Scalable projected light sheet microscopy for high-resolution imaging of living and cleared samples

Abstract

A light sheet microscope can be implemented using a projector configured to project an image of a line of light, so that a projection of the line of light forms a sheet of light. A first set of optical components routes the sheet of illumination light into the sample, so that the sheet of illumination light travels through the sample. A first objective lens accepts light from a depth within the sample that corresponds to the sheet of illumination light that travels through the sample, and routes the accepted light towards at least one camera. And a controller varies a thickness of the line of light by sending corresponding data to the input port of the projector.

Description

SCALABLE PROJECTED LIGHT SHEET MICROSCOPY FOR HIGH-RESOLUTION IMAGING OF LIVING AND CLEARED SAMPLESCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This Application claims the benefit of US Provisional Applications 63 / 567,571 (filed March 20, 2024), 63 / 635,060 (filed April 17, 2024), and 63 / 734,280 (filed December 16, 2024), each of which is incorporated herein by reference in its entirety.STATEMENT REGARDING FEDERALLY-SPONSORED RESEARCH

[0002] This invention was made with government support under MH119423 and TR002151 awarded by the National Institutes of Health. The government has certain rights in the invention.BACKGROUND

[0003] Light sheet fluorescence microscopy (LSFM) has emerged as a powerful and versatile approach for biological imaging, enabling prolonged in vivo observations and facilitating high-resolution imaging of large cleared intact organs. These remarkable capabilities are achieved through LSFM's distinctive orthogonal planar illumination and detection geometry, which minimizes energy load and allows use of fast CMOS camerabased wide-field detection. In recent years, several highly optimized LSFM implementations have emerged for high-speed phenotyping of intact cleared organs at high resolution and high-speed live imaging of cells, embryos and functioning nervous systems. However, the substantial costs associated with implementing these systems, and their limited scalability due to complex architectures, hinder their utility for high-throughput imaging applications and impede broader accessibility within the scientific community. To address some of these constraints, several open-source initiatives have emerged, including openSPIM, OpenSpinMicroscopy and mesoSPIM, which aim to optimize the cost and portability of LSFM. Nevertheless, there is still an unmet need for genuinely scalable low-cost systems, with small compact footprints, to facilitate high-throughput applications in biomedical research.SUMMARY OF THE INVENTION

[0004] One aspect of this application is directed to a first light sheet microscope for obtaining images of a sample. The first light sheet microscope comprises a projector, a first set of optical components, a first objective lens, and a controller. The projector is configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector. The projector has an input port and is configured to change characteristics of the projected image based on data that arrives at the input port. The first set of optical components is positioned to route the first sheet of illumination light into the sample, so that the first sheet of illumination light travels through the sample. The first objective lens is positioned to accept light from a depth within the sample that corresponds to the first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens and towards at least one first camera. And the controller is configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector.

[0005] In some embodiments of the first light sheet microscope, the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to output a thick first line of light. In these embodiments, the controller is also further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light. Optionally, in these embodiments, the thin first line of light has a width of one pixel, and the thick first line of light has a width of at least three pixels.

[0006] Some embodiments of the first light sheet microscope further comprise a movable stage that is controllable by the controller. In these embodiments, the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to output a thick first line of light. The controller is also further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light. And the controller is also further configured to obtain multiple low resolution images of respective thick slices of the sample by sending the first data to theinput port of the projector, and instructing the at least one first camera to capture respective images of the sample while the movable stage is set to each of a plurality of different respective locations in a Z direction.

[0007] Optionally, in the embodiments described in the previous paragraph, each pixel in the low resolution images includes at least 4 native-resolution pixels of the at least one first camera that have been binned together. Optionally, in these embodiments, the controller can be further configured to identify a region of interest from the low resolution images, and to subsequently obtain high resolution images of a plurality of thinner slices of the sample that correspond to the region of interest.

[0008] In some embodiments of the first light sheet microscope, the first set of optical components comprises a second objective lens positioned between an output of the projector and the sample. In some embodiments of the first light sheet microscope, the first set of optical components includes only a second objective lens that is positioned between an output of the projector and the sample.

[0009] In some embodiments of the first light sheet microscope, the first set of optical components comprises a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light, and a second objective lens positioned to route the flattened first sheet of illumination light into the sample.

[0010] In some embodiments of the first light sheet microscope, the first set of optical components comprises a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; a second objective lens positioned to route the flattened first sheet of illumination light into the sample; and a tube lens positioned between the scan lens and the second objective lens. Optionally, the first set of optical components in these embodiments can further comprise an electrically tunable lens positioned between the tube lens and the second objective lens.

[0011] In some embodiments of the first light sheet microscope, the first set of optical components comprises a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; and an electrically tunable lens positioned to route the flattened first sheet of illumination light into the sample.

[0012] In some embodiments of the first light sheet microscope, the projector comprises a MEMS laser projector. In some embodiments of the first light sheet microscope, the first objective lens is mounted so that its optical axis is perpendicular to the first sheet of illumination light that travels through the sample.

[0013] Some embodiments of the first light sheet microscope further comprise the at least one first camera.

