Microlens enhanced multi-modal on-chip imaging

Abstract

An on-chip imaging system useful for fluorescence microscopy, bright-field imaging, and flow cytometry, and methods of making and using the same, are described.

Description

70647- WO-PCT / TECH-2025 - 15TITLEMicrolens Enhanced Multi-Modal On-Chip ImagingInventors: Aniruddha Ray, Somaiyeh Khoubafarin DoustRELATED APPLICATIONS

[0001] This application claims priority to United States Provisional Application No. 63 / 749,437 Hied under 35 U.S.C. § 111(b) on January 24, 2025, the entire disclosure of which is incorporated herein by reference for all purposes.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0002] This invention was made with government support under grant number R15 EB034552 awarded by the Nadonal Institutes of Health. The government has certain rights in this invention.BACKGROUND

[0003] Imaging and biosensing devices are used worldwide for a wide variety of applications from research to healthcare to environmental monitoring. However, these measurements become challenging in resource-limited settings such as remote areas or villages in third-world countries. In resource-limited setdngs, traditional imaging and readout devices are often hard to obtain due to high costs, bulky designs, and the need for trained personnel to operate them.

[0004] Fluorescence microscopy has long been a cornerstone in biological and medical research due to its ability to visualize specific structures, molecules, and processes within cells through the use of fluorescent markers. Fluorescence microscopy plays a crucial role in applications ranging from cellular biology to disease diagnosis and drug discovery, providing a window into the intricate workings of living systems. The evaluation of anti-cancer drugs and understanding cellular responses, such as oxidative stress and cell death, are fundamental to advancing cancer treatment. Traditional fluorescence microscopy is a common tool in these studies but is often limited by its narrow field-of-view, as well as being bulky, complex, and costly. As research continues to evolve, the demand for high-throughput imaging has grown substantially. This capability is important for studying dynamic cellular processes on a large scale, allowing researchers to track real-time changes across a multitude of cells or tissues simultaneously. Such insights are important for advancing the understanding of complex biological behaviors.

[0005] The meet the growing needs, the development of portable fluorescence microscopes has become increasingly important. Traditional fluroscence microscopes, while highly effective, are often large, expensive, and restricted to use in laboratory environments. Portable systems, on the other hand,70647- WO-PCT / TECH-2025 - 15offer researchers the flexibility to conduct imaging studies in a variety of settings, including fieldwork, point-of-care diagnostics, and low-resource environments. This portability opens the door to broader applications, particularly in areas where access to traditional microscopy is limited. Moreover, the high cost of conventional fluorescence microscopes presents a significant barrier, particularly for smaller research institutions or in regions with limited resources. Low-cost microscopy solutions are important for expanding access to these powerful imaging tools without sacrificing performance. By reducing costs, researchers can conduct large-scale studies more efficiently, enabling the kind of high-throughput imaging that is essential for advancing both basic and applied sciences.

[0006] Traditional lab equipment for fluorescence microscopy and cytometry is expensive, which limits accessibility for smaller labs, educational institutions, and point-of-care testing. Most diagnostic imaging devices are also not portable, making it difficult to conduct in-field studies, on-site diagnostics, or rapid analysis outside of laboratory settings. Similarly, fluorescence-based affordable wearable readout devices for direct monitoring in the body is not widely available in the market. Wearable readout devices are important for on-site and real-time health monitoring. There have been efforts to develop low-cost portable readout devices for fluorescence and other modalities. However, so far all cheap, portable systems suffer from poor sensitivity (signal to noise ratio), low resolution, or other drawbacks.

[0007] On-chip fluorescence microscopy has emerged as a solution to some of these challenges. The technique eliminates the need for traditional optical lenses, instead relying on computational algorithms to reconstruct images from the light field captured by the sensor. By doing so, on-chip systems are not only more comptact and portable but also significantly more affordable. These characteristics make on-chip fluorsescnce microscopy ideal for applications requiring mobility, cost-effectievness, and scalability. A key advantage of on-chip fluorescent microscopy is its ability to provide a wide field of view, enabling the imaging of large sample areas in a single shot. This capability is particularly useful for high-throughput applications, such as screening assays or large-scale cell imaging. Despite these advantages, on-chip fluorsecnce microscopy is not without its limitations. One of its main drawbacks is the lower resolution compared to conventional lens-based systems. Additionally, the computrational algorithms used for image reconstruction can introduce challenges, especially when dealing with complex biological samples.

[0008] There is also a need for real-time monitoring. Conventional imaging systems may lack realtime capabilities, making it challenging to monitor dynamic changes in samples, such as the interaction of cells with drug drug responses in cells or environmental changes or even in the body.

[0009] In view of the above challenges and issues, there is a need in the art for new and improved on-chip imaging systems.70647- WO-PCT / TECH-2025 - 15SUMMARY

[0010] Provided is an on-chip imaging system comprising a substrate; a monolayer of microlenses distributed on the substrate; a sensor comprising an array of photodetectors configured to detect light, wherein the substrate is disposed on the sensor; a light source configured to emit excitation light toward a sample; and a prism configured to receive the excitation light from the light source and transmit the excitation light to the sample.

[0011] In certain embodiments, the substrate comprises glass.

[0012] In certain embodiments, the prism is a glass rhomboid prism.

[0013] In certain embodiments, the prism is configured to transmit the excitation light to the sample at an angle greater than a critical angle.

[0014] In certain embodiments, the microlenses comprise borosilicate glass having a diameter in a range of from about 150 to about 180 micrometers and a refractive index of about 1.48.

[0015] In certain embodiments, the microlenses comprise glass spheres with a diameter of about 0.5 mm and a refractive index of about 1.5.

[0016] In certain embodiments, the microlenses comprise a polymeric material with a refractive index of about 1.59 and a diameter of about 42 pm.

[0017] In certain embodiments, the on-chip imaging system further comprises an emission filter configured to selectively transmit wavelengths of light while blocking excitation light.

[0018] In certain embodiments, the microlenses comprise borosilicate glass, barium titanate, glass spheres, or a combination thereof.

[0019] In certain embodiments, the substrate is flexible.

[0020] In certain embodiments, the on-chip imaging system further comprises an emission filter disposed between the sensor and the substrate.

[0021] In certain embodiments, the monolayer is evenly distributed on the substrate.

[0022] In certain embodiments, the on-chip imaging system further comprises a light source configured to emit excitation light at a wavelength capable of exciting fluorophores in a sample.

[0023] In certain embodiments, the on-chip imaging system further comprises a signal processing module configured to transfer data to an external system.

[0024] Further provided is an on-chip imaging system comprising a sensor chip comprising an array of photodetectors configured to detect light and convert the detected fight into electrical signals; a substrate on the sensor chip; and an array of microlenses affixed to the substrate with a gel.

[0025] In certain embodiments, the gel comprises polydimethylsiloxane (PDMS), bovine gelatin, agarose, sodium alginate, or acrylamide.

[0026] In certain embodiments, the microlenses are not submerged in the gel.70647- WO-PCT / TECH-2025 - 15

[0027] In certain embodiments, the sensor chip comprises a CMOS sensor.

[0028] In certain embodiments, the on-chip imaging system comprises an emission fdter between the sensor chip and the substrate, the emission filter being configured to selectively transmit wavelengths of fight while blocking excitation light.

