, wherein the ML model is trained on cell signals generated from the exemplary platform to perform live-dead analysis without needing a capture chamber. Fig. 3C shows the exemplary platform using an immobilization method for buffer solution (e.g., annexin- V), antibodies, and engineered peptide ligands.

[0030] Fig. 3D shows an example instrumentation for the on-chip apoptosis quantification microfluidic platform, in accordance with an illustrative embodiment.

[0031] Fig. 3E shows an example configuration of the sensor / electrodes, in accordance with an illustrative embodiment.

[0032] Figs. 4A - 4G each shows an example platform employing a microfluidic layer with an electrical sensor network, wherein the microfluidic layer includes a single inlet, multiple parallel cell capture chambers, and individual outlets for each chamber in accordance with an illustrative embodiment. The exemplary microfluidic platform can further incorporate impedance spectroscopy, ML model, live-dead assay, sample purification unit, and particle separation unit. Fig. 4A shows the exemplary platform. Fig. 4B shows all possible combinations of cells that can be formed from the three markers through a Venn diagram. Fig. 4C shows the exemplary platform employing impedance spectroscopy and a machine learning (ML) model to perform immunophenotyping. Fig. 4D shows an example optical detection-based module incorporating specific antibodies or antibody cocktails that may be integrated into the exemplary platform. Fig. 4E shows an example live-dead assay integration into the exemplary platform. Fig. 4F shows the exemplary platform employing a sample purification unit (e.g., a small particle or cell lysate separation unit. Fig. 4G shows the exemplary platform employing an electrical sensor-based unit coupled with a particle separation unit, live-dead assay unit, and immunoassay unit to analyze cell secretomes.

[0033] Figs. 5A - 5K show a fabricated microfluidic platform and its performance evaluation in apoptosis quantification in various conditions and environments. Fig. 5A shows a fabricated microfluidic platform employing a microfluidic layer coupled with an electrical sensor network. Fig. 5B shows the brightfield microscope images and mixing index (MI) for three different micromixer types (e.g., serpentine, triangular, linear) at a 100 pL / hr flow rate. Fig. 5C shows the brightfield microscope images and mixing index of the mixing units (e.g., 1st, 20th, 40th, 50th) using color dyes at a 100 pL / hr flow rate. Fig. 5D shows the mixing index of the micromixers at various angles (e.g., 5°, 10°, 15°, 20°) and flow rates (50 pL / hr, 100 pL / hr, 150 pL / hr, 200 pL / hr). Fig. 5E shows the fluorescence microscopy image of thefabricated platform with a neutravidin-functionalized capture chamber. Fig. 5F shows the impact of annexin-V concentration on the capture rate of apoptotic cells. Fig. 5G shows the apoptotic cell capture rate at various flow rates and the saturation point of the fabricated platform when passing the annexin-V solution. Fig. 5H shows the capture rate of live and apoptotic Jurkat cells and flow cytometer measurement thereof. Fig. 51 shows the externalization of phosphatidylserine (PS) using the fabricated platform and flow cytometry for the heat-induced apoptotic sample at 30 minutes, 1 hour 30 minutes, and 6 hours 30 minutes after initial treatment. Fig. 5J shows the frequencies of viable and apoptotic Jurkat cell subpopulations determined by the fabricated platform compared to nominal mix ratios (e.g., 1 : 1, 1:2, 2: 1) determined by a hemocytometer. Fig. 5K shows the quantification (i.e., capture rate) of live and apoptotic cells obtained from the fabricated platform and the fluorescence-based flow cytometer.Detailed Description

[0034] Some references, which may include various patents, patent applications, and publications, are cited in a reference list and discussed in the disclosure provided herein. The citation and / or discussion of such references is provided merely to clarify the description of the disclosed technology and is not an admission that any such reference is “prior art” to any aspects of the disclosed technology described herein. In terms of notation, “[n]” corresponds to the nth reference in the list. For example, [1] refers to the first reference in the list. All references cited and discussed in this specification are incorporated herein by reference in their entirety and to the same extent as if each reference was individually incorporated by reference.

[0035] Example Systems

[0036] Fig. 1 shows a microfluidic platform system 100 (shown as 100a) for on-chip apoptosis quantification using an impedance-basedinstrumented quantification chamber that can provide a direct measure of subpopulation with apoptosis cells within a cell population as the cells flow through the instrumented quantification chamber in the analysis, in accordance with an illustrative embodiment.

[0037] In the example shown in Fig. 1, the system 100a includes a an on-chip apoptosis quantification microfluidic platform 101 that operates with data acquisition instrumentation 103. The on-chip apoptosis quantification microfluidic platform 101 (shown as 101’) includes a mixing channel 102 (shown as “Micromixer” 102), a functionalized chamber 104 (shown as “Capture Chamber” 104), and a set of electrodes 106 (shown as106a, 106b). The system substrate 201 (not shown, see Fig. 2) is formed of polydimethylsiloxane (PDMS) having one or more inlets 110 (e.g., 110a) to receive a plurality of cells (e.g., pre-treated cells treated with some stimuli of interest). Other substrate materials may be used, e.g., glass, plastics, and the like. The data acquisition instrumentation 103 is configured to interrogate the on-chip apoptosis quantification microfluidic platform 101 and provide recording and network interface capabilities. In some embodiments, the data acquisition instrumentation 103 includes a graphical user interface.

[0038] To improve the capture efficiency, the functionalized chamber 104 may be formed, as shown in Fig. 1, in the substrate 202 in an elongated serpentine-shaped channel and is in direct or indirect fluidic communication with the device inlet 110. The functionalized chamber 104 may be configured with a plurality of structures 112 having functionalized surfaces 114 (not shown, see Fig. 2) each coated to bind in combination with a binding buffer (i) an apoptotic binding reagent 116 and (ii) the bounded apoptotic binding reagent (as bounded to the apoptotic cells), to maintain apoptotic subpopulation in the functionalized chamber 104. The structure of the elongated serpentine-shaped channel and additional functionalized structures can facilitate the retention or capture of the apoptotic binding reagent 116 and (ii) the bounded apoptotic binding reagent as bound to the apoptotic cells in the flow.

[0039] In some embodiments, and as shown in Fig. 1, the microfluidic platform 100a includes a second device inlet 110 (shown as 100b) to receive buffer solution. The inputted cell population and a binding reagent (e.g., apoptotic binding reagent 116) can mix in the mixer channel 102 having at least a first channel portion having a first flow direction, a second channel portion having a second flow direction, and a third channel portion having a third flow direction to sufficiently faciliate mixing of the cell population with the binding reagent (e.g., apoptotic binding reagent 116, e.g., biotinylated annexin- V). The bounded apoptotic binding reagent (as bounded to the apoptotic cells) can be capture by the functionalized chamber 104 (e.g., having neutravidin to capture the binding reagent).

[0040] The set of electrodes 106 (shown as first electrode 106a (shown as “Inlet Sensor” 106a” and second electrode 106b (shown as “Outlet Sensor” 106b)) are positioned in the substrate 202 and are located at an inlet 118 and outlet 120 of the functionalized chamber 104. The electrodes 106a, 106b are configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treatedcell population. Fig. IF shows an example electrode 106 configured with an interdigitated arrangement of three metal structures (e.g., electrode pair and common electrode).

[0041] In some embodiments, the electrode 106 includes interdigitated electrode pairs forming sensors in which each pairs include positive and negative output electrode fingers, acting as current sinks, while a common input electrode serves as the current source. The electrodes may be spaced apart, extending throughout the sensor's length and forming pairs with the output electrodes (e.g., 5 pm wide and spaced apart by 5 pm).

[0042] Each electrode may have 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 pairs of positve and negative fingers to produce a waveform with multiple peaks of varying polarities. In some embodiments, the electrode may have more than 24 pairs, e.g., between 25 and 32 pairs, or between 25 and 64 pairs.

[0043] Example Method of Operation

[0044] Fig. 2 shows an example method 200 of operation for the microfluidic platform 100a of Fig. 1. Method 200 includes treating a pre-treated cell population with a binding reagent (e.g., probe, e.g., Annexin V) configured to bind to apoptotic cells in the cell population to form a treated cell population.

[0045] Pre-treated cells refer to a cell population 204, e.g., treated with a therapeutic, reagent, or condition to interest to potentially invoke programmed cell death. In some embodiments, the cells may be pre-disposed for cell death but are treated with a therapeutic or reagent to prevent or reduce with statistical significance the degree of apoptosis in the measured sub-population, e.g., as a way to evaluate, screen, or test the therapeutic or reagent.

[0046] To perform flow cytometry without individual cell treatment, method 200 includes flowing the treated cell population 202 through a microfluidic device (e.g., 100a, 100b, 100c) comprising, including through, a functionalized chamber (shown as “Capture Chamber” 104a) configured with a plurality of functionalized surfaces 206 (shown as 206a, 206b) each coated to bind with the binding reagent and combined cells bound by binding reagents to maintain apoptotic subpopulation in the chamber. In Fig. 2, the functionalized surfaces 206a include wall surfaces of the chamber 104a as well as protrusion structure surfaces that are disposed within the chamber 104a.

[0047] The functionalized capture chamber is designed to expect pre-treated cells that have already been bound to apoptotic binding reagent (biotinylated Annexin V). In the stage preceding the capture chamber (i.e. the mixer), the pre-treated cells are exposed to the binding reagent (biotinylated Annexin V) that is suspended within a buffer (e.g., apoptotic binding reagent). The arrival of the buffer (binding reagent) into the capture chamber is of nomeaningful consequence and may play no direct role in the cell capture function if it had not bound to an apoptotic cell.

[0048] To improve the capture efficiency, in Fig. 2, the protrusion structures (e.g., pillars) includes (i) a first subset of protrusion structures located at a first region in the functionalized chamber, the first subset of protrusion structures having a first density arrangement, (ii) a second subset of protrusion structures located at a second region in the functionalized chamber downstream to the first region, the second subset of protrusion structures having a second density arrangement, wherein the second density arrangement is higher the first density arrangement, as well as a third subset of protrusion structures located at a third region in the functionalized chamber downstream to the second region, the third subset of protrusion structures having a third density arrangement, wherein the third density arrangement is higher the second density arrangement. The first zone (e.g., Stage 1) may have a vertical spacing of 50 pm and a horizontal spacing of 160 pm, the second zone (e.g., Stage 2) may have a vertical spacing still at 50 pm and a lower horizontal spacing of 50 pm, and the third zone (e.g., Stage 3) may have pillars spaced 15 pm apart horizontally. The configuration, in this manner, may be optimized to facilitate interactions with cells ranging from 11 to 15 pm in diameter, ensuring effective engagement with the slender structures in which the first zone may capture unbound binding reagents, and the second and third zones capturing bounded apoptotic cells bound to the binding reagents.

[0049] Method 200 then includes measuring, via a first electrode located at an inlet of the functionalized chamber and a second electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, to evaluate a pre-treatment applied to the pre-treated cell population.

[0050] Cell death is a process that plays a crucial role in various physiological and pathological events in multicellular organisms. Understanding the mechanisms and different types of cell death can be beneficial in deciphering complex biological processes, studying disease pathogenesis, and developing therapeutic strategies. Cell death can be classified into several main types, each with distinct characteristics and underlying mechanisms. Among the various types of cell death, apoptosis is a well-known and extensively studied form. Unlike necrosis, apoptosis is a controlled and regulated process that occurs in response to specific physiological cues or pathological stimuli. It plays a role in various biological processes,including embryonic development, tissue homeostasis, and immune system regulation.Necrosis, a commonly recognized type of cell death, is characterized by cell swelling, plasma membrane rupture, and the release of cellular contents, often leading to inflammation and tissue damage. Autophagy, another recognized type of cell death, is a process that involves the degradation and recycling of cellular components in response to stress or nutrient deprivation. Additionally, emerging types of cell death, such as ferroptosis and necroptosis, have been identified and recognized as distinct forms with unique molecular mechanisms.