[0014] In some embodiments of the first light sheet microscope, the image projected by the projector includes a second line of light, so that a projection of the second line of light forms a second sheet of light that is projected out of the projector; the first set of optical components is positioned to route the second sheet of illumination light into the sample, so that the second sheet of illumination light travels through the sample; and the light sheet microscope further comprises an additional objective lens positioned to accept light from a depth within the sample that corresponds to the second sheet of illumination light that travels through the sample, and route that light through a back aperture of the additional objective lens and towards at least one second camera. Optionally, in these embodiments, the second line of light is parallel to the first line of light, and the second sheet of illumination light is parallel to the first sheet of illumination light.

[0015] In some embodiments of the first light sheet microscope, the first sheet of illumination light is oriented parallel to an upper surface of the sample, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample.

[0016] In some embodiments of the first light sheet microscope, the first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°. Optionally, in these embodiments, an optical axis of the first objective lens is perpendicular to the first sheet of illumination light.

[0017] In some embodiments of the first light sheet microscope, the first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample or the lower surface of the sample.

[0018] Another aspect of this application is directed to a second light sheet microscope for obtaining images of a sample. The second light sheet microscope comprises aprojector, a scan lens, a first set of optical components, and a first objective lens. The projector is configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector. The scan lens is positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light. The first set of optical components is positioned to route the flattened first sheet of illumination light into the sample, so that the flattened first sheet of illumination light travels through the sample. The first objective lens is positioned to accept light from a depth within the sample that corresponds to the flattened first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens and towards at least one first camera.

[0019] In some embodiments of the second light sheet microscope, the projector has an input port and is configured to change characteristics of the projected image based on data that arrives at the input port, and the light sheet microscope further comprises a controller configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector.

[0020] Optionally, in the embodiments described in the previous paragraph, the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to output a thick first line of light; and the controller is also further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light. Optionally, in these embodiments, the thin first line of light has a width of one pixel, and the thick first line of light has a width of at least three pixels.

[0021] Some embodiments of the second light sheet microscope further comprise a movable stage that is controllable by the controller. In these embodiments, the projector has an input port and is configured to change characteristics of the projected image based on data that arrives at the input port, and the light sheet microscope further comprises a controller configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector. In these embodiments, the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to outputa thick first line of light; and the controller is also further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light; and the controller is also further configured to obtain multiple low resolution images of respective thick slices of the sample by sending the first data to the input port of the projector, and instructing the at least one first camera to capture respective images of the sample while the movable stage is set to each of a plurality of different respective locations in a Z direction.

[0022] Optionally, in the embodiments described in the previous paragraph, each pixel in the low resolution images includes at least 4 native-resolution pixels of the at least one first camera that have been binned together. Optionally, in these embodiments, the controller is further configured to identify a region of interest from the low resolution images, and to subsequently obtain high resolution images of a plurality of thinner slices of the sample that correspond to the region of interest.

[0023] In some embodiments of the second light sheet microscope, the first set of optical components comprises an electrically tunable lens positioned between the scan lens and the sample. Optionally, in these embodiments, the scan lens comprises a Pldssl lens.

[0024] In some embodiments of the second light sheet microscope, the first set of optical components comprises an electrically tunable lens positioned between the scan lens and the sample. These embodiments further comprise a controller configured to (a) determine, based on feedback obtained using the at least one first camera, at least one axial position parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust the electrically tunable lens based on the at least one axial position parameter.

[0025] In some embodiments of the second light sheet microscope, the first set of optical components comprises an electrically tunable lens positioned between the scan lens and the sample. These embodiments further comprise a controller configured to (a) determine, based on feedback obtained using the at least one first camera, at least one light sheet offset parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust a projected line offset based on the at least one light sheet offset parameter.

[0026] In some embodiments of the second light sheet microscope, the first set of optical components comprises a tube lens and an illumination objective lens arranged so that the flattened first sheet of illumination light that is output by the scan lens will (a) enter the tube lens, (b) exit the tube lens and enter a back aperture of the illumination objective lens, and (c) exit the illumination objective lens and enter the sample.

[0027] Optionally, the embodiments described in the previous paragraph can further comprise an electrically tunable lens positioned between the tube lens and the illumination objective lens. Optionally, in these embodiments, the scan lens can comprise a Pldssl lens.

[0028] In some embodiments of the second light sheet microscope, the projector comprises a MEMS laser projector.

[0029] Some embodiments of the second light sheet microscope further comprise the at least one first camera.

[0030] In some embodiments of the second light sheet microscope, the first objective lens is mounted so that its optical axis is perpendicular to the flattened first sheet of illumination light that travels through the sample.

[0031] In some embodiments of the second light sheet microscope, the image projected by the projector includes a second line of light, so that a projection of the second line of light forms a second sheet of light that is projected out of the projector; the scan lens is positioned to accept the second sheet of light that is projected out of the projector, and to output a flattened second sheet of illumination light; the first set of optical components is positioned to route the flattened second sheet of illumination light into the sample, so that the flattened second sheet of illumination light travels through the sample; and the light sheet microscope further comprises a second objective lens positioned to accept light from a depth within the sample that corresponds to the flattened second sheet of illumination light that travels through the sample, and route that light through a back aperture of the second objective lens and towards at least one second camera. Optionally, in these embodiments, the second line of light is parallel to the first line of light, and the flattened second sheet of illumination light is parallel to the flattened first sheet of illumination light.