[0029] In certain embodiments, the on-chip imaging system is in the form of a wearable microscope.

[0030] Further provided is an on-chip imaging system comprising a CMOS sensor; an emission filter directly on the CMOS sensor; and a microlens substrate directly on the CMOS sensor, wherein the microlens substrate comprises an array of microlenses affixed to a substrate with a gel.

[0031] Further provided is a method for conducting on-chip imaging, the method comprising transmitting excitation light through a prism to a sample; focusing light emitted from the sample through a microlens substrate comprising an array of microlenses; filtering the focused light through an emission filter; and detecting the focused and filtered light with a sensor chip comprising photodetectors.

[0032] In certain embodiments, the excitation light is transmitted through the prism and to the sample at an angle greater than a critical angle to reject excitation light at a bottom facet of the sample and to reduce unwanted excitation light interference.

[0033] In certain embodiments, the excitation light is transmitted through the prism and to the sample vertically from above the sample.|0034| In certain embodiments, the microlense substrate comprises the array of microlenses affixed to a substrate with a gel.

[0035] In certain embodiments, the microlenses are not submerged in the gel.

[0036] In certain embodiments, the microlenses comprise borosilicate glass having a diameter in a range of from about 150 to about 180 micrometers. In particular embodiments, the borosilicate glass has a refractive index of about 1.48.

[0037] In certain embodiments, the method comprises imaging flowing particles in real-time.

[0038] Further provided is the use of an array of microlenses to focus light from an excited sample prior to collecting the light with a chip sensor to conduct on-chip fluorescence microscopy or bright-field imaging.BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0040] FIG. 1: Schematic view of an on-chip fluorescence imaging platform.

[0041] FIG. 2: Photon density vs. sensor distance from microlens and object distance from the70647- WO-PCT / TECH-2025 - 15microlens.

[0042] FIGS. 4A-4C: Imaging of 1 gm fluorescence particle with (FIG. 4A) and without (FIG. 4B) microlens, and a conventional fluorsecency microscopy image with lOx objective, NA=0.2 (FIG. 4C).

[0043] FIGS. 5A-5C: FIG. 5A shows time series images of cells treated with 50 gM DOX in the first row; on-chip fluorescence microscope. The second row in FIG. 5A shows fluorescence microscope in FITCI channel to measure DCF excitement. The third row in FIG.5A shows fluorescence microscope in TXR channel to monitor cell viability by staining cells with 1 gM PI. The fourth row in FIG.5A shows phase contrast microscope, with lOx objective. The scale bar is 242.6 gm. FIG. 5B shows an intensity graph using on-chip fluorescence microscopy. FIG. 5C shows the same using conventional fluorescence microscopy.

[0044] FIG. 6: Time series images of cells treated with 1 mM H2O2 on-chip fluorescence microscope and intensity graph versus time. The scale bar is 100 gm.

[0045] FIG. 7: Time series of calcium influx in a specific BT-20 cell upon addition of calcium chloride and calcium ionophore. The sequence of fluorescence images of the cell taken at various time points after the addition of calcium chloride (8 mM) and calcium ionophore (5 gM). The graph of fluorescence intensity for the marked cell over time demonstrates a gradual increase in intracellular calcium concentration following calcium ionophore treatment, indicative of effective calcium transport across the cell membrane. The scale bar is 100 gm.

[0046] FIGS. 8A-8D: FIGS.8A-8B show time lapse images of FITC 4-kilodalton Dextran dye flow from a vascular section into a central tissue compartment through a porous interface. FIGS. 8C-8D show 3D fluorescence intensity changes along the line shown in the figure overtime and position. The scale bar is 100 gm. The arrow in FIG.8D shows the flow of dye from vascular channels to the central tissue compartment.100471 FIGS. 9A-9B: FIG. 9A shows time lapse image flow of fluorescent nanoparticle with diameter of 42 nm from vascular section into the central tissue compartment through porous interface. FIG.9B shows fluorescence intensity changes along the line shown in the figure overtime and position. The scale bar is 100 gm.

[0048] FIGS. 10A-10B: FIG. 10A shows time lapse image flow of calcium chloride and calcium ionophore from the vascular section into the central tissue compartment through porous interface. FIG. 10B shows 3D fluorescence intensity changes along the line shown in the figure overtime and position. The scale bar is 100 gm.

[0049] FIG. 11: Imaging set up for virtualization of moving particle through thin capillary as described in Example II herein.

[0050] FIG. 12: Imaging movement of 1 gm fluorescence particles through capillary.70647- WO-PCT / TECH-2025 - 15

[0051] FIG. 13: Illustration of image formation using a microlens substrate.

[0052] FIGS. 14A-14C: Different views of thin microlens substrates made by PDMS. FIG. 14A shows an image of a solid microlens substrate captured by camera. FIG. 14B shows a phase contrast image of a microlens substrate using lOx objective. FIG. 14C shows a cross sectional image of a substrate.

[0053] FIGS. 15A-15C: FIG. 15A shows an image of a flexible substrate. FIG. 15B shows the microlens array can be homogeneous with all microlens with same size and refractive index for imaging at a constant height. The example shown is size 150-180 microns and with a refractive index of 1.48 (borosilicate). FIG. 15C shows the microlens array can be heterogeneous, featuring microlenses of varying sizes and / or refractive indices to achieve increased multiplexing imaging height. The size shown is 150-180 microns and with a refractive index of 1.48 (borosilicate glass) and 1.98 (Barium Titanate glass).

[0054] FIG. 16: Schematic of an on-chip bright field microscope.

[0055] FIG. 17: Time lapse image of morphological changes of BT-20 cells treated with 50 LIM DOX (first row) and 2 mM hydrogen peroxide (second row). The scale bar is 200 pm.

[0056] FIG. 18: Imaging of a fixed cell with 10 pM 3-(5-methoxy-2-methyl-lH-indol-3-yl)-l-(4-pyridinyl)-2-propene-l-one (MOMIPP) using the on-chip bright field microscope, showing methuosis.

[0057] FIG. 19: Il 1 ustration of the on-chip imaging system described herein with either fluorescence excitation light or bright field illumination.|0058| FIGS. 20A-20C: Comparison showing the resolution of the bright field on-chip microscope.FIG. 20A shows a conventional light field microscope image of group 9 in a test target. FIG.20B shows on-chip bright field image of group 9 in the test target without a micerolens substrate. FIG.20C shows an on-chip bright field image of group 9 in the test target using a microlens substrate.

[0059] FIG. 21: On-chip bright field microscope images distinguishing apoptosis from necrosis.

[0060] FIG. 22: Schematic of a wearable microscope utilizing the on-chip imaging system described herein. MS = microlens substrate. L = laser or light emitting diode.

[0061] FIG. 23: Fluorsecent image of microparticles acquired using the wearable microscope.

[0062] FIGS. 24A-24B: Photograph showing a top view of the wearable microscope (FIG. 24A) and photograph of the wearable microscope with light turned on (FIG.24B).DETAILED DESCRIPTION

[0063] Throughout this disclosure, various publications, patents, and published patent specifications are referenced by an identifying citation. The disclosures of these publications, patents, and published patent specifications are hereby incorporated by reference into the present disclosure in their entirety to more fully describe the state of the art to which this invention pertains.