[0051] The exemplary system and method may be employed with binding reagents and buffers for apoptosis. The process of apoptosis includes the externalization of phosphatidyl serine (PS), a phospholipid normally located in the inner leaflet of the plasma membrane [1], During apoptosis, phosphatidylserine is translocated from the inner to the outer leaflet of the plasma membrane, making it accessible to extracellular proteins. The externalization of phosphatidylserine acts as a signal to other cells, facilitating the recognition and clearance of apoptotic cells by neighboring cells or phagocytes. The detection of phosphatidylserine externalization is essential for studying and quantifying apoptosis in biological samples. One commonly used approach is the Annexin V-based apoptosis assay. Annexin V is a calcium-dependent phospholipid-binding protein with a high affinity for phosphatidylserine. In the presence of calcium ions, Annexin V can specifically bind to phosphatidylserine exposed on the outer surface of apoptotic cells. By labeling Annexin V with fluorescent dyes or enzymes, different detection methods have been employed, including flow cytometry, fluorescence microscopy, or enzyme-linked immunosorbent assays (ELISAs), to visualize and quantify apoptotic cells based on PS externalization.

[0052] Apoptosis assays based on Annexin V, as employed in other systems, have revolutionized research and clinical applications by providing tools for investigating cell death pathways, evaluating drug effectiveness, and diagnosing disorders associated with abnormal cell death, contributing to the understanding of apoptosis dynamics and regulatory mechanisms, enabling researchers and clinicians to unravel cell fate decisions and pinpoint potential targets for therapeutic intervention. However, to drive further progress and meet the demands of early disease detection and cell manufacturing capacity, in-line apoptosis assays have emerged as a need to improve the understanding of the intricate mechanisms and dynamic aspects of cell death. The assays may be employed in the assessment and efficacy of therapeutic interventions, e.g., to improve the detection of diseases at an early stage.

[0053] Conducting in-line apoptosis assays can present challenges. Accurate detection of apoptotic markers without disturbing cellular conditions can require highlysensitive and specific techniques, while distinguishing apoptosis from other cellular processes poses additional challenges. Integrating the assay into microfluidic or lab-on-a-chip devices also presents engineering and technical hurdles. These are addressed by the exemplary systems and methods.

[0054] In using an apoptosis assay that focuses on phosphatidylserine externalization as a reliable apoptotic marker, the exemplary system and method can provide insights into apoptotic events and other processes when used with other assays and associated binding buffer, e.g., for evaluation of cell differentiation, development, signaling, and activation. Assessment of phosphatidylserine externalization can provide direct insights into diseases such as cancer, inflammation, immune response, and neurodegenerative disorders, among others. By integrating in-line apoptosis assays with the analysis of phosphatidylserine externalization, researchers can gain a comprehensive and multifaceted understanding of cellular dynamics and disease mechanisms, e.g., in the development of therapeutic interventions and diagnostics.

[0055] In Fig. 2, the method 200 is shown performed at a microfluidic layer coupled with an electrical sensor network, where the microfluidic layer includes inlets (e.g., cell sample inlet, solution inlet), a mixer, and a cell capture chamber. The method 200 includes receiving cells via the cell sample inlet 110a and buffer solution (e.g., annexin- V) via the solution inlet 110b. Through the integrated mixer 102 (i.e., micromixer), the cells and buffer solution (e.g., biotinylated annexin- V) may be thoroughly and uniformly mixed, to ensure a consistent distribution of the buffer solution among the cells. During apoptosis, phosphatidylserine (PS) may be externalized on the cell surface, indicating the apoptotic state. The buffer solution (e.g., biotinylated annexin- V) may bind to the phosphatidylserine on the cell surface, facilitating the identification and analysis of apoptotic cells.Subsequently, the mixture of cells and buffer solution (e.g., biotinylated annexin- V) flows through the cell capture chambers 104. Within the chambers 104, the cells encounter neutravidin-coated pillars (shown as 206b. As a result, the population of cells that bind to buffer solution (e.g., biotinylated annexin- V) may become immobilized upon interaction with the neutravidin coating pillars. The immobilization step facilitates a more focused analysis and characterization of the apoptotic cells in separating apoptotic cells from non-apoptotic cells.

[0056] In Fig. 2, the exemplary platform incorporates an electrical sensor network comprising an inlet sensor 106a and an output sensor 106b positioned at the entrance and exit of the capture chamber 104, to facilitate precise monitoring of cell movement. The networkof sensors positioned at each of the capture chamber may track cell flow at specific locations within the microchip.

[0057] By observing the entry and exit of cells in the capture chamber 116, the exemplary platform may use, in some embodiments, a mass balance equation to determine the count of apoptotic cells within the chamber 104. The electrical sensor network may include interdigitated electrode pairs forming sensors in which each pair include positive and negative output electrode fingers, acting as current sinks, while a common input electrode serves as the current source. The electrodes may be spaced apart, extending throughout the sensor's length and forming pairs with the output electrodes (e.g., 5 pm wide and spaced apart by 5 pm). Each sensor (e.g., 106a, 106b, e.g., consists of 15 pairs, etc.) may produce a waveform with multiple peaks of varying polarities. The unique waveform pattern can provide a distinct code for each sensor. The specific sequence of polarities in the waveform corresponds to the physical arrangement of the input-output electrode finger pairs, giving rise to a distinctive waveform shape. Leveraging the waveform uniqueness, the exemplary platform may employ advanced signal processing techniques to accurately identify the sensor responsible for generating a particular waveform. By utilizing the distinct waveform patterns generated by each sensor, the exemplary platform may achieve precise localization and identification of cells within the capture chamber, providing an accurate assessment of the populations of apoptotic cells.

[0058] Microfluidic pathways of the exemplary platform may accommodate the target cell population (e.g., Jurkat cells, which are associated with T-cell leukemia). The capture chamber stands out with dimensions of 145 mm in length and 1.5 mm in width and can capture cells having an average diameter of approximately 11 - 14 pm (e.g., Jurkat cells or other lines of human T-lymphocyte cells or B- lymphocyte cells.

[0059] To facilitate optimal cell capture, the cell capture chamber (e.g., 104) may include an arrangement of pillars 206b organized in distinct stages with different placements. The placement of the three stages of pillars may serve multiple purposes. Firstly, the placement of the pillar stages may prevent the clogging of immobilized cells at the beginning of the capture chamber. Secondly, the placement of the pillar stages may help the exemplary platform avoid saturation caused by the buffer solution. The placement of the pillar stages may ensure effective cell capture and analysis without compromising the performance of subsequent stages, maintaining the capture process's effectiveness and preventing potential saturation or interference in subsequent stages.

[0060] The exemplary platform may also employ buffer solution-capturing prechamber (not shown, see Fig. 4F) to capture any excess amounts of buffer solution. The prechamber may provide a flow rate that is high enough to prevent the capture of cells while also being slow enough to capture the unbound buffer solution. The design of the pre-chamber may be tailored for a balance between flow rate and capture efficiency, allowing for the selective capture of unbound buffer solution molecules without interfering with the flow of cells.

[0061] By capturing excess buffer solution in the earlier stages, the exemplary platform may maximize the binding capacity and interaction between cells and the neutravidin-coated pillars, while also ensuring the pillars remain available for capturing apoptotic cells. Thirdly, the placement of the pillar stages may maximize the interaction between cells and the neutravidin-coated pillars, while providing sufficient clearance to prevent clogging.

[0062] Furthermore, the pillars not only aid in cell capture but also provide structural support to the chamber ceiling, preventing any potential collapse. The capture chamber may be equipped with auxiliary functionalization ports and channels (not shown; see Fig. 1) that are located near the inlet and outlet of each chamber. In the example shown in Fig. 1, the auxiliary channels 122 may be configured with diamond-shaped micropillar arrays that can serve as efficient particulate filters (e.g., for chemical filtration), e.g., connected to a port 124, e.g., to prevent the entry of pollutants during the multi-step functionalization process, and also ensuring that there is no coagulation or precipitation caused by the sequential infusion of different chemical reagents.

[0063] The setup may enable the delivery of functionalization reagents to the desired chamber, thereby maintaining the integrity of the assay. Once the functionalization process is complete, the auxiliary ports 124 may be sealed to prevent any leakage during assay operations. The microchip may have interfaces that include one or more fluidic inlets and a single outlet 126, providing the necessary connections for smooth fluid flow and comprehensive analysis. In some embodiments, rather than an outlet, a reservoir is included (not shown).

[0064] Impedance Spectroscopy Integration

[0065] Fig. 3 A shows an example platform 100 (shown as 100b) having an electrical sensor network configured with impedance spectroscopy. Impedance spectroscopy may be used for characterizing cells by probing their electrical properties and behaviors across a range of frequencies. Cells, being complex biological entities, exhibit distinct electricalcharacteristics that can be analyzed using impedance measurements. In Fig. 3 A, continuous temporal or frequency-swept signals 302 (shown as 302a, 302b) may be introduced at both the inlet and outlet sensors, providing a way to examine the platform’s behavior across a wide range of frequencies. By capturing the output signals from both sensors, the platform's transfer function may be analyzed and determined, wherein the transfer function provides insights into how the system modifies or responds to input signals that are temporally or spectrally dynamic.

[0066] The analysis of the characteristics of the output signals (e.g., temporal, spatial, and frequency profiles) from both the inlet and outlet sensors may provide the detection of various types of dead cells, including those undergoing necrosis or autophagy. Measurements of this detail may provide a binary live / dead determination at which point along the wide transitory span between healthy and dead that the interrogated cells are found to be in. By incorporating this multi-domain electrical profile, the exemplary platform may achieve a comprehensive and accurate live-dead assay.

[0067] Example Machine learning (ML) and artificial intelligence (Al) integration.

[0068] Fig. 3B shows an example platform employing an ML model, wherein the ML model is trained on cell signals generated from the exemplary platform to perform live-dead analysis without the need for a capture chamber. In Fig. 3B, the exemplary platform 310 may employ electrical sensors (e.g., 106a, 106b) positioned at the beginning and exit of the capture chamber to acquire training data for a training system 307. The first sensor 106a may capture signals from all cells entering the chamber, while the second sensor 106b may record signals from live cells exiting the capture chamber. By training a machine learning model 309 on the signals generated by the exemplary platform, which include signals from both live and a mixture of live-dead cells, the exemplary platform may provide a live-dead analysis without the need for a capture chamber, shown as system 100c.

[0069] The method (e.g., for measuring cell population in a rapid cell characterization analysis) may include flowing a pre-treated cell population across an electrode having a plurality of sensing surfaces (sets of electrode pair and common electrode) to form a measurement signal having measurement waveform defined by observations from each of the plurality of sensing surfaces; and determining, via a trained Al model (e.g., 309), using the measurement signal, a score value of a cell of the pre-treated cell population being alive or the pre-treated cell population being dead or programmed for death, where the trained Al model was trained using training data generated using the on-chip apoptosis quantification microfluidic platform 101.

[0070] The machine learning model can be trained to recognize patterns and distinguish between the different signal characteristics of live and dead cells. By feeding the model a dataset consisting of labeled signals from known live and dead cell populations, the model learns to differentiate between them based on the unique features present in the signals. These features may include amplitude, frequency distribution, waveform shape, or other relevant parameters that can be extracted from the impedance spectroscopy measurements. For example, a machine learning model such as a Support Vector Machines (SVM), Artificial Neural Network (ANN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM) Networks, Random Forests, Gradient Boosting Machines (GBM), Hidden Markov Models (HMM), Gaussian Mixture Models (GMM), Extreme Learning Machines (ELM), and Deep Belief Networks (DBN) can be trained using the labeled dataset. Once trained, the model can then be applied to analyze signals obtained from a similar type of cell population without the need for a capture chamber. The algorithm may classify the cells as either live or dead based on the patterns it has learned during the training phase.

[0071] Moreover, incorporating impedance spectroscopy alongside the machine learning model can further enhance the performance of the live-dead analysis. Impedance spectroscopy provides additional electrical properties and characteristics of the cells, which can be used as input features for the machine learning algorithm. By considering the impedance spectra in combination with the electrical signals from the sensors, the model can achieve a more comprehensive and accurate differentiation between live and dead cells.