[0032] In some embodiments of the second light sheet microscope, the flattened first sheet of illumination light is oriented parallel to an upper surface of the sample, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample.

[0033] In some embodiments of the second light sheet microscope, the flattened first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°. Optionally, in these embodiments, an optical axis of the first objective lens is perpendicular to the flattened first sheet of illumination light.

[0034] In some embodiments of the second light sheet microscope, the flattened first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample or the lower surface of the sample.BRIEF DESCRIPTION OF THE DRAWINGS

[0035] FIG. la is a perspective view of an embodiment of a projected Light Sheet Microscopy (pLSM) microscope.

[0036] FIG. lb depicts the optical paths in the detection arm and one of the illumination arms of the FIG. la pLSM microscope.

[0037] FIG. 1c depicts the data flow paths for controlling the FIG. la pLSM microscope.

[0038] FIGS. 2a and 2b depict the system Point Spread Function (PSF) across the field of view (FOV) for the FIG. la embodiment for 16.7x and lOx detection objectives, respectively.

[0039] FIG. 2c depicts simulated intensity profiles at the waist of light sheets generated by the projection of multiple pixel- width lines.

[0040] FIG. 2d depicts linearly adaptive light sheet offset calibration for correcting the light sheet-focal plane alignments for achieving high-quality imaging throughout large cleared samples using the FIG. la embodiment.

[0041] FIG. 2e depicts a Maximum Intensity Projection (MIP) images of TH+ neurons obtained from a sample using pLSM (with a 2-pixel wide light sheet) and high-end LSFM implementation.

[0042] FIG. 2f depicts three-channel imaging of CLARITY cleared mouse brain.

[0043] FIG. 3 shows a comparative analysis of region- specific neuron counts extracted from pLSM and COLM images.

[0044] FIG. 4 shows a quantification of total number of detected cells and Sox2 signal intensity distribution.

[0045] FIGS. 5A-E depicts an example setup for aligning the illumination arm, showing the alignment target in position to assess the alignment accuracy of the components.

[0046] FIG. 6A-B depict aligning the illumination and detection in the system.

[0047] FIG. 7 depicts a microscope control graphical user interface (GUI).

[0048] FIGS. 8a and 8b depict two different pLSM hardware configurations.

[0049] FIG. 8c depicts automated multi-resolution imaging of a sample in large volume using the FIG. 8a or the FIG. 8b embodiments.

[0050] FIG. 9 is detail of a pLSM configuration that relies on the parallel processing of images obtained using two light sheets to increase the imaging speed.

[0051] FIG. 10 depicts an embodiment in which a sheet of illumination light enters the sample through the upper surface of the sample at an oblique angle.

[0052] FIG. 11 depicts another embodiment that is the inverted version of the FIG. 10 embodiment.

[0053] FIG. 12 depicts another embodiment in which two sheets of illumination light enter the sample through the upper surface of the sample at an oblique angle.

[0054] FIG. 13 depicts another embodiment that is the inverted version of the FIG. 12 embodiment.

[0055] FIG. 14 depicts a variation of the FIG. 10 embodiment in which the scan lens and the tube lens in the illumination arm are omitted.

[0056] Various embodiments are described in detail below with reference to the accompanying drawings, wherein like reference numerals represent like elements.DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0057] SECTION 1

[0058] This application describes a versatile and scalable LSFM framework, projected Light Sheet Microscopy (pLSM), that achieves high imaging performance while minimizing the spatial, optical, electronics and computational footprint, and lowering the implementation costs by an order of magnitude. This can be achieved by repurposing readily available consumer-grade components, optimized optics, an over-network control architecture, and software-driven light sheet modulation.

[0059] We performed extensive characterization of the pLSM framework, along with a comprehensive array of applications in high-resolution mapping of cleared mouse brains and post-mortem human brain pathological samples, and molecular phenotyping of human induced pluripotent stem cell (iPSC)-derived brain and vessel organoids. Additionally, we report a method for live imaging of sparsely (2.5%) labeled bacterial pellicle biofilms directly at an air-liquid interface, revealing the intricate cellular dynamics in a multi-layered architecture. Overall, the pLSM framework has the potential to make high resolution LSFM widely accessible, affordable, and scalable for high-throughput biomedical applications.

[0060] This application describes a versatile LSFM imaging framework called projected LSFM (pLSM), which offers a scalable solution without compromising the imaging performance. Our approach leverages off-the-shelf components, including pocket LASER projectors as multi-color illumination sources, Nvidia Jetson Nano boards for electronic control, 3D-printed imaging chambers, and optimized scan and detection optics. Additionally, we have developed a network-based imaging workflow using native Python and Jupyter Notebook, enabling remote control of multiple microscopes at scale. Altogether, pLSM provides high-resolution imaging, akin to that of conventional light sheet systems, with significantly reduced mechanical, optical, electronics and computational complexity and size.