[0064] On-chip microscopy integrates optical microscopy techniques directly onto a microchip or70647- WO-PCT / TECH-2025 - 15compact platform. Instead of using bulky traditional microscopes, on-chip microscopy miniaturizes and automates imaging systems. In general, samples are illuminated, light diffraction patterns are captured by an image sensor, and then computational algorithms reconstruct the image. In fluorescence on-chip microscopy, fluorophores (such as fluorescent dyes) in samples are excited, and the emitted light is detected by integrated sensors. The emitted fluorescence is detected by a sensor (such as a CMOS or CCD sensor). Image processing and computational algorithms are used to reconstruct high-resolution images from raw sensor data.

[0065] In accordance with the present disclosure, a cost-effective on-chip imaging system that combines a wide field of view with high resolution and a favorable signal-to-noise ratio has been developed. The on-chip imaging system utilizes a two-dimensional microlens array, placed underneath the sample, to capture the emission, and can be used for fluorescence microscopy or bright field imaging. By exciting the sample through a prism interface at an angle greater than the critical angle, the excitation light is successfully rejected at the bottom facet of the sample substrate, enhancing image clarity in fluorescence microscopy. Various experimental parameters have also been improved, such as sample-to-microlens and microlens-to-sensor distances, as described in the examples herein.

[0066] FIG. 19 depicts the on-chip imaging system 10 with either fluorescence excitation light or bright field illumination. Referring to FIG. 19, the on-chip imaging system 10 includes a sensor 12. The sensor 12 is a sensor chip that includes photodectors, such as CMOS or CCD elements, with signal processing circuity to capture and process visual information. The sensor 12 converts light signals into electrical signals for real-time imaging or data analysis. The electrical signals are processed by on-chip electronics, which may include amplifiers, analog-to-digial converters, and signal processors, to generate a digital image. In some non-limiting examples, the sensor 12 is a complementary metal-oxide-semiconductor sensor (CMOS) sensor. A CMOS sensor is a type of image sensor that converts light into electrical signals to produce digital images. Each pixel in a CMOS sensor has a photodiode that collects light and converts it into an electrical charge. The electrical signal from each pixel is amplified directly on the CMOS sensor chip. However, other sensors 12 are possible and encompassed within the scope of the present disclosure.

[0067] Referring still to FIG. 19, the on-chip imaging system 10 may include an emission filter 14 directly on the sensor 12. The emission filter 14, when present, selectively transmits specific wavelengths of light emitted by a sample 16 while blocking unwanted wavelengths, such as excitation light or ambient noise. The presence of the emission filter 14 is for fluorescence imaging. The emission filter 14 may be integrated into the sensor 12 or may be a distinct layer applied over the sensor 12.

[0068] Referring still to FIG. 19, the on-chip imaging system 10 includes a microlens array on a substrate, together referred to herein as a microlens substrate 18. The microlens substrate 18 includes an70647- WO-PCT / TECH-2025 - 15array of microlenses which may be, for example, glass or another transparent material having a desired refractive index. The microlenses may be dispersed in a gel matrix and evenly dispersed on the glass substrate. Suitable gels for the gel matrix include, but are not limited to, polydimethylsiloxane (PDMS), bovine gelatin, agarose, sodium alginate, acrylamine, and matrigel. The gel serves as an adhesive layer to secure the microlenses in place on the substrate. The gel matrix may optionally further include a solvent, such as an alcohol, to reduce the viscosity of the gel. Non-limiting example solvents include ethanol, methanol, isopropyl alcohol (IP A), dichloromethane (DCM), or toluene. As one non-limiting example, the microlenses may be borosilicate solid glass, with diameters ranging from 150 to 180 micrometers and a refractive index of 1.48, and the microlenses may be dispersed in a gel matrix of PDMS. However, other combinations are possible and encompassed within the scope of the present disclosure.

[0069] The microlens substrate 18 may be made by spin coating a thin, uniform layer of the gel onto the substrate. This allows for even spreading of the microlenses on the substrate, resulting in the microlenses being affixed to the substrate by the gel but not submerged in the gel (as verified in FIG. 14C).The gel may optionally be allowed to semi-dry for a period of time, ranging from about 10 minutes to about 30 minutes, before the microlenses are spread on top of the gel-coated substrate and allowed to dry.Allowing the gel to semi-dry before spreading the microlenses on top of a thick layer (>>50 microns) of gel creates a flexible microlens substrate whereas spreading the microlenses on glass with a thin layer (<50 microns) of PDMS creates a solid (not flexible) microlens substrate. The flexible microlens substrate 18a and solid microlens substrates 18b are each shown in the photographs in FIG. 19.

[0070] As one non-limiting example method to fabricate a flexible microlens substrate 18a, a mold is prepared on a glass surface, and a thick layer of PDMS is poured into the mold. The PDMS is allowed to partially cure to achieve a semi-dry state. Subsequently, microlenses are evenly distributed over the surface to form a monolayer. The substrate is then fully cured either under ambient conditions or in an oven. After complete curing, the flexible substrate is carefully peeled off from the glass mold. As one non-limiting example method for fabricating the solid substrate 18b, a thin layer of PDMS is evenly spread on a glass surface using a spin coater. Once the PDMS layer reaches a semi-dry state, microlenses are added to its surface. The substrate is then fully cured under similar conditions to finalize the solid microlens substrate 18b.

[0071] The microlens substrate 18 serves to focus the light emitted from the sample following excitation. The microlens substrate 18 is critical for high resolution and sensitivity, as demonstrated by at least FIGS. 4A-4C and FIGS. 20A-20C. FIGS. 20A-20C show a comparison showing the resolution of the bright field on-chip microscope. FIG.20A shows a conventional light field microscope image of group 9 in the test target. FIG. 20B shows on-chip bright field image of group 9 in the test target without a microlens substrate. FIG.20C shows an on-chip bright field image of group 9 in the test target using a70647- WO-PCT / TECH-2025 - 15microlens substrate 18. The resolution of the on-chip microscope is less than 1 pm. FIGS. 4A-4C show the imaging of a 1 pm fluorescent particle using both the on-chip imaging system as a fluorescence microscope and a conventional fluorescence microscope. The sensitivity of the on-chip imaging system 10 is increased incredibly by the presence of the microlens substrate 18.

[0072] Referring again to FIG. 19, a sample 16 to be analyzed may be placed on the microlens substrate 18. As will be described in more detail below, excitation light is passed through the sample 16, focused by the microlens substrate 18, filtered by the emission filter 14, and collected and processed by the sensor 12.

[0073] Referring still to FIG. 19, the on-chip imaging system 10 includes a source 22, 24 of excitation light and a prism 20. The excitation light may be fluorescene excitation light 22 or bright field illumination 24. The source 22, 24 of the excitation light may be a laser, LED, or other suitable light source. The prism 20 serves to channel the excitation light from the source 22, 24 onto the sample 16 and to reduce unwanted excitation light interference. When the on-chip imaging system 10 is being used for fluorescence microscopy, the excitation light is directed through the prism 20 toward the sample 16 at an angle greater than a critical angle to reject excitation light at the bottom facet of the sample 16 (i.e., to reject unwanted light after sample excitation). For bright field imaging, the sample is illuminated vertically from the top. When the on-chip imaging system 10 is being used for bright field illumination, the excitation light is directed through the prism 20 and toward the sample 16 vertically from the top instead of at an angle. This allows for high-contrast imaging.