[0072] In addition to apoptosis, the method may be employed to determine a score value of a cell of the pre-treated cell population having one of set of pre-defined phenotypes, where the trained Al model (e.g., 314) was trained using training data generated using the on- chip apoptosis quantification microfluidic platform 101, where at least one of the functionalized chamber was coated with a reagent (e.g., antibodies, aptamers, DNA probes, and engineered proteins.) associated with one of the set of pre-defined phenotypes.

[0073] Machine Learning. The trained Al model (e.g., 314) can be implemented using one or more artificial intelligence and machine learning operations. The term “artificial intelligence” can include any technique that enables one or more computing devices or computing systems (i.e., a machine) to mimic human intelligence. Artificial intelligence (Al) includes but is not limited to knowledge bases, machine learning, representation learning, and deep learning. The term “machine learning” is defined herein to be a subset of Al that enables a machine to acquire knowledge by extracting patterns from raw data. Machine learningtechniques include, but are not limited to, logistic regression, support vector machines (SVMs), decision trees, Naive Bayes classifiers, and artificial neural networks. The term “representation learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, or classification from raw data. Representation learning techniques include, but are not limited to, autoencoders and embeddings. The term “deep learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, classification, etc., using layers of processing. Deep learning techniques include but are not limited to artificial neural networks or multilayer perceptron (MLP).

[0074] Machine learning models include supervised, semi-supervised, and unsupervised learning models. In a supervised learning model, the model learns a function that maps an input (also known as feature or features) to an output (also known as target) during training with a labeled data set (or dataset). In an unsupervised learning model, the algorithm discovers patterns among data. In a semi-supervised model, the model learns a function that maps an input (also known as a feature or features) to an output (also known as a target) during training with both labeled and unlabeled data.

[0075] Neural Networks. An artificial neural network (ANN) is a computing system including a plurality of interconnected neurons (e.g., also referred to as “nodes”). This disclosure contemplates that the nodes can be implemented using a computing device (e.g., a processing unit and memory as described herein). The nodes can be arranged in a plurality of layers, such as an input layer, an output layer, and, optionally, one or more hidden layers with different activation functions. An ANN having hidden layers can be referred to as a deep neural network or multilayer perceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, each layer is made of a plurality of nodes, where each node is connected to all nodes in the previous layer. The nodes in a given layer are not interconnected with one another, i.e., the nodes in a given layer function independently of one another. As used herein, nodes in the input layer receive data from outside of the ANN, nodes in the hidden layer(s) modify the data between the input and output layers, and nodes in the output layer provide the results. Each node is configured to receive an input, implement an activation function (e.g., binary step, linear, sigmoid, tanh, or rectified linear unit (ReLU), and provide an output in accordance with the activation function. Additionally, each node is associated with a respective weight. ANNs are trained with a dataset to maximize or minimize an objective function. In some implementations, the objective function is a costfunction, which is a measure of the ANN’S performance (e.g., an error such as LI or L2 loss) during training, and the training algorithm tunes the node weights and / or bias to minimize the cost function. This disclosure contemplates that any algorithm that finds the maximum or minimum of the objective function can be used for training the ANN. Training algorithms for ANNs include but are not limited to backpropagation. It should be understood that an ANN is provided only as an example machine learning model. This disclosure contemplates that the machine learning model can be any supervised learning model, semi-supervised learning model, or unsupervised learning model. Optionally, the machine learning model is a deep learning model. Machine learning models are known in the art and are therefore not described in further detail herein.

[0076] A convolutional neural network (CNN) is a type of deep neural network that has been applied, for example, to image analysis applications. Unlike traditional neural networks, each layer in a CNN has a plurality of nodes arranged in three dimensions (width, height, and depth). CNNs can include different types of layers, e.g., convolutional, pooling, and fully-connected (also referred to herein as “dense”) layers. A convolutional layer includes a set of filters and performs the bulk of the computations. A pooling layer is optionally inserted between convolutional layers to reduce the computational power and / or control overfitting (e.g., by downsampling). A fully connected layer includes neurons, where each neuron is connected to all of the neurons in the previous layer. The layers are stacked similarly to traditional neural networks. GCNNs are CNNs that have been adapted to work on structured datasets such as graphs.

[0077] Transformer Models. A transformer is a deep learning architecture developed based on the multi-head attention mechanism. Text is converted to numerical representations referred to as tokens, and each token is converted into a vector, e.g., via a look-up from an embedding table. At each layer, each token is then contextualized within the scope of the context window with other (unmasked) tokens via a parallel multi-head attention mechanism, allowing the signal for key tokens to be amplified and less important tokens to be diminished. Transformers have tokenizers, which convert text into tokens; a single embedding layer, which converts tokens and positions of the tokens into vector representations; transformer layers, which carry out repeated transformations on the vector representations, extracting more and more linguistic information. These consist of alternating attention and feedforward layers. In some embodiments, the transformer may include an un-embedding layer, which converts the final vector representations back to a probability distribution over the tokens. Transformer layers can be one of two types: encoder and decoder.

[0078] Other Supervised Learning Models. A logistic regression (LR) classifier is a supervised classification model that uses the logistic function to predict the probability of a target, which can be used for classification. LR classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize an objective function, for example, a measure of the LR classifier’s performance (e.g., an error such as LI or L2 loss), during training. This disclosure contemplates that any algorithm that finds the minimum of the cost function can be used. LR classifiers are known in the art and are therefore not described in further detail herein.

[0079] A Naive Bayes’ (NB) classifier is a supervised classification model that is based on Bayes’ Theorem, which assumes independence among features (i.e., the presence of one feature in a class is unrelated to the presence of any other features). NB classifiers are trained with a data set by computing the conditional probability distribution of each feature given a label and applying Bayes’ Theorem to compute the conditional probability distribution of a label given an observation. NB classifiers are known in the art and are therefore not described in further detail herein.

[0080] A k-NN classifier is an unsupervised classification model that classifies new data points based on similarity measures (e.g., distance functions). The k-NN classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize a measure of the k-NN classifier’s performance during training. This disclosure contemplates any algorithm that finds the maximum or minimum. The k-NN classifiers are known in the art and are therefore not described in further detail herein.

[0081] Example Computing System. The exemplary system and method may be implemented (1) as a sequence of computer-implemented acts or program modules running on a computing system and / or (2) as interconnected machine logic circuits or circuit modules within the computing system. The implementation is a matter of choice dependent on the performance and other requirements of the computing system. Accordingly, the logical operations described herein are referred to variously as state operations, acts, or modules. These operations, acts, and / or modules can be implemented in software, in firmware, in special purpose digital logic, in hardware, and any combination thereof. It should also be appreciated that more or fewer operations can be performed than shown in the figures and described herein. These operations can also be performed in a different order than those described herein.

[0082] The computer system is capable of executing the software components described herein for the exemplary method or systems. In an embodiment, the computingdevice may comprise two or more computers in communication with each other that collaborate to perform a task. For example, but not by way of limitation, an application may be partitioned in such a way as to permit concurrent and / or parallel processing of the instructions of the application. Alternatively, the data processed by the application may be partitioned in such a way as to permit concurrent and / or parallel processing of different portions of a data set by the two or more computers. In an embodiment, virtualization software may be employed by the computing device to provide the functionality of a number of servers that are not directly bound to the number of computers in the computing device. For example, virtualization software may provide twenty virtual servers on four physical computers. In an embodiment, the functionality disclosed above may be provided by executing the application and / or applications in a cloud computing environment. Cloud computing may comprise providing computing services via a network connection using dynamically scalable computing resources. Cloud computing may be supported, at least in part, by virtualization software. A cloud computing environment may be established by an enterprise and / or can be hired on an as-needed basis from a third-party provider. Some cloud computing environments may comprise cloud computing resources owned and operated by the enterprise as well as cloud computing resources hired and / or leased from a third-party provider.

[0083] In its most basic configuration, a computing device includes at least one processing unit and system memory. Depending on the exact configuration and type of computing device, system memory may be volatile (such as random-access memory (RAM)), non-volatile (such as read-only memory (ROM), flash memory, etc.), or some combination of the two.

[0084] The processing unit may be a standard programmable processor that performs arithmetic and logic operations necessary for the operation of the computing device. While only one processing unit is shown, multiple processors may be present. As used herein, processing unit and processor refers to a physical hardware device that executes encoded instructions for performing functions on inputs and creating outputs, including, for example, but not limited to, microprocessors (MCUs), microcontrollers, graphical processing units (GPUs), and application-specific circuits (ASICs). Thus, while instructions may be discussed as executed by a processor, the instructions may be executed simultaneously, serially, or otherwise executed by one or multiple processors. The computing device may also include a bus or other communication mechanism for communicating information among various components of the computing device.

[0085] Computing devices may have additional features / functionality. For example, the computing device may include additional storage, such as removable storage and nonremovable storage, including, but not limited to, magnetic or optical disks or tapes. Computing devices may also contain network connection(s) that allow the device to communicate with other devices, such as over the communication pathways described herein. The network connection(s) may take the form of modems, modem banks, Ethernet cards, universal serial bus (USB) interface cards, serial interfaces, token ring cards, fiber distributed data interface (FDDI) cards, wireless local area network (WLAN) cards, radio transceiver cards such as code division multiple access (CDMA), global system for mobile communications (GSM), long-term evolution (LTE), worldwide interoperability for microwave access (WiMAX), and / or other air interface protocol radio transceiver cards, and other well-known network devices. Computing devices may also have input device(s) such as keyboards, keypads, switches, dials, mice, trackballs, touch screens, voice recognizers, card readers, paper tape readers, or other well-known input devices. Output device(s) such as printers, video monitors, liquid crystal displays (LCDs), touch screen displays, displays, speakers, etc., may also be included. The additional devices may be connected to the bus in order to facilitate the communication of data among the components of the computing device. All these devices are well known in the art and need not be discussed at length here.

[0086] The processing unit may be configured to execute program code encoded in tangible, computer-readable media. Tangible, computer-readable media refers to any media that is capable of providing data that causes the computing device (i.e., a machine) to operate in a particular fashion. Various computer-readable media may be utilized to provide instructions to the processing unit for execution. Example tangible, computer-readable media may include but is not limited to volatile media, non-volatile media, removable media, and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data. System memory, removable storage, and non-removable storage are all examples of tangible computer storage media. Example tangible, computer-readable recording media include, but are not limited to, an integrated circuit (e.g., field-programmable gate array or applicationspecific IC), a hard disk, an optical disk, a magneto-optical disk, a floppy disk, a magnetic tape, a holographic storage medium, a solid-state device, RAM, ROM, electrically erasable program read-only memory (EEPROM), flash memory or other memory technology, CD- ROM, digital versatile disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices.

[0087] In light of the above, it should be appreciated that many types of physical transformations take place in the computer architecture to store and execute the software components presented herein. It also should be appreciated that the computer architecture may include other types of computing devices, including hand-held computers, embedded computer systems, personal digital assistants, and other types of computing devices known to those skilled in the art.

[0088] In an example implementation, the processing unit may execute program code stored in the system memory. For example, the bus may carry data to the system memory, from which the processing unit receives and executes instructions. The data received by the system memory may optionally be stored on the removable storage or the non-removable storage before or after execution by the processing unit.

[0089] Immobilizing Annexin- V, Antibodies, and Engineered Peptide Ligands

[0090] Fig. 3C shows an exemplary system 100 (shown as lOOd) configured with the immobilization of annexin-V or aptamers in the capture chamber. Immobilizing buffer solution (e.g., annexin-V), aptamers, or other molecular probes onto the capture chamber may improve the performance of the apoptosis assay. The modification enables direct detection of apoptotic cells, providing valuable insights into the early stages of programmed cell death.

[0091] The engineered peptide ligands, protein scaffolds, or antibodies may be immobilized onto the capture chamber to specifically detect dead cells. By incorporating the alternative detection methods, the exemplary platform lOOd may eliminate the need for a micromixer and simplify the workflow by allowing the sample to be infused into the capture chamber. Immobilizing the appropriate molecular probes onto the capture chamber facilitates the binding and recognition of apoptotic or dead cells, and may enhance the sensitivity and accuracy of the assay. By expanding the detection option, the exemplary platform lOOd may provide a more comprehensive analysis of cell viability and cell death mechanisms. The approach can provide specificity and versatility, e.g., to target specific cellular markers associated with apoptosis or cell death.