[0061] In our extensive characterization of the pLSM system, we analyze the achievable lateral and axial resolutions and demonstrate its robustness and versatility by performing high-resolution multi-color imaging of large mouse and Alzheimer’ s disease compared with neurotypical human brain samples cleared and labelled using multiple methods. Moreover, we establish scalable pLSM mounting and imaging capabilities for human iPSC-derived brain and vessel organoids at sub-cellular resolution, addressing their sample-to-sample variability. Furthermore, we developed and utilized a pLSM live imaging assay for sampling bacterial dynamics directly at the air-liquid interface in sparsely (2.5%) labeled pellicle biofilms, yielding insights into their dynamic layered architecture with significant biomedical implications. Overall, we demonstrated high imaging performance of pLSM in diverse experimental contexts, from brain structural imaging to live bacterial biofilms, while significantly reducing the implementation complexity and costs by an order of magnitude.

[0062] Results

[0063] pLSM implementation

[0064] FIGS, la and lb are, respectively, a 3D model and an optical path diagram of an embodiment of a pLSM system. This embodiment repurposes inexpensive, commercially available “pocket” LASER projectors 10 for generation of thin scanned light sheets of up to three wavelengths. Table 1 is a list of components that can be used to make the FIG. la embodiment.

[0065] Table 1 1 Detailed system components list for pLSM.Part Item Usage CountIllumi- AnyBeam / Ultimems Laser source 2 nation ProjectorAC254-045-A Plossl scan lens 4AC508-075-A Tube lens 2Mitutoyo lOx Plan Objective 2ApoDT12A Angle Bracket for Projector Mounting 4DT12 Translation Stages for Projector Mounting 4LCP01B 60mm Mounting Brackets 6LCP09 60mm Cage Plates 4ER6-P4 Cage Assembly Rods, 6" Long, 06 mm, 4 Pack 2TR1-P5 Optical Post, 5 Pack 2PH1E Post Holder 4SM1L10 Scan Lens Mounting Tube 2SM2A6 Scan Lens and Objective Mounting Adapter 4SM2L10 Optics Mounting 6SM2V05 Objective Adjustable Lens Tube 2KCB2C Mirror Mount (Optional for compact assembly) 2PF20-3-G01 Aluminum Mirror (Optional for compact 2 assembly)Detection TTL200 Tube lens 1GS3-U3-89S6M-C Camera 1Mitutoyo lOx Plan Objective 1ApoASI 16.6X / 0.4NA Objective (Optional) 1LCP01B 60mm Mounting Brackets 2LCP09 60mm Cage Plates 2ER6-P4 Cage Assembly Rods, 6" Long, 06 mm, 4 Pack 1PH1E Post Holder 2SM2A6 Objective Mounting Adapter 1SM2L10 Optics Mounting 1SM2V05 Objective Adjustable Lens Tube 1SM2A20 Tube Lens Mounting Adapter 1SM2L20 Tube Connecting Tube Lens and Camera 1SM2L30 Tube Connecting Tube Lens and Camera 1SM2A31 Camera Mounting Adapter 1LCP90F Emission Filter Mounting Plate 1Sample KDC 101 Translation Motor Stage Control 3Mounting PT3 / g Translation Motor Stage for Imaging 1AB90E Angle Bracket for Mounting 1DT12 Translation Stages for Manual Sample 2MountingDT12A Angle Bracket for DTI 2 2QRP02 Rotational Stage for Manual Sample Mounting 1Others Nvidia Jetson Nano System Control Circuit Board 1Micro SD Memory Card for Control System 1

[0066] In the embodiment illustrated in FIGS, la-b, two illumination arms (illumination arm 1 and illumination arm 2) generate opposing light sheets that are employed to illuminate the sample, one from the left and one from the right in FIG. lb. (The components within the illumination arm 2 are identical to the components within illuminationarm 1, but with a mirror-image orientation.) These opposing light sheets should be aligned. Note that while FIGS, la-b depict two illumination arms 1 and 2, in alternative embodiments only a single illumination arm is used (in which case the second illumination arm should be omitted).

[0067] The sample is positioned within an oil chamber, and a 3D motorized stage (not shown) is used for multi-tile imaging of large samples mounted in quartz glass cuvettes within a 3D-printed immersion chamber.

[0068] The illuminated plane is captured by an orthogonally arranged in-focus detection objective 50, a tube lens 55, and a camera 60 (e.g., a consumer-grade CMOS camera). Optionally, a multi-band filter 52 is positioned between the detection objective 50 and the tube lens 55 or, in alternative embodiments, between the tube lens 55 and the camera 60. For the detection arm, we utilized a long-working distance air objective 50 (Mitutoyo lOx Plan Apo) or multi-immersion objective (ASI 16.67X / 0.4NA), a 200 mm focal length tube lens 55 (TTL-200, Thorlabs), a multi-band pass emission filter 52, and a low-cost consumergrade CMOS camera 60 (GS3-U3-89S6M-C, FLIR, 3.45 pm pixel) for wide-field detection of the illuminated plane. Notably, this camera provided performance comparable to high-end sCMOS cameras, albeit with lower quantum efficiency.