[0074] Referring now to FIG. 22, the on-chip imaging system can be configured into a wearable microscope. FIG.22 shows a schematic of the wearable microscope. This device employs light sources, such as lasers or light-emitting diodes (LEDs) to excite the sample. Upon illumination, some of the light interacts directly with the sample, while a portion is coupled into the underlying substrate. The evanescent wave, along with the light that leaks out from the substrate, also contribute to the excitation of the sample. This multi-faceted approach enhances the overall excitation efficiency.

[0075] Referring still to FIG. 22, the emitted fluorescence from the sample is subsequently collected by the substrate and focused onto a CMOS camera, ensuring high-resolution imaging. The camera is controlled using a microcontroller, such as an ESP32. The data can be transferred wirelessly via wifi or Bluetooth or USB. A battery, such as a lithium polymer battery (3.7V 2000 mAh), is used to power the microcontroller, camera, and light sources.

[0076] FIG. 23 shows an illustrative example of the wearable system’s capability, presenting an image of a fluorescent sample composed of 1 -micron fluorescent beads, taken using an example wearable microscope. Additionally, FIGS.24A-24B show a photograph of an example wearable device (FIG.24A) with certain components annotated and a photograph of the wearable device in an operational state (FIG.70647- WO-PCT / TECH-2025 - 1524B).

[0077] Using the on-chip imaging system described herein, the interactions between anti-cancer drugs and breast cancer cells have been monitored as an example application. Specifically, as described in the examples herein, changes in oxidative stress were tracked using the fluorescent probe 2’, 7’-dichlorofluorescin diacetate (DCF-DA), which is highly sensitive to reactive oxygen species. By introducing chemotherapeutic drugs, such as doxorubicin and hydrogen peroxide, at varying concentrations, their effects on the cancer cells over time were able to be observed. The system captured dynamic changes in fluorescence intensity across hundreds of cells simultaneously, allowing the monitoring of the onset of oxidative stress and necrosis. Additionally, calcium imaging was successfully performed, demonstrating the system’ s versatility in capturing calcium flux dynamics in response to external stimuli such as calcium chloride and calcium ionophore. In another experiment, a microfluidic device was used to create a coculture model that simulates the tumor microenvironment, allowing for the study of cell interactions in a controlled manner, and to study endothelial cell permeability for different dyes. This approach demonstrates the ability of the on-chip imaging system as a powerful tool for high-throughput screening of anti-cancer drugs, offering the ability to accelerate the drug development process.

[0078] The on-chip imaging system is a multimodal on-chip microscope that can be operated in fluorescence or transmission (brightfield) mode. Integrating a 2D microlens array provides for a wide field-of-view, high resolution (way beyond the diffraction limit), and unmatched sensitivity, all packaged in a portable and low-cost device, making advanced imaging more accessible than ever. Its multi-modal capabilities allow for both brightfield and fluorescence imaging, while an integrated cytometry function opens doors for cell counting applications. The on-chip imaging system is effectively one of the cheapest and easiest to use microscopes (and cytometers) in the world and has a very large field of view which allows for high-throughput measurements. As described in the examples herein, experiments have been performed with the on-chip imaging system on chemical imaging in cells, drug screening, and imaging in vitro tissue models. The two dimensional microlens array can drastically improve the resolution and sensitivity of any optical imaging device incorporating the on-chip imaging system, including wearables.

[0079] The on-chip imaging system can be used in a wide variety of applications, including, but not limited to: point of care diagnostics, enabling rapid and accurate detection of pathogens or cellular abnormalities, improving patient care in remote or resource-limited regions (for example, detecting malaria by imaging blood smears, or detecting cancer by imaging biopsy samples, or detecting pathogens for infectious disease with imaging); environmental monitoring, facilitating quick and effective detection of microorganisms or pollutants in water or soil samples; agriculture applications, allowing for the analysis of plant cells or pathogens to assess crop health, supporting sustainable farming practices; research labs, enabling flexible, high-throughput experiments with on-site data collection, entrancing research capabilities70647- WO-PCT / TECH-2025 - 15without relying on traditional laboratory constraints; wearable devices for monitoring analytes in skin, blood, sweat, or other bodily fluids; and educational activities, providing students and researchers in resource-limited instututions with access to sophisticated imaging and cytometry tools, enabling learning and innovation through hands-on experience. The on-chip imaging system can provide access to at-home diagnostics. This accessibility may transform personal healthcare, making early diagnostics available and affordable to a broader population, thereby improving health outcomes and reducing the burden on traditional healthcare facilities. The on-chip imaging system can be used for imaging (microscope), sensing (biosensor), and counting particles (cytometer). The on-chip imaging system can be easily implemented as a wearable device as well.

[0080] In sum, the on-chip imaging system described herein has the advatanges of ease of implementation, ease of fabrication, high signal, high resolution, and ultra-low cost. The system has a low cost and high resolution with a wide field of view and a low-cost flow cytometer for counting. The system is multi-modal in that it may be used with bright field illumination or for fluorescence. The system is thus multifunctional, combining bright field, fluorescence, and flow cytometry in a single device. The system has a wide field of view and high throughput, allowing imaging and analysis of a large field of view or in continuous flow. The system is cost-effective yet robust, significantly less expensive than lab-scale microscopes or cytometers. Furthermore, the on-chip design means fewer external alignments, simpler workflows, and easier maintenance than lab-scale devices.

[0081] EXAMPLES

[0082] Example I - Low cost, high sensitivity, high resolution, multi-modal imaging, and cytometry on a chip

[0083] The on-chip imaging system combines a wide field of view with high resolution and signal-to-noise ratio, allowing for the imaging of hundreds of cells simultaneously in a cost-effective manner. This platform integrates a two-dimensional microlens array below the sample, efficiently collecting fluorescence emissions. The excitation is applied at an angle exceeding the critical angle through a prism interface, reducing unwanted excitation fight interference. Important experimental parameters, such as the distances between the sample, microlens, and sensor, were tailored for maximum imaging sensitivity. This example demonstrates the utility of this imaging system for studying the effects of chemotherapeutic drugs on BT-20 beast cancer cells. Specifically, the levels of reactive oxygen species (ROS) were assessed using 2’, 7’-dichlorofluorescin diacetate (DCF-DA) as a fluorescent probe. The cancer cells, incubated with DCF-DA, were treated with doxorubicin and hydrogen peroxide at various concentrations, and changes in fluorescence intensity were tracked over time to observe cellular responses. Additionally, calcium imaging was demonstrated using Fluo-3, a calcium-sensitive dye, to observe the rapid influx of calcium ions into BT-20 breast cancer cells after treatment with calcium chloride and calcium ionophore. Another70647- WO-PCT / TECH-2025 - 15experiment was performed using a microfluidic chip to model in vitro tissue structures that stimulate the tumor microenvironment. FITC -labeled dextran and fluorescent nanoparticle with a diameter of 42 nm were introduced into the microfluidic device to assess endothelial cell permeability. By monitoring changes in fluorescence intensity over time, the permeability was quantified, offering important insights into the dynamics of tissue barrier function and the effectiveness of endothelial barriers under varying conditions. This on-chip fluorescence microscope effectively captures real-time dynamic changes in fluorescence intensity across hudnreds of cells, providing a powerful high-throughput method for monitoring oxidative stress, studying calcium signaling dynamics, and evaluating cell viability. The approach has the ability to significantly improve the efficiency of drug screening and expedite the development of new therapeutic agents.