[0092] Example Data Acquisition Instrumentation

[0093] Fig. 3D shows an example instrumentation (e.g., 103) for the on-chip apoptosis quantification microfluidic platform 101.

[0094] In Fig. 3D, cells 302 may be driven through the device at a constant flow rate with a syringe pump. An electrical sensor network (e.g., 106) may be excited using a sine wave 305 generated from the lock-in amplifier 306, and the resulting current signals 308A-B may be first converted to voltage signals 310A-B using transimpedance amplifiers 312, thensubtracted from each other by a differential amplifier 314 and the signal amplitude 316 is measured using a lock-in amplifier 306. The decoding process may then identify individual sensor signals in the device output signal 316. The output signal 316 may be correlated 318 with a template library 320 consisting of signature waveforms, such as 322A, corresponding to each and every coded sensor in the network using a custom-built algorithm 318. A correlation peak 324 may be used to identify the matching template and the specific sensor that detected the cell. The specific case in the figure demonstrates the decoding of a signal produced by the sensor with the Code (e.g., 2,3).

[0095] During an exemplary assay, the sample may be driven through the functionalized device by a syringe pump at a controlled flow rate and followed by a brief phosphate-buffered saline (PBS) wash to clear the device of remaining cells. The electrical signal 316 from the device may be acquired via electronic hardware and analyzed using a computer 326. To determine a capture location for each cell processed on the device, the output signal 316 may be processed using a decoding algorithm 318.

[0096] The algorithm may first review a part of the recorded electrical waveform, identify different code signals present, and classify them into different sensor groups. Once each sensor group contains a sufficient number of code signal instances, signals were normalized and averaged to form a library of code templates that correspond to each and every sensor in the network. The generation of templates based on recorded signals from the sample itself made the templates specific to both the sample and the device, thereby increasing accuracy. The templates may then be used to process all sensor data by correlating the output signal with the template library. Because the code signals are specifically designed to be mutually orthogonal, the system may classify the sensor signals robustly with minimal crosstalk but also resolve signal interferences through an iterative process called successive interference cancellation. At the end of this decoding process, the original output waveform may be decomposed into data from individual sensors, which was then used to calculate cell capture statistics across the whole device. Specifically, the number of captured cells in each chamber may be obtained, by subtracting the exit node cell count from the entry node cell count.

[0097] Fig. 3E shows an example configuration of the sensor / electrodes (e.g., 106). The sensing strategy may be based on the Microfluidic coded orthogonal detection by electrical sensing (CODES) scheme, which uses micromachined electrode patterns to multiplex spatiotemporal cell data across a microfluidic device. In Fig. 3E, a three-electrode Coulter counter may be shaped to form distinct electrode patterns (i.e., sensors) at sixdifferent nodes to monitor cell passage between microfluidic chambers. Each sensor may be composed of an array of 5 pm wide finger electrodes separated by 5 pm gaps (344) and may produce a specific 31 -bit digital code, which may be implemented by an interdigitated arrangement of three electrodes: two sensing electrodes (340, 342) to set the bit polarity (positive for " 1" and negative for "0") and one common electrode (344) meandering in between to excite the sensor network (340, 342). Cells flowing over one of the sensors may sequentially modulate the local impedance between adjacent finger electrodes via the Coulter principle and generate a distinct bipolar electrical waveform dictated by the surface electrode pattern. In addition, embodiments may use sensor codes designed to be mutually orthogonal (Gold sequences), and therefore, embodiments may 1) reliably discriminate sensor signals from each other in the output signal and 2) resolve interfer- ing signals when multiple cells are coincidentally detected by the same or different sensors. Moreover, in the case of cell debris or aggregates, the electrical signal generated by sensors does not match any of the templates constructed based on single cell signals, and therefore, these data are discarded and do not affect the assay performance.

[0098] Additional Microfluidic Platform

[0099] Fig. 4A shows an example platform employing a microfluidic layer with an electrical sensor network, wherein the microfluidic layer includes a single inlet, multiple parallel cell capture chambers, and individual outlets for each chamber. Alternatively, the outlets can be merged into a single outlet at the end of each chamber. The exemplary platform may conduct multiplex immunoassays.

[0100] The exemplary platform may intake a cell sample via the inlet, initiating the flow of cells through the parallel cell capture chambers. Within these chambers, the cells may encounter antibody-coated pillars. Each chamber may be coated with either a specific antibody or a combination of antibodies targeting a particular number of surface markers. Consequently, the target cell population carrying the specific target antigen may become immobilized upon interaction with the corresponding antibody-coated pillars within the capture chamber. This immobilization step may facilitate a focused analysis and characterization of the target cell population with specific surface markers.

[0101] As an illustrative example, the exemplary platform shown in Fig. 4A may conduct a simultaneous analysis of three markers (referred to as a 3-plex immunoassay). As shown in Fig. 4A, each chamber is coated with either a specific antibody or a combination of antibodies according to the distinct combinations formed from the three markers. Thecaptured cells in each chamber represent a group of different combinations of the three markers, which can be extrapolated using conventional set theory.

[0102] The initial step of immunoassay involves determining the total number of unique combinations that can be formed from the three markers. In this example, the number of unique combinations for the three markers is eight (i.e., 23 markers). Fig. 4B shows all the possible distinct combinations of three markers through a Venn diagram. Based on this, the number of parallel capture chambers of the exemplary platform (shown in Fig. 4A) should correspond to the number of distinct combinations. To monitor the number of captured cells in each chamber, an electrical sensor network may be placed as shown in Fig. 4A.

[0103] Table 1 shows the groups of multiple combinations of the three markers represented by each chamber, wherein the combinations may be represented through a Venn diagram shown in Fig. 4B. In Table 1, A, B, C denotes an antibody marked with A, B, C, respectively; Antibodies (.) denotes a group of multiple antibodies with which a chamber is coated; and (,)posand (,)negdenotes positive and negative state of an antibody, respectively.Table 1

[0104] As shown in Table 1, the first chamber (i.e., Chamber 1) of the exemplary platform may be left uncoated, acting as the negative control for the exemplary platform andensuring an equal or similar representation of cell arrivals in each chamber. To determine the fractions of cells expressing specific antigens or combinations of antigens, the remaining seven chambers (e.g., Chambers 2 - 8) in the microchip may represent a group of different combinations of the seven states of the three markers.

[0105] By solving linear equations using the known number of captured cells in these chambers as shown in Table 1, the exemplary platform may ascertain the proportions of cells expressing specific antigen or antigen combinations. This computational process may measure and quantify the proportions of cells that possess distinct antigenic profiles within the different combinations. Furthermore, the population expressing AnegBnegCneg(all negative) can be determined by counting the number of cells leaving all the antibody-coated chambers. This flexible concept can be scaled up to accommodate any number of markers for multiplex immunophenotyping.

[0106] Overall, the exemplary platform may incorporate distinct antibody-coated chambers to capture cells expressing various combinations of marker states. By analyzing the captured cells, the exemplary platform provides insights into complex immune cell populations and their phenotypic characteristics.

[0107] Machine learning with impedance spectroscopy integration. The exemplary platform incorporates an electrical sensor network, with each capture chamber equipped with an entry and exit sensor. The entry sensor captures signals from cells as they enter the respective capture chamber, while the exit sensor detects signals from cells leaving the antibody-coated chamber, representing cells that do not express the target antigen. By running a cell population through the exemplary platform, signals from both target antigenexpressing cells and cells that do not express the target antigen may be obtained to generate a large dataset.

[0108] This large dataset may be used to train a machine learning model to accurately distinguish between the different cell populations. To process electrical signals generated by the exemplary platform, a data processing pipeline may be developed to amplify, digitize, and analyze the waveform in real-time. At the front end, trans-impedance amplifiers are used to condition the signal, which is then processed by a lock-in amplifier to produce a raw signal stream for digitization and analysis on a computer. Once the signal is digitized, it may be processed by a signal processing operation (e.g., a suite of Python algorithms), beginning with the extraction of valid cell events from the raw stream, which may include idle sensor noise. This extraction may be achieved by applying a sliding window to compute the signal power and saving windows that exceed a calculated threshold. The threshold may bedetermined during a first few seconds of runtime (e.g., 5 seconds) by sampling several thousand windows, selecting a user-defined number of lowest-power windows (e.g., 100 windows), and averaging them to establish a noise floor power.

[0109] An empirically set threshold above the noise floor may then be used for cell signal detection. Each detected cell segment may be refined by discarding minor idle sections at the leading and trailing edges of the signal. This refinement may be accomplished by locating the first and last peaks, following these to their first inflection points, and trimming the segment accordingly. The refined segments may then be passed to an ML model for interpretation. Processing over a user-defined number of cells per second (e.g., 700 cells per second), this model may be designed with two cascaded convolutional neural networks (CNNs) optimized for signal processing. In the first network, a region proposal network (RPN) may be used to identify the temporal location of each cell event, estimating a bounding box and reporting cell speed. The second network, the sensor classification network (SCN), may classify each bounding box generated by the RPN, reporting the corresponding sensor identification (ID) for each waveform. Both networks may consist of four convolutional layers and may be designed to process waveforms normalized to a user-defined number of samples (e.g., 200 samples). The SCN may output three nodes, each representing one of the three sensors embedded in the microchip. The model outputs — cell event sensor ID and cell speed — may then be utilized by the chamber capture counting routine and the feedback control algorithm, respectively.

[0110] Alternative machine learning models, such as Support Vector Machines (SVMs), Artificial Neural Networks (ANNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) Networks, Random Forests, Gradient Boosting Machines (GBMs), Hidden Markov Models (HMMs), Gaussian Mixture Models (GMMs), Extreme Learning Machines (ELMs), and Deep Belief Networks (DBNs), may also be implementable for signal data extraction from the exemplary platform. In the final stage of the pipeline, the aggregated cell event sensor IDs may be mapped to their respective sensors.[OHl] To further enrich the dataset, impedance spectroscopy techniques may be integrated into the exemplary platform. Fig. 4C shows an example platform employing impedance spectroscopy and a machine learning (ML) model to perform immunophenotyping. Subpanel (a) shows the exemplary platform monitoring cell responses and generating impedance data in real time. Subpanel (b) shows an exploded view of the placement of the inlet, outlet, and sensor in the exemplary platform, wherein the platform utilizes the trained ML model to achieve precise and rapid immunophenotyping.

[0112] Impedance spectroscopy involves applying a small amplitude alternating current (AC) signal to the cell sample and measuring the resulting voltage response, by varying the AC signal frequency, the electrical properties of cells may be explored over a wide frequency range, as different electrical phenomena affect impedance at different frequencies. This may enable a comprehensive analysis of cell behavior. For instance, impedance magnitude and phase angle at specific frequencies may differ between cell subpopulations, reflecting variations in cellular composition and dielectric properties. Additionally, the impedance spectrum - capturing impedance across frequencies - may present distinct features that aid in distinguishing cell subpopulations. Statistical methods or pattern recognition algorithms may be employed by the ML model to extract and analyze these features to improve differentiation accuracy.

[0113] Continuous temporal or frequency-swept signals (not shown) are applied to both the inlet and outlet sensors, allowing for a thorough exploration of the system’s behavior across a broad frequency range. By capturing output signals from both sensors shown in subpanel (a), the system’s transfer function may be determined, providing valuable insights into the system’s response to temporally or spectrally dynamic input signals (e.g., continuous temporal, frequency-swept). To deepen this understanding, characteristics of output signals from the inlet and outlet sensors may be compared to analyze their temporal, spatial, and frequency profiles. This comprehensive examination may enable the gathering of detailed electrical impedance data, yielding more information about the captured cells and their immunophenotype. By employing impedance spectroscopy, the exemplary platform may achieve enhanced cell population characterization, enabling a more thorough understanding of their specific attributes.