[0069] We sought to develop cost-effective alternatives to expensive multi-laser engines, galvo scanners, and associated controllers for generation of thin illumination light sheets. To achieve this, we used commercially available “pocket” laser projectors 10 for generation of high-quality scanned light sheets with 3-color imaging capabilities. These projectors (e.g., AnyBeam / Ultimems Projectors) provide substantial benefits of fully integrated 3-color laser diodes (TO-8 set with blue: 440-460 nm, green: 515-530 nm, and red: 632-642 nm) with sufficient power needed for imaging, MEMS 2D- scanners, and associated electronics in a compact assembly.

[0070] The projector 10 is configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector. A scan lens 20 (e.g., a Plbssl scan lens) is positioned to accept the first sheet of light that is projected out of the projector 10, and outputs a flattened first sheet of illumination light. A first set of optical components 32-40 is positioned to route the flattened first sheet of illumination light into the sample, so that the flattened first sheet of illuminationlight travels through the sample. In the embodiment depicted in FIGS, la-b, the first set of optical components 32 includes a tube lens 32, a folding mirror, and the illumination objective 40. These components are arranged so that the flattened first sheet of illumination light that is output by the scan lens will (a) enter the tube lens, (b) exit the tube lens and enter a back aperture of the illumination objective lens, and (c) exit the illumination objective lens and enter the sample. But in alternative embodiments, a different set of components can serve as the first set of optical components.

[0071] A first objective lens 50 (i.e., the detection objective) is positioned to accept light from a depth within the sample that corresponds to the flattened first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens 50, and towards the camera 60.

[0072] Adapting pocket projectors 10 for light-sheet generation presented practical challenges in controlling the various light-sheet properties, positioning, and scanning speeds. To address these challenges, we devised a control framework utilizing the Python X library (Python-Xlib) to directly modulate various scanning parameters, using the controller depicted in FIG. 1c. The control code was developed using Python and is executed on an Nvidia Jetson Nano board. Remote operation of the system is enabled through a Jupyter Notebookbased graphical user interface (GUI) utilizing the SSH remote protocol. The FLIR CMOS camera is controlled by the Spinnaker SDK package, while the LASER projector via the Xlib library generates the scanned light sheets. The acquired camera data is streamed to a solid- state hard drive through a high-speed USB 3.0 port.

[0073] This framework allows the generation of light sheets as multi-pixel (1 or more, on-demand) wide projected lines of specific height and location (FIG. 2a-c). Importantly, this approach also offers the advantage of easily modulating the light sheet’s thickness and field-of-view (FOV) by controlling the pixel widths of the projected line that is generated by the pocket projector 10. More specifically, the controller depicted in FIG. 1c commands the projector 10 to display a line by sending appropriate commands to an input port (e.g., an HDMI port) of the projector 10. When the line that is being displayed is projected out of the projector 10, the result is a sheet of light. The controller can vary the thickness of the sheet of light by sending appropriate commands to an input port of the projector 10 to vary the thickness of the line that is being displayed by the projector. For example, if the controller commands the projector 10 to display a line that is 1 pixel wide, the result will be a thin sheetof light. And if the controller commands the projector 10 to display a line that is 2 or more pixels wide, the result will be correspondingly thicker sheets of light. Downstream to the projector 10, we employed a Plossl lens 20 (implemented as a pair of achromats, AC254-045- A) as an effective substitute for expensive f-theta scan lenses, an achromat doublet (AC508- 075-A) as a tube lens 32, and an air illumination objective 40 (Mitutoyo lOx Plan Apo).

[0074] The pLSM control framework and the imaging workflows were implemented on a cost-effective Nvidia Jetson Nano board running the Linux operating system. The Spinnaker SDK package for image grabbing and Python-Xlib for light sheet projection. Notably, the pLSM control software control system is stored on an SD card inserted in the Nano board, thus allowing straightforward duplication of the entire system as a cloned disk image of the operating system. The acquired images are streamed to a USB 3.0 high-speed solid-state hard drive. The control GUI was implemented using IPyWidgets within the Jupyter Notebook framework, enabling remote operation via a SSH connection from any machine. This network-controlled framework allows for easy scalability without requiring the full capacity of a remote machine.

[0075] pLSM characterization

[0076] FIGS 2a-f depict various characterizations of the FIGS, la-b embodiment. More specifically, FIGS. 2a and 2b depict the system Point Spread Function (PSF) across the field of view (FOV) for the FIG. la embodiment for 16.7x (FIG. 2a) and lOx (FIG. 2b) detection objectives, respectively. FIG. 2c depicts simulated intensity profiles of at the waist of light sheets generated by projection of multiple pixel-width lines. FIG. 2d depicts linearly adaptive light sheet offset calibration for correcting the light sheet-focal plane alignments for achieving high-quality imaging throughout large cleared samples using the FIGS, la-b embodiment. FIG. 2e depicts a Maximum Intensity Projection (MIP) images of TH+ neurons obtained from a sample using pLSM (with a 2-pixel wide light sheet) and high-end LSFM implementation. And FIG. 2f depicts three-channel imaging of CLARITY cleared mouse brain with Thyl -eGFP, GFAP (Alexa 555) and Podocalyxin / CD31 (Alexa 647). Scale bar: 100 pm.