[0084] Materials and methods

[0085] Materials

[0086] Borosilicate solid glass microlenses with a refractive index of 1.48 were purchased from Cospheric (Santa Barbara, CA, USA). Copolymer microlens suspensions and green fluorescence polymer microlenses with a size of 1 pm were acquired from Thermo Scientific. 2 ’,7 ’-Dichlorofluorescein diacetate (DCF-DA) was purchased from Sigma- Aldrich. Propidium iodide (PI) was used for cell viability assessment. The Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), and streptomycin / penicillin for cell culture were obtained from Gibco (Terhmo Fisher Scientific). All solutions were prepared in ACS reagent-grade water, and all chemicals and materials were used as received.

[0087] Preparation of microlenses

[0088] To fabricate a monolayer of microlenses on a substrate, various types of microlenses were distributed evenly on a clean glass coverslip. The microlenses included dry borosilicate glass with diameters ranging from 150 pm to 180 pm and a refractive index of 1.48, glass spheres with a diameter of 0.5 mm and a refractive index of 1.5, as well as polymeric microlenses with a refractive index of 1.59 and a diameter of 42 pm suspended in water. Ethanol was used to moisten the dry borosilicate glass and glass spheres before spreading them evenly on the coverslip. To avoid clumping and ensure a uniform monolayer, a handheld plasma generator was applied for 50 seconds before dispersing the microlenses. Finally, the particles were left to air dry, resulting in a monolayer substrate of microlenses on the coverslip.

[0089] Design of the on-chip imaging platform

[0090] The on-chip imaging configuration utilized an optoelectronic sensor array to record the fluorescent emission from target micro-objects. In this platform, for detection of the fluorescence signal, 8-megapixel complementary metal-oxide-semiconductor sensor (CMOS), with pixel size of 1.12 pm and the active imaging area of 3.68 mm x 2.76, was used. Furthermore, to better control this vertical distance, the sensor was decapped such that the protective cover glass of the chip was removed.70647- WO-PCT / TECH-2025 - 15

[0091] In this imaging platform, an inexpensive filter (from Roscolux) was directly positioned onto a CMOS sensor to reject excitation light and then microlens substrate was placed on top of the filter.Fluorsecent cells / paerticles were placed in close contact on top of the microlens substrate. To create the required dark-field background, in addition to absorption filter, a glass rhomboid prism (Edmund Optics) was used to create total-inernal reflection (TIR) for the excitation light at the surface of the coverslip which hosts samples of interest. The fluorescent cells / particles were then pumped at their proper excitation wavelength using a laser with the wavelength of 450 nm and 304 nm through the side facet of a prism as illustrated in FIG. 1. Fluorsecent emission from the excited cells / particles, however, does not entirely obey TIR and therefore can be directly detected over the entire FOV of the sensor array.

[0092] Image formation of microlens

[0093] Microlens-assisted imaging techniques overcome the diffraction limitations of conventional optical microscopes by employing mechanisms such as the generation of enlarged virtual images, photonic nano-jets, and the conversion of evanescent waves into propagating waves. Image formation with microlenses depends on factors including microlens size, refractive index, and the distance between the microlens and the object. When the object lies beyond the focal length of the microlens, a real image that is smaller and inverted is formed. Conversely, if the object is within the focal length, an enlarged virtual image is formed, maitaining the same orientation as the original object. These behaviors were verified through simulations and experiments using 1 gm polystyrene particles.

[0094] Evaluating photon density on CMOS sensor

[0095] It is advantageous to optimize the photon density on the CMOS sensor to maximize the sensitivity of the imaging system and improve the signal-to-noise ratio. Photon density, defined as the number of photons per unit area on the sensor, depends on the solid angle of light collected by the lens and the area covered by the focused light on the sensor plane.

[0096] The solid angle (Q) can be calculated using the following formula, which takes into account the geometry of the system:where o is the distance between the object and the microlens, and R is the radius of the microlens. The general formula for the focal length (f) of a spherical lens with a nonnegligible thickness, where n is the lens refractive index, isf -nR‘ (2)

[0097] The general formula for the image distance in the case of a spherical lens where thickness is70647- WO-PCT / TECH-2025 - 15nonnegligible is identical to the thin lens equation:Here, I is image distance from the microlens.

[0098] The planar angle a can be calculated as the inverse tangent of the ratio between R and i:

[0099] The area occupied by the focused light is the area of the circle formed by the light on the sensor plane. The radius r of this circle is:

[0100] The area A of the cicle is:A = TTF2(6)

[0101] Finally, the photon density can be calculated by the following equation, as a function of the solid angle of the lens and the area occupied by the focused light. By fine-tuning the configuration to minimize A, it is ensured that the photons converse effectively, thereby enhancing the photon density (p) on the CMOS sensor, as indicated by:

[0102] The photon density calculations were performed for different values of o and d (sensor distance from the microlens). For this calculation, the microlens with refractive index of 1.48 and radius of 75 pm was considered. These data were compiled and plotted, shown in FIG. 2. The maximum value of p can then be expressed as a function in terms of two variables, o and d. This is done by setting d to the image distance i in equation (3). This leads to a zero value for r from equation (5). This causes the area covered by the light from the lens to be 0 according to equation (6). This means that the area occupied by the fluorescence emission photons is an infinitesimal point. Along the curve on the o, d plane (equation (8)), all p values will be infinite, as all the emitted photons will converge to a single point on the sensor.70647- WO-PCT / TECH-2025 - 15

[0103] Cell line and preparation

[0104] BT-20 breast cancer cells were used in this example. The BT-20 cells were grown as an adherent monolayer in culturing flasks containing DMEM supplemented with 1 % streptomycin / penicillin and 10% FBS. The cells were incubated at 37 °C in a humidified atmosphere with 5% CO2 until they reached the desired confluency.

[0105] Cell experiments with DCF in presence of doxorubicin and H2O2

[0106] BT-20 breast cancer cells were incubated with DCF-DA, a cell-permeant sensor dye that is a reduced form of fluorescein used as an indicator of reactive oxygen species (ROS) in cells. The nonfluorescent H2DCFDA is converted into the fluorescent 2’, 7’ -dichlorofluorescein (DCF) upon cleavage of its acetate groups by intracellular esterases. To prepare the working concentration of DCF-DA, 1.25 mg of DCF-DA was dissolved in 0.5 mL dimethyl sulfoxide (DMSO), followed by the addition of 2 mL distilled water to make a 1 mM solution. From this stock, concentrations of 10-30 pM were prepared for cell incubation. BT-20 cells were incubated with DCF-DA for a total of 2 hours, during which doxorubicin (DOX) treatment at a concentration of 50 pM was introduced at different time points. Specifically, DOX was added during the final phase of the 2-hour incubation: 15, 30, and 45 minutes before the end of the incubation period, for different cell samples. The fluorescence intensity was recorded after each incubation to assess ROS levels, allowing for the effects of increasing DOX exposure time to be evaluated while keeping the total incubation time constant. The same method was used to incubate the cells with hydrogen peroxide (H2O2) for the time interval of 15, 30, and 60 minutes. Cell viability was also assessed by staining the cells with 1 pM propidium iodide (PI) to observe their viability during the experiment.