[0114] Once the machine learning model is trained using this enhanced dataset, the immunoassay process may be simplified by utilizing a sensor network without the need for individual capture chambers, as shown in subpanel (b), offering a straightforward and efficient solution for conducting immunoassays based on the trained model. This advancement may provide faster and more accurate analysis of cell populations, paving the way for improved diagnostic and research applications.

[0115] Antibody cocktails and optical detection methods or other sensing modalities. The exemplary platform should immobilize specific antibodies or antibody cocktails targeting surface markers in microfluidic chambers or wells. Fig. 4D shows an example optical detection-based module incorporating specific antibodies or antibody cocktails that may be integrated into the exemplary platform.

[0116] The optical detection method begins when the sample is introduced into the module, allowing for incubation to facilitate the immobilization of the target cell population or molecules onto the antibody-coated surface. Subsequently, the module may be washed to remove any unbound sample components. In order to quantify the target cell population or detect the presence and expression level of specific molecules such as cytokines, interleukins, chemokines, and other signaling molecules, the optical detection method may involve capturing images of the capture chambers or wells both before and after washing the sample. These images may serve as the basis for obtaining precise measurements. By analyzing the captured images, the optical detection method may derive quantitative data regarding the target cell population or the levels of the specific molecules of interest as shown in Fig. 4D.

[0117] Moreover, the optical detection method can be performed in real-time by utilizing high-speed cameras or other techniques, including surface plasmon resonance (SPR), fluorescence spectroscopy, immunofluorescence microscopy, and so on. By leveraging optical detection methods, the exemplary platform may enhance the accuracy and efficiency of the measurements, facilitating a comprehensive understanding of the biological system under study. This optical detection may help the exemplary platform to simultaneously perform immunoassays for both cells and signaling molecules or molecules secreted by cells. This comprehensive method may enable a more thorough understanding of the target system and provide insights into the interaction between cells and their secreted molecules.

[0118] Live-dead assay integration. A combination of live and dead assay (i.e., live- dead assay) may be integrated into the exemplary platform, enabling a comprehensive analysis of cell populations by simultaneously providing information on cell viability and cell surface markers. The live-dead assay can either be focused on apoptosis detection or can encompass a viability assay incorporating impedance spectroscopy.

[0119] Fig. 4E shows an example live-dead assay integration into the exemplary platform. The integration can be achieved in two ways: (1) Strategic placement of the live- dead assay before the immunoassay, allowing for assessment of the overall viability of the entire sample (shown in subpanel a); (2) Strategic placement of the live-dead assay before each immunocapture chamber, enabling quantification of the viability of individual cell populations (shown in subpanel b). In both approaches, integration provides the additional advantage of removing unwanted dead or apoptotic cells from the subsequent immunoassay process. This integration enhances the reliability and accuracy of the immunoassay results byensuring that only viable cells are analyzed while simultaneously providing valuable viability data for comprehensive cell characterization.

[0120] Sample purification integration. An issue for both live-dead assays and immunoassays is the potential for false counting caused by cell debris, lysate, or non-cellular particles, which can compromise the accuracy of the assay results. To address or reduce this issue, a sample purification unit may be integrated into the exemplary platform.

[0121] Fig. 4F shows an example platform employing a sample purification unit (e.g., a small particle or cell lysate separation unit). As shown in subpanels (a) and (b), the purification unit is positioned before the subsequent assay unit or units, facilitating an accurate quantification of viability for cell populations. The purification unit may incorporate particle separation operations, such as DLD (Deterministic Lateral Displacement), Herringbone, inertial, or centrifugal microfluidic operations. By implementing this integration, the exemplary platform may separate and exclude cell debris and small particles from the subsequent assay analysis.

[0122] The particle separation operations may selectively isolate the target cells or analytes of interest, while eliminating unwanted debris or particles that may interfere with the assay measurements. This purification step may enhance the accuracy and reliability of the assay results by ensuring that only the desired cellular components are considered in the analysis. Furthermore, this integration can be applied to both the live-dead assay and the immunoassay platforms, improving the overall performance and specificity of the assays. This approach may provide more accurate and reliable data, enhancing the quality of the analysis and facilitating more precise interpretations of cellular viability or immune-related measurements.

[0123] Simultaneous integration of live-dead assay, immunoassay, cell secretome profiling, and Glucose-Lactose detection. The detection of cell secretomes, including cytokines, interleukins (IL), and growth factors, plays a role in understanding the various stages of cell proliferation, cell differentiation, and activation. Additionally, valuable information regarding cellular metabolism, energy status, and response to different conditions or treatments can be derived from monitoring glucose and lactose levels in the cell media. To further advance the exemplary platform’s analytical capabilities, an electrical sensor-based unit may be integrated with the exemplary live-dead assay and immunoassay platform. Fig. 4G shows an example platform employing an electrical sensor-based unit coupled with a particle separation unit, live-dead assay unit, and immunoassay unit for analysis of cell secretomes. This integrated platform may enable the simultaneous monitoring of cytokines,IL, growth factors, glucose, and lactose. By employing electrical sensors, accurate detection of these analytes can be achieved.

[0124] Among many detection methods, one method for detecting cytokines, IL, and growth factors may involve the use of specialized working electrodes that selectively interact with the target molecules, along with reference electrodes that may maintain a consistent electrical potential. To facilitate cytokine, IL, and growth factor detection, the working electrode may be modified with specific antibodies or antibody-based capture probes. When the sample contains the target molecules, they bind to the immobilized antibodies, resulting in measurable changes in the electrical signal. By precisely measuring and analyzing these changes, the concentration of the target cytokine or IL may be quantified. As glucose and lactose detection rely on enzyme-based reactions, glucose oxidase or lactose oxidase enzymes may be immobilized on the working electrode. When glucose or lactose is present in the sample, enzymatic reactions may occur, generating measurable electrical signals. Monitoring and analyzing these signals may enable the determination of glucose or lactose concentration in the sample.

[0125] To ensure accuracy and reliability, reference electrodes may be employed as a baseline for electrical measurements. Commonly used reference electrodes, such as Ag / AgCl electrodes or calomel electrodes, may provide a stable electrical potential serving as a reference point against which changes in the working electrode can be measured. This may ensure accurate and reliable detection results.

[0126] The integration of electrical sensor-based cytokine, IL, growth factor, glucose, and lactose detection unit with the exemplary live-dead assay and immunoassay platform may provide information about sample viability, immunophenotypic characteristics, cellular metabolism and response to various conditions or treatments. This capability may be valuable for a wide range of research and diagnostic applications, contributing to advancements in various fields.

[0127] Experimental Results and Additional Examples

[0128] A study was conducted to develop an advanced microchip specifically designed for use with annexin-V-based apoptosis assays. The study developed a proof of concept, incorporating a combination of an electrical sensor network and a microfluidic capture chamber. The integration enables the precise detection of phosphatidylserine (PS) externalization, a critical marker in the apoptosis process. By harnessing the power of electrical sensors, the microchip enables accurate monitoring and unparalleled quantification of apoptotic events with outstanding sensitivity. This technology will provide researchers andclinicians with an unprecedented platform to unravel the intricacies of apoptosis and its implications across diverse realms of research and diagnostics. Furthermore, the microchip extends its applications beyond apoptosis, encompassing vital areas such as cell differentiation, development, signaling, and activation. By offering invaluable insights, the exemplary transformative microchip holds tremendous potential in driving breakthroughs in biomedical research, drug discovery, cell therapeutics, and clinical applications, propelling us towards a new era of scientific advancements and improved patient outcomes.

[0129] Fabricated platform. The study fabricated an advanced microchip (i.e., platform) combining a microfluidic layer with an electrical sensor network. Fig. 5A shows a fabricated microfluidic platform employing a microfluidic layer coupled with an electrical sensor network. As shown in Fig. 5 A, the microfluidic layer accommodated two inlets, a mixer, and a cell capture chamber.

[0130] The fabricated platform started by intaking cells and a buffer containing biotinylated annexin-V via the two inlets. Through the integrated mixer, the cells and biotinylated annexin-V were mixed, ensuring a consistent distribution of the biotinylated annexin-V among the cells. During apoptosis, phosphatidylserine (PS) was externalized on the cell surface, indicating the apoptotic state. The biotinylated annexin-V bound to PS on the cell surface, facilitating the identification and analysis of apoptotic cells. Subsequently, the mixture of cells and biotinylated annexin-V flowed through the cell capture chambers. Within these chambers, the cells encountered neutravidin-coated pillars. As a result, the population of cells that bound to biotinylated annexin-V became immobilized upon interaction with the neutravidin coating pillars. This immobilization step enabled a more focused analysis and characterization of the apoptotic cells.

[0131] The fabricated platform incorporated an electrical sensor network comprising two sensors positioned at the entrance and exit of the capture chamber, enabling precise monitoring of cell movement. This network was based on the Microfluidic CODES platform [2-3], which tracked cell flow at specific locations within the microchip. By observing the entry and exit of cells in the capture chamber, the fabricated platform used a mass balance equation to determine the count of apoptotic cells within the chamber.

[0132] An electrical sensor network, employed by the fabricated platform, comprised interdigitated electrode pairs forming sensors. These pairs consisted of positive and negative output electrode fingers, acting as current sinks, while a common input electrode served as the current source. The electrodes were 5 pm wide and spaced apart by 5 pm, extending throughout the sensor's length and forming pairs with the output electrodes. Each sensorconsisted of 15 pairs (shown in Fig. 5 A), resulting in a waveform with multiple peaks of varying polarities. This unique waveform pattern provided a distinct code for each sensor. The specific sequence of polarities in the waveform corresponded to the physical arrangement of the input-output electrode finger pairs, giving rise to a distinctive waveform shape. Leveraging this waveform uniqueness, the fabricated platform employed advanced signal processing techniques to accurately identify the sensor responsible for generating a particular waveform. By integrating the Microfluidic CODES platform and utilizing the distinct waveform patterns generated by each sensor, the fabricated platform achieved precise localization and identification of cells within the capture chamber, providing accurate assessment of the populations of apoptotic cells.

[0133] The fabricated platform’s microfluidic pathways were fine-tuned to accommodate the target cell population, specifically Jurkat cells, which were associated with T-cell leukemia. These cells possessed an average diameter of approximately 11-14 pm. At the core of the fabricated platform, the capture chamber had dimensions of 145 mm in length and 1.5 mm in width. To facilitate optimal cell capture, the chamber incorporated an ingenious arrangement of pillars. These pillars, each with a diameter of 60 pm, were organized into three distinct stages. The first two stages shared a vertical spacing of 50 pm, but their horizontal spacing diverged. The first stage showcased a spacing of 160 pm, while the second stage tightened the gap to 50 pm. In the final act, the third stage revealed a 15 pm spacing between pillars, optimized to facilitate the interaction of cells ranging from 11 to 15 pm with these slender structures (shown in Fig. 5A). The channel height over the capture region (i.e., chamber) was 20 pm, creating an optimal environment for cell capture and analysis.

[0134] Micromixer evaluation. In the initial phase of platform development, the study focused on developing a mixer that would facilitate the achievement of a homogeneous blend between the cell sample and annexin-V buffer solution. To accomplish this, the study created three distinct micromixer designs: serpentine, triangular, and linear. These designs were tailored to incorporate two inlets, allowing for a separate supply of the working fluids.

[0135] Each of the micromixer configurations featured its unique geometry and dimensions. The serpentine micromixer boasted an extensive mixing length of 280 mm, while the triangular design settled at a respectable 150 mm. On the other hand, the linear micromixer, optimized for its specific purpose, had a more compact mixing length of 40 mm. Altogether, these micromixer designs comprised a total of 50 mixing units, providing opportunities for thorough blending.

[0136] To ensure optimal mixing performance, the study focused on the dimensions of each mixing unit, taking advantage of the distinct flow patterns induced by the different geometries employed. For example, the serpentine micromixer had a major axis diameter of 300 pm and a minor axis diameter of 140 pm for both inner and outer walls, while the triangular micromixer featured a curved vertex with a 70 pm diameter of curvature, enabling enhanced fluid flow and mixing capabilities. As for the channel width, the study set it at 50 pm for both the serpentine and triangular configurations. However, the linear micromixer presented a slight variation in its channel width. The linear micromixer featured a narrower passage with a width of 100 pm, which widened to 300 pm in the subsequent wider region. These chosen dimensions ensured that each mixing unit harnessed the distinct flow patterns induced by the specific micromixer geometry, leading to an efficient and effective blending process.