[0077] For FIGS. 2a and 2b, the light sheet thickness was varied by projection of multi-pixel illumination lines. Scale bar: 50pm. A single-pixel projected line achieves a thickness of approximately 5 pm at the thinnest waist. Use of multiple pixel projected linesyielded thicker light sheets with more uniformity across the FOV. Note that, consistent with previous reports, the achieved axial FWHM uniformity (for 1 projection line, right plots) was better than theoretical expectation due to the use of Air objective for illumination through a chamber filled with oil, thus causing PSF elongation due to smooth spherical aberrations.

[0078] FIG. 2c depicts simulated intensity profiles of at the waist of light sheets generated by projection of multiple pixel-width lines. Notably, increasing the number of projected lines improves the uniformity of the intensity throughout the light-sheet crosssection.

[0079] FIG. 2d depicts linearly adaptive light sheet offset calibration for correcting the light sheet-focal plane alignments for achieving high-quality imaging throughout large cleared samples. The large sample size can introduce refractive index inhomogeneity which leads to changes in the detection focal plane and light sheet trajectory. The lower left panel illustrates choosing of a few 3D ‘anchor points’ with manually defined optimal light sheet offsets, and spatial interpolation to estimate optimal light sheet offsets across the entire 3D sample. Right panels demonstrate the effects of calibration in correcting the light sheet-focal plane alignment to yield focused images. Scale bar: 100 pm.

[0080] FIG. 2e depicts a Maximum Intensity Projection (MIP) images of TH+ neurons obtained from the same sample using pLSM (with a 2-pixel wide light sheet) and high-end LSFM implementation COLM. Scale bar: 100pm. The right column shows quantitative analysis and comparison of the intensity profiles of a cell imaged by pLSM and COLM. The pLSM image was preprocessed by applying median filtering (with four neighboring pixels) followed by Gaussian filtering (G= I ) to eliminate hot pixels.

[0081] We first characterized the lateral and axial resolution of pLSM by imaging 500 nm fluorescent beads using two different detection objectives and illumination light sheets of varying thickness generated by the projection of multi-pixel wide lines. As shown in FIG.2a-b, the pLSM achieved -0.9 pm and -1.7 pm lateral resolutions for the two objectives, and the best axial resolution of -5 pm at the light sheet waist, comparable to or better than other implementations (Table 2). Furthermore, for multi-pixel projected lines, we observed sublinear increases in the thickness of the light sheet (FIG. 2b, axial FWHM plots) with enhanced uniformity. Importantly, this modulation of light sheet thickness and field-of-view uniformity can be accomplished through pLSM software by controlling the pixel width of thelines projected with the projector, providing a convenient means for rapid adaptation to diverse experimental requirements.

[0082] Additionally, our findings also indicate that this approach of projecting multiple lines to generate thicker light sheets yields a flatter intensity profile across the crosssection, compared with an expected Gaussian distribution (FIG. 2c). Notably, consistent with previous studies, the use of an Air objective for illumination through an oil chamber results in spherically-aberrated beams with more uniform FOV compared with the theoretical expectation (FIG. 2a-b, axial FWHM plots), although less uniform than that can be achieved by axial scanning approaches. The scanning speed of the light sheet in pLSM, and consequently the imaging speed, is mainly determined by the projector refresh rate (60 Hz for the utilized model), independent of the pixel widths employed. Consequently, pLSM presents a highly adaptable multi- scale scheme capable of accommodating various imaging needs, spanning from high axial resolution settings to high-volume acquisition rates with thicker light sheets, which allow use of larger z-step settings to reduce the total number of optical sections to cover a volume.

[0083] To enable high quality imaging of large intact samples, we optimized the imaging workflow of pLSM to address the sample-induced optical aberrations at different locations. A key improvement was the implementation of a linearly adapting light- sheet offset correction procedure (FIG. 2d). By sampling optimal parameters at a few (typically 5- 10) anchor 3D positions and applying 3D linear interpolation, the pLSM system generates a smoothed light-sheet offsets map to compensate for optical shifts resulting from refractive index mismatches. Importantly, this approach also allows compatibility of pLSM with various tissue clearing methods.

[0084] Next, we conducted a direct comparison of pLSM imaging quality with a tissue-clearing optimized light sheet microscopy implementation (CLARITY-Optimized Light Sheet Microscopy, COLM) as a representative conventional light sheet system. Note that the COLM system has been extensively used and cited for high-resolution imaging of varied types of large cleared samples, including whole mouse brains, spinal cord samples and cancerous tissues, among others. Quantification of the intensity distribution, as shown in FIG. 2e, demonstrated that pLSM achieves comparable imaging quality. Furthermore, we demonstrated the multi-color imaging capability of pLSM by performing 3-channel imaging of a mouse brain sample containing three markers Thyl-eGFP, GFAP (Alexa Fluor 555) andPodocalyxin / CD31 (Alexa Fluor 647) (FIG. 2f) using blue (440-460 nm), green (515-530 nm) and red (632-642 nm) light.

[0085] In summary, the detailed characterization demonstrated that pLSM provides high-quality imaging, while offering a highly compact, cost-effective, and scalable platform for rapid high-resolution imaging of biological samples.