[0107] Cell experiments with Fluo-3 in presence of calcium chloride and calcium ionophore

[0108] Cells were incubated with 5 pM Fluo-3, a calcium-sensitive fluorescent dye, for one hour in the dark to ensure sufficient uptake of the dye without premature activation. Subsequently, calcium chloride (8 mM) and calcium ionophore (5 pM) were added to initiate calcium influx for imaging.

[0109] Establishing tumor-EC co-culture in the microfluidic device

[0110] A microfluidic device was utilized to establish a co-culture model that mimics the tumor microenvironment. To prepare the microfluidic device, it was first rinsed thoroughly with deionized (DI) water and degassed using nitrogen gas at 7 psi for 15 minutes to eliminate any trapped air bubbles.Following this, the tissue compartment was pre-treated with 5% Matrigel solution prepared in serum-free DMEM and incubated at 37 °C for 20 minutes. The device was then perfused with fibronectin (200 pm / mL in phosphate-buffered salien (PBS)) and incubated for 2 hours at 37 °C to promote cell attachment by generating a homogeneous coating. All devices were rinsed with fresh endothelial cell growth medium (EGM) prior to cell seeding.

[0111] Human umbilical vein endothelial cells (HUVECs) were prepared at a concentration of 3.5 x70647- WO-PCT / TECH-2025 - 15IO7cells / mL and introduced into the vascular channels using a precision syringe pump at a flow rate of 5 pL / min. This process was repeated to seed both vascular channels, and the microfluidic devices were then incubated for 4 hours to facilitate cell attachment. After this initial attachment, the vascular channels were manually rinsed with EGM to flush out debris and unattached cells. Subsequently, continuous perfusion was established in both vascular channels using an automated syringe pump (PHD ULTA, Harvard Apparatus, Holliston, MA) to maintain physiological flow conditions with shear stress between 0 and 1.75 dyne / cm2, which mimics in vivo microvascular shear rates.

[0112] After a 24-hour incubation to allow HUVEC adhesion and alignment along the direction of flow, BT-20 cell line suspended in 5% Matrigel in serum-free medium, were seeded into the tissue compartment using a precision syringe pump. The BT-20 cells were seeded at a concentration of 107cells / mL, and the device was incubated for an additional 48 hours to allow the formation of a 3D tumor structure. Continuous medium perfusion was maintained through the vascular channels, simulating blood flow and nutrient exchange, while bolus injections of fresh media were administered every 8 hours to the tissue compartment to support tumor cell growth.

[0113] The imaging capabilities of the on-chip fluorescence microscope were evaluated by imaging 1 pm fluorescent particles. In this setup, the fluorescent particles were placed in close contact with the microlens substrate, forming a sandwich structure that positioned the particles within the focal length of the microlenses. This configuration allowed the formation of larger virtual images of the particles, which were captured by the sensor. Additionally, images of the fluorescent particles were captured without using the microlens substrate, and these images showed very blurry signals that made it impossible to distinguish or identify any individual particles. The images obtained using the on-chip fluorescence microscope were then compared to those captured with a conventional fluorescence microscope. The on-chip system demonstrated comparable resolution to the conventional microscope while offering a significantly larger field of view, effectively validating the utility of the approach for imaging small fluorescent particles. The results of imaging fluorescent particles are shown in FIGS. 4A-4C.

[0114] The results obtained from the developed on-chip fluorescence microscope have implications for anti-cancer drug screening. For imaging the BT-20 cells, the experimental setup involved seeding the cells in a modified petri dish. To accommodate the prism for excitation, the walls of the petri dish were cut to create a flat structure with a glass bottom in the center. The cells were then seeded in this modified petri dish and allowed to grow. When the cells were ready for imaging, the glass bottom section of the petri dish was filled with PBS and sealed with a glass cover to maintain a stable environment. A drop of immersion oil was placed on top of the cover glass to match the refractive index, and a prism was positioned directly on top. Additionally, the microlens substrate was prepared on a coverslip and placed under the glass-bottomed petri dish to collect the fluorescence emission from the cells effectively. Next, an experiment was70647- WO-PCT / TECH-2025 - 15performed where BT-20 cells were incubated with both DCF-DA and doxorubicin (DOX). The incubation time for DOX was varied, with cells being incubated with 50 pM DOX for 0 min, 15 min, 30 min, 45 min, and 60 min. After each incubation period, the fluorescence intensity was measured. It was observed that the fluorescence intensity increased with longer DOX incubation times. This increase in intensity is attributed to the fact that doxorubicin induces oxidative stress in the cells, leading to the production of reactive oxygen species. As the cells were incubated with DOX for longer durations, more ROS were generated, which in turn increased the fluorescence signal of the DCF-DA probe (FIG. 5A, first row). In addition, the intensity changes were monitored using conveitonal fluorescence microscopy, which showed a similar trend to the results obtained with the on-chip fluorescence microscope (FIG. 5A, second row). FIGS.5B-5C also present the graph of fluorescence intensity versus time for both on-chip fluorescence microscopy and conventional fluorescence microscopy, showing similar trends in intensity changes over time. To assess cell viability during the experiment, the cells were also stained with 1 pM propdium iodide (PI), and the results, shown in FIG. 5A (third row), indicate that the cells remained viable throughout the imaging period. Imaging with PI was performed using conventional fluorescence microscopy in TXR channel, confirming cell viability during the entire agreement. Subsequently, the same method was performed using hydrogen peroxide instead of DOX. The BT-20 cells were also incubated with hydrogen peroxide, and the fluorescence intensity was measured at different incubation times (15 min, 30 min, and 60 min). As shown in FIG. 6, a similar increase in fluorescence intensity was observed, which was due to the increased level of oxidative stress caused by hydrogen peroxide. This further demonstrates the capability of the system to monitor oxidative stress in real time, whether induced by chemotherapeutic agents such as DOX or other oxidative stress inducers such as hydrogen peroxide. This demonstrates the ability of the system to effectively monitor drug-induced oxidative stress in real time, providing valuable insights into cellular responses to chemotherapeutic agents.

[0115] Another experiment involved imaging changes in calcium levels within the cells. Fluo-3 effectively measures intracellular calcium levels by binding to free calcium ions, resulting in a fluorescence signal. Following incubation with fluo-3, calcium chloride and calcium ionophore were added to the cells, and imaging began immediately. Calcium ionophore plays an important role in facilitating the transport of calcium ions across cell membranes, significantly increasing intracellular calcium concentration, which led to an immediate and notable increase in fluorescence intensity, indicating a rapid influx of calcium ions into the cells. FIG.7 illustrates the time series of calcium influx in a specific BT-20 cell upon addition of calcium ionophore. The top images in FIG.7 show initial and final fluorescence images of the cells taken after the addition of calcium chloride and calcium ionophore. The bottom image in FIG. 7 shows a graph of fluorescence intensity for the marked cell over time, demonstrating a gradual increase in intracellular calcium concentration following ionophoire treatment, indicating effective calcium transport across the cell70647- WO-PCT / TECH-2025 - 15membrane.