[0137] To evaluate and compare the mixing performance of the micromixers, the study conducted experiments using two color samples, Red and Blue, introduced through the respective inlets. The flow rate at each inlet was set to 100 pL / hr. Brightfield fluorescence images were captured using a digital camera, both at the inlet and the end of the mixing unit. These images were then utilized to calculate the mixing index (MI), which served as a metric to assess the effectiveness of each micromixer design.

[0138] Fig. 5B shows the brightfield microscope images and mixing index (MI) for three different micromixer types (e.g., serpentine, triangular, linear) at a 100 pL / hr flow rate. Subpanel (a) shows the brightfield microscope images of the mixers. Subpanel (b) shows the evaluation of the mixing index for the three types of mixers with 50 mixing units at the channel exit. All the mixing index (MI) values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured MI values.

[0139] The MI calculation was based on the RGB components (red, green, and blue) extracted from the acquired images. The study split the image and measured each resulting image using grayscale values ranging from 0 (completely black) to 255 (completely white), with each pixel representing the light intensity. After each mixing unit, the study selected a square region of interest (ROI) measuring 30 pm by 30 pm in the red stream and measured the color intensity. To quantify the red color, the study utilized the red measurement from the red channel, while the blue measurement corresponded to the blue color. Specifically, for the red color, the MI can be calculated using the Equation 1.100 %(Eq. 1)

[0140] In Equation 1, RROI represents the amount of red measured in the ROI, BROI represents the amount of blue measured in the ROI, whereas R unmix 3.11(1 Bunmix denote the amount of red and blue, respectively, in an unmixed state, measured at the beginning of the channel.

[0141] Based on the results shown in Fig. 5B, the study identified the Triangular micromixer as the optimal choice for the fabricated platform in terms of mixing performance. To maximize the performance of the triangular micromixer, the study sought to optimize its geometry and the number of mixing units it featured. At first, the study evaluated the MI at various mixing units and found, expectedly, that mixing performance increased with the number of mixing units used. The first unit produced an MI of 12.47, whereas 50 units produced an MI of 94.27. As the latter was sufficient for the use, the micromixer was configured to feature 50 mixing units.

[0142] Fig. 5C shows the brightfield microscope images and mixing index of the mixing units (e.g., 1st, 20th, 40th, 50th) using color dyes at a 100 pL / hr flow rate. All the mixing index (MI) values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured MI values. The study evaluated the MI at various mixing units, as shown in Fig. 5C, and determined that utilizing 50 mixing units was sufficient for the fabricated platform.

[0143] Fig. 5D shows the mixing index of the micromixers at various angles (e.g., 5°, 10°, 15°, 20°) and flow rates (50 pL / hr, 100 pL / hr, 150 pL / hr, 200 pL / hr). All the mixing index (MI) values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured MI values.

[0144] As shown in subpanel (a), an angle of 15° degrees yielded satisfactory results while maintaining a compact footprint. Larger angles tended to require more space, which could be impractical for the fabricated platform. In addition to angle evaluation, the study explored the impact of flow rate on the MI, as shown in subpanel (b). After conducting a thorough assessment, the study concluded that a flow rate of 100 pL / hr at each inlet represented the ideal configuration for the fabricated platform, striking a balance between mixing efficiency and practical considerations. By considering the MI results, the number of mixing units, the optimal angle, and the suitable flow rate, the study ensured the successful design and operation of the micromixer within the fabricated microfluidic platform.

[0145] Cell capture chamber evaluation. Next, the capture chamber was functionalized with neutravidin using the PDMS-avidin immobilization protocol to capture the biotinylated annexin- V-bound apoptotic cells within the chamber. The infusion of immobilization reagents across the capture chamber was facilitated by utilizing auxiliary ports located at the entry and exit points of the chamber. The process involved the utilization of various reagents, including APTES, glutaraldehyde, neutravidin, BSA, and glycine.

[0146] To initiate the process, the fabricated platform was wetted with ethanol immediately after bonding the PDMS glass using oxygen plasma assistance. APTES solution in Fethanol (3 % v / v) was then introduced into the fabricated platform and allowed to incubate at room temperature for 1 hour. Subsequently, the microchip was washed with ethanol and placed in a vacuum oven at 110 °C for 1 hour. Afterward, the fabricated platform underwent a wash with DI water, followed by infusion with a glutaraldehyde solution in DI water (3 % v / v). Another 1-hour incubation at room temperature was carried out, followed by washing the microchip with DI water and PBS. Next, a solution of neutravidin (1 mg / mL) in PBS was introduced into the fabricated platform, and the platform was incubated for 4 hours. The fabricated platform was then rinsed with PBS and incubated with a solution consisting of 3 % BSA blocking buffer and 1.5 mg / mL glycine for 1 hour to block nonspecific binding sites. Finally, the fabricated platform was washed with PBS to remove any unbound neutravidin.

[0147] To ensure successful immobilization of neutravidin within the capture chamber, the study undertook a series of steps. The process began with the functionalization of the capture chamber with neutravidin, followed by the infusion of Cy5 biotin conjugate, a fluorophore-conjugated biotin, into the chamber. Subsequently, fluorescence microscopy was employed to analyze the emitted fluorescence.

[0148] Fig. 5E shows the fluorescence microscopy image of the fabricated platform with a neutravidin-functionalized capture chamber. The study confirmed the effective coating of neutravidin in each capture chamber. Additionally, the observed uniform fluorescence signal throughout the chamber indicated a consistent and even distribution of the coating (shown in Fig. 5E), providing further evidence of successful neutravidin immobilization.

[0149] Then, the study investigated the impact of annexin-V concentration on the capture rate using Jurkat cells subjected to apoptosis induction. The apoptosis was induced by incubating the cells at 70 °C for 15 minutes, followed by a subsequent incubation at 37 °C for 12 hours. Different concentrations of annexin-V were introduced during the apoptotic cell- annexin-V mixing stage. The study quantified the total number of cells captured within thecapture chamber as a measure of the efficacy of annexin-V at each concentration. This analysis identified the optimal concentration of annexin-V for the fabricated platform.

[0150] Fig. 5F shows the impact of annexin-V concentration on the capture rate of apoptotic cells. All the capture rate values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured capture rate values.

[0151] The capture rate was determined by calculating the ratio of the total number of cells immobilized in the functionalized chamber to the total number of cells processed through the microchip. In Fig. 5F, as the concentration of annexin-V was reduced, the study observed a decrease in the capture rate, with a significant reduction of up to 36.55 % at the 2 pg / ml annexin-V concentration. This decrease can be attributed to the diminished availability of annexin-V molecules for binding to the surface of apoptotic cells. To ensure sufficient annexin-V availability and maximize capture efficiency, the study selected 10 pg / ml as the optimal annexin-V concentration.

[0152] Next, the study assessed the impact of cell flow rate on the capture efficiency. Fig. 5G shows the apoptotic cell capture rate at various flow rates and the saturation point of the fabricated platform when passing the annexin-V solution. All the capture rate values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured capture rate values.

[0153] The right balance should be struck, as excessively high flow rates may result in insufficient interaction time, leading to false negatives where target apoptotic cells may remain uncaptured. Conversely, excessively low flow rates lead to non-specific binding, causing false positives as non-target cells adhere to the chamber surface. The study conducted a series of experiments using different flow rates in the fabricated platform. Apoptotic cells and annexin-V were driven through the fabricated platform using a syringe pump, with flow rates ranging from 100 to 600 pl / hr for both inlets.

[0154] As shown in subpanel (a), at a flow rate of 100 pl / hr, the study achieved a capture rate of 94.10 %. This optimal flow rate provided ample interaction time for effective capture while minimizing non-specific adhesion artifacts and sedimentation issues that were more prominent at slower flow rates. However, as the flow rate was increased to 600 pl / hr, the capture rate decreased to 48.10 % (shown in subpanel a). After evaluation, the study determined that the optimal operating flow speed for the fabricated platform was 100 pl / hr, ensuring a high capture rate of 94.10 % while mitigating potential drawbacks associated with slower flow rates.

[0155] In addition, the study conducted an investigation to determine the saturation point of the fabricated platform concerning the amount of annexin-V needed. The presence of unbound biotinylated annexin-V in the sample solution, alongside the annexin-V captured by cells, may reduce the number of available neutravidin binding sites. Thus, determining the saturation point may be crucial for optimizing the fabricated platform configuration. To characterize the phenomenon, the study ran the microchip for an extended period using apoptotic Jurkat cells and an annexin-V buffer solution with a concentration of 10 pg / ml. During the operation, the study measured the capture rate by running the microchip with specific and quantified amounts of annexin-V. This approach provided precise evaluation of the capture efficiency at each distinct quantity of annexin-V introduced into the fabricated platform.

[0156] As the annexin-V solution passed through the microchip, the study observed a gradual decrease in the capture rate. As shown in subpanel (b), initially, the capture rate remained steady at approximately 96 % until 0.33 pg of annexin-V had been processed by the fabricated platform. However, as the amount of annexin-V increased to 1.17 pg, the capture rate dropped significantly to a mere 3 % (shown in subpanel b). This finding indicated the saturation point of the fabricated platform in terms of the amount of annexin-V used in the experimental setup.

[0157] To evaluate the specificity of the assay, the study implemented optimized surface chemistry and conducted experiments using both viable and heat induced apoptotic Jurkat cells. The experiments involved directing these cell populations through separate fabricated platforms at a consistent flow rate of 100 pl / hr. Throughout the experiments, the study employed fabricated platforms that had been functionalized with neutravidin.

[0158] The fabricated platform-based assay demonstrated good selectivity. Fig. 5H shows the capture rate of live and apoptotic Jurkat cells and flow cytometer measurement thereof. All the capture rate values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured capture rate values.

[0159] Among the population of live cells, 12.30 % were successfully captured by the fabricated platform (shown in subpanel a). In contrast, 93.60 % of the apoptotic cells were effectively captured (shown in subpanel b). These findings were further validated by flow cytometer analysis of the same sample shown in subpanel (b), providing additional evidence of the assay's robust performance. The fabricated platform-based assay distinguished between live and apoptotic Jurkat cells, highlighting the successful validation of its specificity. Byemploying optimized surface chemistry and meticulous experimental procedures, the study validated the specificity of the fabricated platform-based assay.

[0160] Apoptosis assay of Leukemic T Cell Line. Heat-induced apoptosis is a well- established characteristic of heat-related illnesses, where cellular damage becomes irreversible within the temperature range of 46 °C to 60 °C [4], The extent of damage is proportional to the duration of exposure. As temperatures escalate from 60 °C to 100 °C, protein coagulation rapidly occurs, leading to irreversible harm to crucial cytosolic and mitochondrial enzymes, as well as nucleic acid-histone complexes. If temperatures surpass 105 °C, tissues undergo boiling, vaporization, and carbonization.

[0161] The study induced apoptosis in Jurkat cells by subjecting them to an incubation temperature of 70 °C for 15 minutes. Following this incubation period, the cells were subsequently maintained at 37 °C for further analysis or immediately utilized to determine the percentage of apoptotic cells by evaluating the externalization of phosphatidylserine (PS) using the fabricated platform and flow cytometry.

[0162] Fig. 51 shows the externalization of phosphatidylserine (PS) using the fabricated platform and flow cytometry for the heat-induced apoptotic sample at 30 minutes, 1 hour 30 minutes, and 6 hours 30 minutes after initial treatment. All the capture rate values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured capture rate values.

[0163] As shown in subpanel (a), approximately 72 % of the cells displayed phosphatidyl serine (PS) externalization within 30 minutes after the heat treatment. The fabricated platform detected around 66% of apoptotic cells based on PS externalization. With an extended incubation time following the heat treatment, the study observed a progressive increase in the proportion of cells exhibiting PS externalization. Over time, there was a gradual and incremental rise in the number of cells displaying this characteristic.