[0086] High-resolution mapping of large intact samples cleared with multiple techniques

[0087] We conducted high-resolution imaging of intact mouse brain samples cleared using the iDISCO and CLARITY techniques. In one experiment, our focus was on mapping the dopamine (DA) system throughout the entire mouse brain using pLSM imaging. We stained intact mouse brain samples with an antibody against tyrosine hydroxylase (TH), a widely-used marker of DA neurons, and an Alexa Fluor 647 secondary antibody. The sample was cleared using the iDISCO method and imaged with red projected lines (632-642 nm) and a Mitutoyo lOx (Plan Apo) objective. The results demonstrated successful high-quality imaging of the intact mouse brain sample, enabling the complete reconstruction of the DA system. Additionally, we utilized pLSM (with the same specifications) to map the entire micro-vasculature in an intact mouse brain. Vasculature labeling was achieved using a cocktail of antibodies against Podocalyxin, CD31, and Acta2 (all detected using the same Alexa Fluor 647 secondary antibody).

[0088] We further investigated whether the whole brain datasets acquired using pLSM could provide quantitative insights comparable to those obtained from optimized LSFM implementations such as COLM. To explore this, we performed high-resolution imaging of the same TH-labeled mouse brain sample using both pLSM and the COLM system. The total imaging times with the two systems were similar (COLM, 10X Obj: -6 hours, pLSM, lOx Obj: -5 hours).

[0089] High-resolution pLSM imaging was performed, and quantitative analysis of intact mouse brain samples cleared with iDISCO was performed. High-resolution imaging of TH+ dopamine neurons in an iDISCO+ cleared brain sample were obtained. Some images showed an overview of the entire brain, while other images showed visualizations of specific neuronal features at various depths. High-resolution imaging of the complete mouse brain vasculature in an iDISCO cleared sample was performed. Some images showed a renderingof the entire brain, while other images showed a 3D rendering of a smaller volume. Additionally, 100pm thickness Maximum Intensity Projections (MIPs) were depicted at different imaging depths. Segmentation of TH+ dopamine neurons in datasets acquired by pLSM and COLM imaging of the same sample was performed. The point clouds were color- coded according to the regional annotations from the Allen Brain Atlas at level 5.

[0090] FIG. 3 shows a comparative analysis of region- specific neuron counts extracted from pLSM and COLM images. The top 40 regions at ABA level 8 were included in the analysis. The Pearson correlation coefficient of 0.966 indicates a high degree of agreement between the two imaging methods. We utilized the suite WB pipeline to map TH+ DA neurons. The point clouds generated from the two systems showed strong agreement across all brain regions, demonstrating the compatibility of pLSM-acquired data with whole brain analysis approaches.

[0091] To assess the compatibility of pLSM with different tissue clearing methods, we also performed whole -brain imaging of CLARITY-cleared mouse brain samples from the Thyl-eYFP transgenic mouse line. The samples were cleared using the passive CLARITY method and imaged with green-color (515-530 nm) projected lines and Mitutoyo lOx and ASI 16.67X / 0.4 NA objectives. The pLSM system achieved high-resolution imaging of CLARITY cleared intact mouse Thyl-eYFP brain. Some of the obtained images showed a rendering of the whole brain images, and other images showed a 3D rendering of the deeper half of the brain. Some of these images were obtained using an ASI 16.67X / 0.4 NA objective.

[0092] Next, we demonstrated the applicability of pLSM for high-resolution mapping of large post-mortem human brain samples. Adopting the SHANEL clearing method, we labeled human brain samples with vasculature markers (Podocalyxin, CD31, and Acta2) and performed rapid high-resolution 2-color imaging with pLSM to enable detailed reconstructions. Importantly, pLSM acquired images were amenable to standard vasculature tracing and quantification approaches.

[0093] Lastly, we demonstrated the applicability of pLSM in mapping the pathological distribution of amyloid precursor protein (APP) aggregates in large volumes of Alzheimer’s disease brain tissue, compared with a neurotypical source. The brain samples were stained with an antibody against APP and processed with the SHANEL method. ThepLSM system allowed high-resolution visualization of the APP aggregates throughout large post-mortem human brain samples from Alzheimer’s disease and neurotypical samples. More specifically, high-resolution 2-channel imaging of a large human brain sample (6.57 x 3.24 x 4.84 mm3) stained with antibodies against Podocalyxin / CD31 and Acta2. Some images showed a 0.5 x 0.5 x 0.5 mm3volume of raw data and vasculature tracings at three different depths within the sample, and colorbar maps were used to show extracted vessel diameters. Other images showed high-resolution imaging of amyloid precursor protein (APP) distribution and its aggregation in large volumes of Alzheimer’s disease and neurotypical brain samples.

[0094] Validation of APP staining was performed. Confocal imaging of Alzheimer's Disease and neurotypical brain tissue sections were stained with a- APP (Y188) antibody and an empty channel for autofluorescence. We obtained images of Maximum Intensity Projection (MIP) of 100pm stack, and zoom-in images MIP of 20pm stack. The observed APP signal and aggregates were distinct from the autofluorescence, confirming their nonlipofuscin origin. We also obtained images showing secondary antibody alone control for Alzheimer's Disease tissue section (MIP of 20pm stack). These control experiments validated the specificity of APP signals observed in the former images.