[0116] Endothelial cell permeability of molecules and nanoparticles

[0117] Another experiment conducted involved in vitro tissue modeling with a microfluidic chip that simulates tissue structures in the body. FITC 4-kilodalton dextran dye with a concentration of 2.5 mM was infused through the vascular section at a controlled flow rate and the fluorescence intensity changes were monitored over time and position to evaluate permeability across the endothelial barrier. Fluorescence intensity was imaged for half an hour at a flow rate of 1 pL / min using an automated syringe pump. The results are shown in FIGS.8A-8C. Another fluorescent nanoparticle with a diameter of 42 nm was injected through the vascular section, and the changes in fluorescence intensity were imaged. The permeability of endothelial cells for both was then compared. The results are shown in FIG. 9.

[0118] Due to the small size of the dextran dye, their permeability across the endothelial cells is significantly greater than that of the fluorescent nanoparticles. In another experiment BT-20 cells incubated with Fluo-3 were exposed to calcium ionophore and calcium chloride through the vascular section, and the permeability of endothelial cells for their flow into the central tissue compartment was imaged by observing changes in fluorescence intensity. FIGS. 10A-10B shows the flow of calcium chloride and calcium ionophore from the vascular section to the central tissue compartment, visualizing their permeability across the endothelial cells.

[0119] Conclusion

[0120] In this example, a cost-effective on-chip imaginge systed useful as a fluorescence microscope with a wide field of view and which can facilitate high-throughput imaging for anti-cancer drug screening has been demonstrated. The system’s use of a two-dimensional microlens array allows for efficient collection of fluorescence emission, producing high-resolution images of hundreds of cells simultaneously. This approach overcomes the limitations of conventional fluorescence microscopy, including its small field of view, complexity, and high cost. By adjusting key parameters such as the sample-to-microlens and micro! ens-to-sensor distances, a maximum sensitivity for monitoring oxidative stress in BT-20 breast cancer cells is achieved. The results show that the system effectively captured dynamic fluorescence intensity changes induced by chemotherapeutic agents such as doxorubicin and hydrogen peroxide, highlighting the induction of oxidative stress and necrosis in the cells. The increase in fluorescence intensity with longer incubation times demonstrates the system’s capability to track real-time cellular responses. Additionally, calcium imaging was successfully performed, demonstrating the system’s versatility in capturing calcium flux dynamics in response to external stimuli such as calcium chloride and calcium ionophore. Moreover, the experiments using a microfluidic device to simulate the tumor microenvironment demonstrate the system’ s ability to investigate endothelial cell permeability. The microfluidic device allowed for tire creation of a controlled flow environment, mimicking physiological70647- WO-PCT / TECH-2025 - 15conditions, which facilitated the assessment of the permeability of endothelial barriers for different fluorescent dyes. This setup provided quantitative insights into how various agents influence endothelial integrity and transport dynamics. As a result, this on-chip fluorescence microscopy technique presents a powerful tool for high-throughput screening of anti-cancer drugs, providing an efficient and scalable solution to evaluate therapeutic efficacy and understand cellular responses. The combination of wide field of view, high resolution, and cost-effectiveness makes this system highly useful for accelerating the drug development process, ultimately contributing to advances in cancer treatment research.

[0121] Example II - Flow cytometry application of on-chip fluorescence microscope

[0122] The feasibility of performing flow cytometry and cell counting using the on-chip imaging system as a fluorescence microscope was evaluated. For this purpose, an imaging system was used to visualize particles moving through capillaries. As shown in FIG. 11, the experimental setup for on-chip fluorescence microscopy involved a laser to excite the sample, with excitation directed at an angle above the critical angle to effectively reject unwanted light after sample excitation. Imaging was performed using a CMOS sensor, with an absorption filter from Roscolux positioned on top of the sensor to block residual excitation light. To enhance the collection of fluorescence emission, which is isotropic, a microlens substrate was placed on top of the absorption filter. The aim was to image 1 -micrometer fluorescent particles moving through a capillary using this setup to characterize on-chip fluorescence microscope performance. A thin capillary with an outer diameter of 80 micrometers and an inner diameter of 50 micrometers was used to flow the particles, and it was placed in close contact with the microlens substrate. A prism was positioned on top of the capillary to channel the excitation light effectively on the capillary. Multiple capillaries can be placed next to each other to image numerous fluorescent particles within a single field of view, demonstrating the high-throughput capability of the on-chip fluorescence microscope.

[0123] As shown in FIG. 12, 1 pm fluorescence particles pass through the capillary and their movement was imaged using this set up.

[0124] Making permanent microlens substrate using different gels

[0125] A substrate containing microlenses was fabricated to enhance optical imaging capabilities. The microlenses used were made of borosilicate solid glass, with diameters ranging from 150 to 180 micrometers and a refractive index of 1.48. To ensure effective virtual imaging, it is important that the microlens substrate is in close contact with the sample, without any gap. This configuration allows the sample to be positioned within the focal length of the microlens, enabling the microlens to form a magnified virtual image of the object. FIG. 13 shows a schematic of image formation using microlens. To fabricate a functional microlens substrate, the microlenses should not be submerged into the subtrate, and the top surface should remain free of gels to allow for ideal imaging conditions.

[0126] Different gels can affix the microlenses to the coverslip, including PDMS, bovine gelatin.70647- WO-PCT / TECH-2025 - 15agarose, sodium alginate, and acrylamide, which serves as an adhesive layer to secure the microlenses in place. Some of the gels, such as PDMS, are so viscious that they need to be made thinner. For this purpose, different solvents such as isopropyl alcohol (IPA), dichloromethane (DCM), or toluene can be used. By using a spin coater, a very thin, uniform layer of gels can be created, which then allows for the even spreading of microlenses on the substrate. This method improves the consistency of the gel layer thickness, ensuring that it was not submerged and was properly affixed for imaging purposes.

[0127] To make the microlens substrate, gel was spread on a coverslip using a spin coater, and microlenses were spread on another coverslip by making them wet with ethanol. A sandwich structure was attempted to be made, so that the microlenses could transfer from then ethanol-treated coverslip to the other coverslip with the gel substrate and be fixed in place. However, this technique did not work as not all of the microlenses transferred to the other coverslip with the gel. In another method, a substrate was made with any kind of gel and allowed to semi-dry at different time intervals (10 min, 20 min, 30 min). After semidrying the gel substrate, the microlenses were spread on top of the gel substrate and allowed to dry. Using these approaches, two different kinds of substrates were formed as described below.

[0128] Solid microlens substrate

[0129] This involved forming a thin layer of gel with microlenses on top, which remained fixed on a coverslip to create a solid substrate. FIGS. 14A-14C show different views of thin microlens substrate made by PDMS. FIG. 14C is a cross sectional image of a substrate to verify that the microlens is not submerged in the PDMS layer.

[0130] Flexible substrate

[0131] This involved creating a substrate using a thick layer of gel, allowing it to semi-dry, and then spreading the microlenses on top. Once the substrate was completely dry, it was peeled off from the glass, resulting in a flexible microlens substrate. FIG. 15A shows a flexible substrate.