[0164] As shown in subpanel (b), after 1 hour and 6 hours of incubation at 37 °C, approximately 93 % and 98 % of the cells, respectively, exhibited PS externalization, while the fabricated platform detected approximately 89 % and 97 % for their respective time points.

[0165] To comprehensively evaluate the time-dependent PS externalization subsequent to the initial heat treatment at 70 °C, the study conducted a quantitative analysis comparing flow cytometry with the fabricated platform to assess the kinetics of PS externalization over time and validate the effectiveness of fabricated platform in accurately capturing these changes.

[0166] To assess the accuracy and precision of the fabricated platform, the study conducted experiments using precisely calibrated suspensions containing a mixture of viable Jurkat cells and apoptotic Jurkat cells. The apoptotic cells were obtained by subjecting them to a 15-minute heat treatment at 70 °C, followed by a 12-hour incubation period. The study evaluated different ratios of viable and apoptotic cells, including 1 : 1, 1 :2, and 2: 1. Each mixed sample was processed using identically designed platforms.

[0167] Fig. 5J shows the frequencies of viable and apoptotic Jurkat cell subpopulations determined by the fabricated platform compared to nominal mix ratios (e.g., 1 : 1, 1 :2, 2: 1) determined by a hemocytometer. All the capture rate values were measured and were the average of three experiments. The error bars indicated the standard error associated with the measured capture rate values.

[0168] As shown in Fig. 5J, the fabricated platform demonstrated high accuracy in reporting the intended mixing ratios of viable and apoptotic cells. However, certain factors, such as impurities in the cell populations and inherent sample impurities, contributed to discrepancies between the results determined by the fabricated platform and the nominal mixing ratios.

[0169] Furthermore, the study performed a comparative analysis between the fabricated platform and a commercial fluorescence-based flow cytometer. For flow cytometry analysis, the study used FITC-conjugated annexin-V as a labeling probe, specifically targeting phosphatidylserine (PS) on the cell surface. The samples consisted of viable Jurkat cells and apoptotic Jurkat cells obtained 12 hours after the initial heat treatment, with a 1 : 1 ratio.

[0170] Fig. 5K shows the quantification (i.e., capture rate) of live and apoptotic cells obtained from the fabricated platform and the fluorescence-based flow cytometer. As shown in subpanel (a), the distinct subpopulations of live and apoptotic cells were identified based on fluorescence signals. As shown in subpanel (b), the error bars indicated the standard error associated with the measured capture rate values. All the capture rate values were measured and were the average of three experiments.

[0171] By gating the flow cytometry measurements based on PS expression, the study determined the frequency of viable and apoptotic cells as shown in Fig. 5K. Using flow cytometry data as the benchmark, the study found that the percentage of annexin-V negative (viable Jurkat cells) and annexin-V positive (apoptotic Jurkat cells) expressors was 47.8 % and 52.2 %, respectively. Comparing these results to fabricated platform, the study achievedan average error rate of less than or equal to 7 % (n = 3) (shown in subpanel b), indicating an accuracy of over 93 %.

[0172] The measurement errors can be attributed to several factors. Firstly, the capture-based apoptosis assay technique in the study employed a different sensing modality than the laser-scatter technique used by flow cytometers, leading to different discrimination criteria. Secondly, the sample preparation process resulted in cell loss and the inclusion of non-target cell particles, such as lysed residues.

[0173] Discussion

[0174] Immunophenotyping is a fundamental technique in immunology that involves the identification and characterization of immune cell populations based on their surface markers or intracellular molecules. In the traditional approach to immunophenotyping, flow cytometry and fluorescence microscopy have been widely used. Flow cytometry allows the analysis of individual markers by labeling cells with specific antibodies conjugated to fluorochromes. By measuring the emitted fluorescence, the expression levels of a single marker on different cell populations can be determined. Similarly, fluorescence microscopy provides visual information on individual markers but lacks the quantitative capabilities of flow cytometry.

[0175] Both flow cytometry and fluorescence microscopy, the traditional methods of immunophenotyping, have inherent limitations when it comes to the simultaneous analysis of multiple markers. A challenge is the issue of spectral overlap between fluorochromes, which imposes restrictions on the number of parameters that can be assessed using flow cytometry. The emission spectra of different fluorochromes can overlap, making it difficult to accurately distinguish the signals emitted by each fluorochrome. Consequently, this compromises the reliability and accuracy of the data obtained. Similarly, fluorescence microscopy is limited by the availability of distinct fluorochromes with non-overlapping spectra, further hindering comprehensive analysis. While advanced techniques such as compensation controls, multicolor flow cytometry panels, and specialized software for data analysis have been developed to overcome these limitations, there are still challenges associated with multiplex immunophenotyping. These limitations have impeded researchers' ability to perform comprehensive analyses of complex immune cell populations and their functional characteristics. The inability to simultaneously examine multiple markers restricts the depth of understanding in immunophenotyping studies.

[0176] To address the limitations of traditional multiplex immunophenotyping methods, the study developed the exemplary platform which offers several advantages overconventional techniques. Firstly, the exemplary platform provides the capability to simultaneously analyze multiple markers, allowing for a comprehensive assessment of complex cell populations. This multiplexing capability is crucial in understanding the intricate interactions and functional characteristics of cells. Furthermore, the platform-based immunoassay enables real-time analysis, eliminating the need for time-consuming sample processing and batch analysis. This real-time capability allows for immediate insights into the cell populations, facilitating prompt decision-making in research and clinical settings. The portability of the exemplary platform is another advantage. The exemplary platform offers flexibility in terms of sample collection and analysis, enabling immunophenotyping to be performed in diverse settings, such as point-of-care clinics, field research, or resource-limited areas. This portability empowers researchers and healthcare professionals with the ability to perform on-site immunophenotyping, leading to faster diagnosis, monitoring, and treatment decisions. The scalability of the platform-based immunoassay may expand to accommodate a wide range of markers, facilitating the analysis of extensive antibody panels. This scalability plays a role in keeping pace with the evolving understanding of immune cell populations and their profound implications in diverse diseases and conditions.

[0177] In conclusion, the exemplary platform-based immunoassay for multiplex immunophenotyping represents a breakthrough in this field. Equipped with real-time capabilities, exceptional portability, remarkable scalability, and the ability to simultaneously analyze multiple markers, the exemplary platform pioneers the way for enhanced research, diagnosis, and personalized treatment strategies in diagnostics, therapeutics, and personalized medicine.

[0178] Conclusion

[0179] The construction and arrangement of the systems and methods as shown in the various implementations are illustrative only. Although only a few implementations have been described in detail in this disclosure, many modifications are possible (e.g., variations in sizes, dimensions, structures, shapes, and proportions of the various elements, values of parameters, mounting arrangements, use of materials, colors, orientations, etc.). For example, the position of elements may be reversed or otherwise varied, and the nature or number of discrete elements or positions may be altered or varied. Accordingly, all such modifications are intended to be included within the scope of the present disclosure. The order or sequence of any process or method steps may be varied or re-sequenced according to alternative implementations. Other substitutions, modifications, changes, and omissions maybe made in the design, operating conditions, and arrangement of the implementations without departing from the scope of the present disclosure.

[0180] The present disclosure contemplates methods, systems, and program products on any machine-readable media for accomplishing various operations. The implementations of the present disclosure may be implemented using existing computer processors, or by a special purpose computer processor for an appropriate system, incorporated for this or another purpose, or by a hardwired system. Implementations within the scope of the present disclosure include program products, including machine-readable media for carrying or having machine-executable instructions or data structures stored thereon. Such machine- readable media can be any available media that can be accessed by a general-purpose or special-purpose computer or other machine with a processor. By way of example, such machine-readable media can comprise RAM, ROM, EPROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code in the form of machineexecutable instructions or data structures, and which can be accessed by a general purpose or special purpose computer or other machine with a processor.

[0181] When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or a combination of hardwired or wireless) to a machine, the machine properly views the connection as a machine-readable medium, thus, any such connection is properly termed a machine-readable medium. Combinations of the above are also included within the scope of machine-readable media. Machine-executable instructions include, for example, instructions and data that cause a general-purpose computer, special-purpose computer, or special-purpose processing machine to perform a certain function or group of functions.

[0182] Although the figures show a specific order of method steps, the order of the steps may differ from what is depicted. Also, two or more steps may be performed concurrently or with partial concurrence. Such variation will depend on the software and hardware systems chosen and on the designer's choice. All such variations are within the scope of the disclosure. Likewise, software implementations could be accomplished with programming techniques with rule-based logic and other logic to accomplish the various connection steps, processing steps, comparison steps, and decision steps.

[0183] Machine Learning. In addition to the machine learning features described above, the analysis system can be implemented using one or more artificial intelligence and machine learning operations. The term “artificial intelligence” can include any technique thatenables one or more computing devices or computing systems (i.e., a machine) to mimic human intelligence. Artificial intelligence (Al) includes but is not limited to knowledge bases, machine learning, representation learning, and deep learning. The term “machine learning” is defined herein to be a subset of Al that enables a machine to acquire knowledge by extracting patterns from raw data. Machine learning techniques include, but are not limited to, logistic regression, support vector machines (SVMs), decision trees, Naive Bayes classifiers, and artificial neural networks. The term “representation learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, or classification from raw data. Representation learning techniques include, but are not limited to, autoencoders and embeddings. The term “deep learning” is defined herein to be a subset of machine learning that enables a machine to automatically discover representations needed for feature detection, prediction, classification, etc., using layers of processing. Deep learning techniques include but are not limited to artificial neural networks or multilayer perceptron (MLP).

[0184] An artificial neural network (ANN) is a computing system including a plurality of interconnected neurons (e.g., also referred to as “nodes”). This disclosure contemplates that the nodes can be implemented using a computing device (e.g., a processing unit and memory as described herein). The nodes can be arranged in a plurality of layers, such as an input layer, an output layer, and optionally one or more hidden layers with different activation functions. An ANN having hidden layers can be referred to as a deep neural network or multilayer perceptron (MLP). Each node is connected to one or more other nodes in the ANN. For example, each layer is made of a plurality of nodes, where each node is connected to all nodes in the previous layer. The nodes in a given layer are not interconnected with one another, i.e., the nodes in a given layer function independently of one another. As used herein, nodes in the input layer receive data from outside of the ANN, nodes in the hidden layer(s) modify the data between the input and output layers, and nodes in the output layer provide the results. Each node is configured to receive an input, implement an activation function (e.g., binary step, linear, sigmoid, tanh, or rectified linear unit (ReLU) function), and provide an output in accordance with the activation function. Additionally, each node is associated with a respective weight. ANNs are trained with a dataset to maximize or minimize an objective function. In some implementations, the objective function is a cost function, which is a measure of the ANN’S performance (e.g., error such as LI or L2 loss) during training, and the training algorithm tunes the node weights and / or bias to minimize the cost function. This disclosure contemplates that any algorithm that finds themaximum or minimum of the objective function can be used for training the ANN. Training algorithms for ANNs include but are not limited to backpropagation. It should be understood that an artificial neural network is provided only as an example machine learning model. This disclosure contemplates that the machine learning model can be any supervised learning model, semi-supervised learning model, or unsupervised learning model. Optionally, the machine learning model is a deep learning model. Machine learning models are known in the art and are therefore not described in further detail herein.

[0185] A convolutional neural network (CNN) is a type of deep neural network that has been applied, for example, to image analysis applications. Unlike traditional neural networks, each layer in a CNN has a plurality of nodes arranged in three dimensions (width, height, depth). CNNs can include different types of layers, e.g., convolutional, pooling, and fully-connected (also referred to herein as “dense”) layers. A convolutional layer includes a set of filters and performs the bulk of the computations. A pooling layer is optionally inserted between convolutional layers to reduce the computational power and / or control overfitting (e.g., by downsampling). A fully-connected layer includes neurons, where each neuron is connected to all of the neurons in the previous layer. The layers are stacked similarly to traditional neural networks. GCNNs are CNNs that have been adapted to work on structured datasets such as graphs.