[0132] Cell death pathway classification using on-chip bright field microscope

[0133] Identification of cell death mechanisms is important for various applications such as understanding cell signaling, interactions with pathogens, therapeutic responses, drug discovery, drug resistance, and the development of disease such as cancers and neurodegenerative disorders. In multicellular organisms, cell death typically occurs through either apoptosis, a programmed mechanism to maintain homeostasis, or necrosis, which is triggered by factors such as infection, injury, or hypoxia. Each cell death process has unique morphological characteristics that distinguish it from others. Apoptosis, a non-inflammatory process, shows features such as chromatin condensation, cell membrane shrinkage, blebbing, and the formation of apoptotic bodies, whereas necrosis is marked by cell membrane rupture and leakage of cytoplastmic contents triggering inflammation. In addition to apoptosis and necrosis, other non-apoptotic cell death processes such as necroptosis, pyroptosis, methuosis, and ferroptosis have also been70647- WO-PCT / TECH-2025 - 15identified, each with distinct hallmarks. In this example, the on-chip imaging system was used as an on-chip bright field microscope to classify cell death pathways according to the morphological changes observed in cells. Different cell death processes were induced using known cytotoxic agents, and a deep learning -based approach can be emplouyed to automatically classify these mechanisms. For bright field imaging, the microlens substrate was located on top of the CMOS sensor to collect transmitted light from the sample, and the cells were seeded on a specially designed petri dish without walls. A prism was placed on top of the sample, and an LED was used to illuminate the sample vertically from the top. The bright field imaging setup is shown in FIG. ISA. This setup allowed for high-contrast imaging to observe morphological changes directly, providing important information on cell death pathways under different treatment conditions.

[0134] For this purpose, the BT-20 cells were treated with doxorubicin (DOX) at a concentration of 50 pM to induce apoptosis and with 5 mM hydrogen peroxide to induce necrosis. The on-chip bright field microscope was used to observe the distinct morphological changes resulting from each treatment.

[0135] FIG. 17 illustrates the morphological changes of BT-20 cells treated with DOX (first row) and hydrogen peroxide (second row).

[0136] Additionally, methuosis, which can be induced using the compound 3-(5-methoxy-2-methyl- I H-mdol-3-yl )- l-(4-pyridinyl )-2-propene- 1 -one (MOMIPP) (10 pM), may be studied in the same manner. Methuosis is characterized by the formation of large intracellular vacuoles and is distinct from apoptosis and necrosis. FIG. 21 shows an on-chip bright field microscope application to distinguish apoptosis from necrosis. Observations using conventional microscopy indicate that treating BT-20 cells with MOMIPP leads to extensive vacuolization, ultimately resulting in methuosis, a non-apoptotic, vacuolization-driven form of cell death. FIG. 18 illustrates the imaging of a fixed cell with 10 pM MOMIPP using the on-chip bright field microscope.

[0137] Certain embodiments of the devices, systems, compositions, and methods disclosed herein are defined in the above examples. It should be understood that these examples, while indicating particular embodiments of the invention, are given by way of illustration only. From the above discussion and these examples, one skilled in the art can ascertain the essential characteristics of this disclosure, and without departing from the spirit and scope thereof, can make various changes and modifications to adapt tire devices, systems, compositions, and methods described herein to various usages and conditions. Various changes may be made and equivalents may be substituted for elements thereof without departing from the essential scope of the disclosure. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the essential scope thereof.

Claims

70647- WO-PCT / TECH-2025 - 15CLAIMSWhat is claimed is:

1. An on-chip imaging system comprising:a substrate;a monolayer of microlenses distributed on the substrate;a sensor comprising an array of photodetectors configured to detect light, wherein the substrate is disposed on the sensor;a light source configured to emit excitation light toward a sample; anda prism configured to receive the excitation light from the light source and transmit the excitation light to the sample.

2. The on-chip imaging system of claim 1 , wherein the substrate comprises glass.

3. The on-chip imaging system of claim 1, wherein the prism is a glass rhomboid prism.

4. The on-chip imaging system of claim 1 , wherein the prism is configured to transmit the excitation light to the sample at an angle greater than a critical angle.

5. The on-chip imaging system of claim 1, wherein the microlenses comprise borosilicate glass having a diameter in a range of from about 150 to about 180 micrometers and a refractive index of about 1.48.

6. The on-chip imaging system of claim 1, wherein the microlenses comprise glass spheres with a diameter of about 0.5 mm and a refractive index of about 1.5.

7. The on-chip imaging system of claim 1, wherein the microlenses comprise a polymeric material with a refractive index of about 1.59 and a diameter of about 42 gm.

8. The on-chip imaging system of claim 1 , further comprising an emission filter configured to select vely wavelengths of light while blocking excitahon light.

9. The on-chip imaging system of claim 1, wherein the microlenses comprise borosilicate glass, glass spheres, polymeric microlenses, or a combination thereof.

10. The on-chip imaging system of claim 1, wherein the substrate is flexible.70647- WO-PCT / TECH-2025 - 1511. The on-chip imaging system of claim 1 , further comprising an emission filter disposed between the sensor and the substrate.

12. The on-chip imaging system of claim 1, wherein the monolayer is evenly distributed on the substrate.

13. The on-chip imaging system of claim 1 , further comprising a light source configured to emit excitation light at a wavelength capable of exciting fluorophores in a sample.

14. The on-chip imaging system of claim 1, further comprising a signal processing module configured to transfer data to an external system.

15. The on-chip imaging system of claim 1, wherein the on-chip imaging system is in the form of a wearable microscope.

16. An on-chip imaging system comprising:a sensor chip comprising an array of photodetectors configured to detect light and convert the detected light into electrical signals;a substrate on the sensor chip; andan array of microlenses affixed to the substrate with a gel.

17. The on-chip imaging system of claim 16, wherein the gel comprises polydimethylsiloxane (PDMS), bovine gelatin, agarose, sodium alginate, matrigel, or acrylamine.

18. The on-chip imaging system of claim 16, wherein the microlenses are not submerged in the gel.

19. The on-chip imaging system of claim 16, wherein the sensor chip comprises a CMOS sensor.

20. The on-chip imaging system of claim 16, further comprising an emission filter between the sensor chip and the substrate, the emission filter being configured to selectively transmit wavelengths of light while blocking excitation light.70647- WO-PCT / TECH-2025 - 1521. An on-chip imaging system comprising:a CMOS sensor;an emission filter directly on the CMOS sensor; anda microlens substrate directly on the CMOS sensor, wherein the microlens substrate comprises an array of microlenses affixed to a substrate with a gel.

22. A method for conducting on-chip imaging, the method comprising:transmitting excitation light through a prism to a sample;focusing light emitted from the sample through a microlens substrate comprising an array of microlenses;filtering the excitation light through an emission filter; anddetecting the focused light with a sensor chip comprising photodetectors.

23. The method of claim 22, wherein the excitation light is transmitted through the prism and to the sample at an angle greater than a critical angle to reject excitation light at a bottom facet of the sample and to reduce unwanted excitation light interference.

24. The method of claim 22, wherein the excitation light is transmitted through the prism and to the sample vertically from above the sample for bright field imaging.

25. The method of claim 22, wherein the microlense substrate comprises the array of microlenses affixed to a substrate with a gel.

26. The method of claim 25, wherein the microlenses are not submerged in the gel.

27. The method of claim 22, wherein the microlenses comprise borosilicate glass having a diameter in a range of from about 150 to about 180 micrometers.

28. The method of claim 27, wherein the borosilicate glass has a refractive index of about 1.48.

29. The method of claim 22, comprising imaging flowing particles in real-time.

30. Use of an array of microlenses to focus light from an excited sample prior to collecting the light with a sensor chip to conduct on-chip fluorescence microscopy or bright-field imaging.

Images