[0186] Other Supervised Learning Models. A logistic regression (LR) classifier is a supervised classification model that uses the logistic function to predict the probability of a target, which can be used for classification. LR classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize an objective function, for example, a measure of the LR classifier’s performance (e.g., an error such as LI or L2 loss), during training. This disclosure contemplates that any algorithm that finds the minimum of the cost function can be used. LR classifiers are known in the art and are therefore not described in further detail herein.

[0187] A Naive Bayes’ (NB) classifier is a supervised classification model that is based on Bayes’ Theorem, which assumes independence among features (i.e., the presence of one feature in a class is unrelated to the presence of any other features). NB classifiers are trained with a data set by computing the conditional probability distribution of each feature given a label and applying Bayes’ Theorem to compute the conditional probability distribution of a label given an observation. NB classifiers are known in the art and are therefore not described in further detail herein.

[0188] A k-NN classifier is an unsupervised classification model that classifies new data points based on similarity measures (e.g., distance functions). The k-NN classifiers are trained with a data set (also referred to herein as a “dataset”) to maximize or minimize a measure of the k-NN classifier’s performance during training. This disclosure contemplates any algorithm that finds the maximum or minimum. The k-NN classifiers are known in the art and are therefore not described in further detail herein.

[0189] A majority voting ensemble is a meta-classifier that combines a plurality of machine learning classifiers for classification via majority voting. In other words, the majority voting ensemble’s final prediction (e.g., class label) is the one predicted most frequently by the member classification models. The majority voting ensembles are known in the art and are therefore not described in further detail herein.

[0190] It is to be understood that the methods and systems are not limited to specific synthetic methods, specific components, or to particular compositions. It is also to be understood that the terminology used herein is for the purpose of describing particular implementations only and is not intended to be limiting.

[0191] As used in the specification and the appended claims, the singular forms “a,” “an” and “the” include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another implementation includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent “about,” it will be understood that the particular value forms another implementation. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint.

[0192] “Optional” or “optionally” means that the subsequently described event or circumstance may or may not occur, and that the description includes instances where said event or circumstance occurs and instances where it does not.

[0193] Throughout the description and claims of this specification, the word “comprise” and variations of the word, such as “comprising” and “comprises,” means “including but not limited to,” and is not intended to exclude, for example, other additives, components, integers or steps. “Exemplary” means “an example of’ and is not intended to convey an indication of a preferred or ideal implementation. “Such as” is not used in a restrictive sense, but for explanatory purposes.

[0194] Disclosed are components that can be used to perform the disclosed methods and systems. These and other components are disclosed herein, and it is understood that when combinations, subsets, interactions, groups, etc. of these components are disclosed while specific reference of each various individual and collective combinations and permutation of these may not be explicitly disclosed, each is specifically contemplated and described herein, for all methods and systems. This applies to all aspects of this application, including, but not limited to, steps in disclosed methods. Thus, if a variety of additional steps can be performed, it is understood that each of these can be performed with any specific implementation or combination of the disclosed methods.

[0195] The following patents, applications and publications as listed below and throughout this document are hereby incorporated by reference in their entirety herein.[1] K. Balasubramanian, B. Mirnikjoo and A. J. Schroit, “Regulated externalization of phosphatidylserine at the cell surface: implications for apoptosis”, J Biol Chem., 282(25), 18357-18364 (2007).[2] R. Liu, N. Wang, F. Kamili and A. F. Sari oglu, “Microfluidic CODES: a scalable multiplexed electronic sensor for orthogonal detection of particles in microfluidic channels”, Lab on a Chip, 16, 1350 (2016).[3] R. Liu, W. Waheed, N. Wang, O. Civelekoglu, M. Boya, C. H. Chu and A. F. Sarioglu,“Design and modeling of electrode networks for code-division multiplexed resistive pulse sensing in microfluidic devices”, Lab on a Chip, 17, 2650 (2017).[4] B. Leber, U. Mayrhauser, B. Leopold, S. Koestenbauer, K. Tscheliessnigg, V. Stadlbauer and P. Stiegler, “Impact of temperature on cell death in a cell-culture model of hepatocellular carcinoma”, Anticancer Res., 32(3), 915-21(2012).

Claims

What is claimed:

1. A method comprising: treating a pre-treated cell population with a binding reagent configured to bind to apoptotic cells in the cell population to form a treated cell population; flowing the treated cell population through a microfluidic device comprising, including through, a functionalized chamber configured with a plurality of functionalized surfaces each coated to bind with the binding reagent and combined cells bound by binding reagents to maintain apoptotic subpopulation in the chamber; and measuring, via a first electrode located at an inlet of the functionalized chamber and a second electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, to evaluate a pre-treatment applied to the pre-treated cell population.

2. The method of claim 1 further comprising: mixing the cell population and the binding reagent in a mixing space.

3. The method of claim 1 or 2, wherein the functionalized chamber includes protrusion structures configured with a portion of the functionalized surfaces.

4. The method of claim 3, wherein the protrusion structures includes (i) a first subset of protrusion structures located at a first region in the functionalized chamber, the first subset of protrusion structures having a first density arrangement, and (ii) a second subset of protrusion structures located at a second region in the functionalized chamber downstream to the first region, the second subset of protrusion structures having a second density arrangement, wherein the second density arrangement is higher the first density arrangement.

5. The method of any one of claims 1-4, wherein the binding reagent includes biotinylated Annexin V and an Annexin V binding buffer.

6. The method of any one of claims 1-4, wherein the binding reagent is an immunoassay probe.

7. The method of any one of claims 1-6, comprising: analyzing the set of measurement signals to determine at least one of (i) the number of cells entering the inlet to the functionalized chamber, (ii) the number of cells exiting at the outlet of the functionalized chamber, (iii) the number of apoptotic cells or a number of non- apoptotic cells in the pre-treated cell population, and (iv) a combination thereof.

8. The method of any one of claims 1-7, comprising: flowing the treated cell population through a second functionalized chamber located downstream to the functionalized chamber, wherein the second functionalized chamber is configured with a second plurality of functionalized surfaces each coated to bind a second binding reagent different from the binding reagent; and measuring, via a third electrode located at an inlet of the second functionalized chamber and a fourth electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the second functionalized chamber and a number of cells exiting at the outlet of the second functionalized chamber to determine a number of cells having correspondence to the second binding reagent.

9. The method of any one of claims 1-8, comprising: flowing the treated cell population through a third functionalized chamber located in parallel to the functionalized chamber, wherein the third functionalized chamber is configured with a third plurality of functionalized surfaces each coated to bind a third binding reagent different from the binding reagent; and measuring, via a fifth electrode located at an inlet of the third functionalized chamber and a sixth electrode located at an outlet of the third functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the third functionalized chamber and a number of cells exiting at the outlet of the third functionalized chamber to determine a number of cells having correspondence to the third binding reagent.

10. A microfluidic device comprising: a substrate having a device inlet configured to receive a plurality of pre-treated cells;a functionalized chamber formed in the substrate, the functionalized chamber being in fluidic communication with the device inlet, the functionalized chamber configured with a plurality of functionalized surfaces each coated to bind (i) an apoptotic binding agent (loose reagent not bounded) and (ii) the apoptotic binding reagent when bounded to apoptotic cells of the pre-treated cells, to maintain apoptotic subpopulation in the functionalized chamber; and a set of electrodes positioned in the substrate, wherein the set of electrodes includes (i) a first electrode located at an inlet of the functionalized chamber and (ii) a second electrode located at an outlet of the functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non- apoptotic cells in the pre-treated cell population.

11. The microfluidic device of claim 10 comprising: a mixing channel defined in the substrate between the device inlet and the functionalized chamber, the mixing channel having at least a first channel portion having a first flow direction, a second channel portion having a second flow direction, and a third channel portion having a third flow direction.

12. The microfluidic device of claim 11, wherein the first channel portion, the second channel portion, and the third channel portion forms a serpentine or meandering channel structure.

13. The microfluidic device of any one of claims 10 - 12, wherein the functionalized chamber includes protrusion structures configured with a portion of the functionalized surfaces.

14. The microfluidic device of claim 13, wherein the protrusion structures includes (i) a first subset of protrusion structures located at a first region in the functionalized chamber, the first subset of protrusion structures having a first density arrangement, and (ii) a second subset of protrusion structures located at a second region in the functionalized chamber downstream to the first region, the second subset of protrusion structures having a seconddensity arrangement, wherein the second density arrangement is higher the first density arrangement.

15. The microfluidic device of any one of claims 10 - 14, wherein the first electrode and the second electrode each includes an interdigitated arrangement of three metal structures.

16. The microfluidic device of any one of claims 10 - 15, comprising at least two capture zones including a second functionalized chamber formed in the substrate, the second functionalized chamber being in fluidic downstream communication with the functionalized chamber, the second functionalized chamber configured with a second plurality of functionalized surfaces each coated to bind a second binding reagent to maintain a second cell subpopulation associated with the second binding reagent in the functionalized chamber; and a second set of electrodes positioned in the substrate, wherein the second set of electrodes includes (i) a third electrode located at an inlet of the second functionalized chamber and (ii) a fourth electrode located at an outlet of the second functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the second functionalized chamber and a number of cells exiting at the outlet of the second functionalized chamber to determine a number of cells associated with the second binding reagent.

17. The microfluidic device of claim 16 comprising at least three capture zones including: a third functionalized chamber formed in the substrate, the third functionalized chamber being in fluidic parallel communication with the functionalized chamber, the third functionalized chamber configured with a third plurality of functionalized surfaces each coated to bind a third binding reagent to maintain a third cell subpopulation associated with the third binding reagent in the third functionalized chamber; and a third set of electrodes positioned in the substrate, wherein the third set of electrodes includes (i) a fifth electrode located at an inlet of the third functionalized chamber and (ii) a sixth electrode located at an outlet of the third functionalized chamber, configured to respectively measure a set of measurement signals having correspondence to a number of cells entering at the inlet to the third functionalized chamber and a number of cells exiting at the outlet of the third functionalized chamber to determine a number of cells associated with the third binding reagent.

18. The microfluidic device of any one of claims 10 - 16, wherein the set of electrodes are coupled to an instrument for impedance spectroscopy.

19. The microfluidic device of any one of claims 10 - 16 further comprising: a filter for sample or chemical purification.

20. A method of measuring cell population in a rapid cell characterization analysis, comprising: flowing a pre-treated cell population across an electrode having a plurality of sensing surfaces (sets of electrode pair and common electrode) to form a measurement signal having measurement waveform defined by observations from each of the plurality of sensing surfaces; determining, via a trained Al model, using the measurement signal, a score value of a cell of the pre-treated cell population having one of set of pre-defined phenotypes, wherein the trained Al model was trained using training data generated from the method of any one of claims 1-10 or the system of any one of claims 1-17, wherein at least one of the functionalized chamber was coated with a reagent associated with one of the set of pre-defined phenotypes.

21. A method of measuring cell population in a rapid cell viability analysis comprising: flowing a pre-treated cell population across an electrode having a plurality of sensing surfaces to form a measurement signal having measurement waveform defined by observations from each of the plurality of sensing surfaces; determining, via a trained Al model, using the measurement signal, a score value of a cell of the pre-treated cell population being alive or the pre-treated cell population being dead or programmed for death, wherein the trained Al model was trained using training data generated from the method of any one of claims 1-10 or the system of any one of claims 1-17.

22. A method comprising: providing a pre-treated cell population; flowing the treated cell population through a microfluidic device comprising, including through, a functionalized chamber configured with a plurality of functionalizedsurfaces (e.g., immobilized eptide ligands, protein scaffolds, or antibodies) each coated to bind to the pre-treated cell; and measuring, via a first electrode located at an inlet of the functionalized chamber and a second electrode located at an outlet of the functionalized chamber, respectively, a set of measurement signals having correspondence to a number of cells entering at the inlet to the functionalized chamber and a number of cells exiting at the outlet of the functionalized chamber to determine a number of apoptotic cells or a number of non-apoptotic cells in the pre-treated cell population, to evaluate a pre-treatment applied to the pre-treated cell population.

Images