[0095] In conclusion, these extensive imaging experiments with large and diverse tissue types and clearing methods substantiate the broad applicability of the pLSM imaging framework for high-resolution structural imaging. The incorporation of the linearly adaptive light-sheet offset correction procedure effectively compensates for the misalignments of the detection focal plane with the illumination light sheet, ensuring high imaging quality throughout the sample. Importantly, despite its highly cost-effective nature, pLSM generates quantitative insights comparable to those obtained from highly optimized LSFM systems.

[0096] Phenotyping of multiple human iPSC-derived brain and vessel organoids

[0097] Recent advances in brain and vessel organoids are enabling unprecedented access to some of the molecular, cellular and developmental mechanisms underlying complex diseases. However, these in vitro 3D preparations often exhibit considerable sample-to- sample variability, necessitating the use of large sample sizes for robust comparisons across different experimental conditions. We demonstrated the effectiveness of the scalable pLSM framework for imaging of numerous organoid samples at sub-cellular resolution.

[0098] We first optimized the sample mounting for rapid imaging of several organoids in a single imaging session. This was achieved manually by immobilizing labeled organoid samples in a line by embedding them in 1% agarose gel, which can then be processed with various tissue clearing methods, including the FDISCO method used in these experiments. The cleared organoids ensemble was mounted in a cuvette in a line for unhindered optical access from all sides and imaged with pLSM at high resolution. This approach allowed us to perform mapping of Sox2 (a marker of progenitor cells) expression in several (8 demonstrated) brain organoid samples in a single imaging experiment. pLSM imaging produced high-resolution images of the entire preparation, facilitating systematic quantification. The total imaging time was ~5.5h (pLSM / 16x).

[0099] We performed pLSM imaging of an ensemble of brain organoid samples, immunostained with anti-Sox2 antibody and cleared with FDISCO technique. Volumetric renderings showed the raw data, segmentation, and Sox2 expression point clouds. We obtained a zoomed- in view of brain organoid samples, showing raw data and the reconstructed expression point cloud. We also obtained multi-color pLSM imaging of vessel organoids stained with anti-CD31 antibody and ZOl-mEGFP reporter expression. The peripheral sprouting vessels exhibited lower ZOl-mEGFP signal.

[0100] FIG. 4 depicts quantification of total number of detected cells and Sox2 signal intensity distribution, showing the variability across brain organoid samples generated in the same batch. Comparative analysis across samples revealed significant heterogeneity in shape, cell number, and Sox2 expression profiles, underscoring the importance of such approaches for generating statistically significant insights. Furthermore, we applied the same manual sample mounting and one-step imaging strategy to perform two-color high-resolution imaging of human iPSC-derived vessel organoids labeled with CD31 antibody (an epithelial cell marker) and ZOl-mEGFP expression (a tight junction protein reporter). This approach unveiled sub-cellular resolution morphological details across multiple vessel organoid samples. Additionally, a comparison of CD31 staining and ZOl-mEGFP expression pattern revealed lower levels of ZOl-mEGFP signal in the peripheral sprouting vessels, consistent with their lack of tight gap junctions.

[0101] Overall, the results from these experiments, combined with rigorous quantification, demonstrated the scalability and effectiveness of the compact pLSM imaging framework for enabling quantitative phenotyping of different types of organoid samples athigh resolution. Through its ability to capture a comprehensive view of the molecular and cellular heterogeneity within these complex 3D structures, pLSM has the potential to facilitate crucial insights into the intricate mechanisms that underlie various diseases.

[0102] Sampling the cellular dynamics in sparsely labeled bacterial pellicle biofilms at air-liquid interface

[0103] To gain insights into the organization of bacterial biofilms and their underlying cellular dynamics, we utilized the light sheet thickness modulation and live imaging capabilities of pLSM. Biofilms, representing the most prevalent form of multicellular organization in the bacterial world, hold significance for various applications such as biotechnology (e.g., waste degradation, chemical production), clinical relevance, and biogeochemical cycling in water and soil ecosystems. The intricate interactions, including social and electrical connections among bacterial cells within biofilms, result in emergent properties beyond individual components. Therefore, understanding the complex cellular dynamics of biofilm formation, maintenance, and function is crucial.

[0104] Recent live imaging studies on bacterial biofilms have provided crucial insights into their dynamic nature, including remarkable transformations from 2D branched morphology to densely packed 3D clusters, as well as the intricate cellular dynamics underlying biofilm formation. However, understanding the cellular dynamics across the deep layered architecture of pellicle biofilms which form at air-liquid interface remains limited. To address this gap, we first developed and optimized a gentle live imaging assay for sampling of cellular dynamics in sparsely labeled (2.5% fluorescently labeled by mScarlet expression) pellicle biofilms formed by Pseudomonas aeruginosa at air-liquid interface. This involved cultivating pellicles in quartz glass cuvettes for unhindered optical access to image crosssections through thickness, providing good resolution across depths and a large field-of-view. To minimize photo toxicity, we aimed to perform imaging of a single illuminated plane (instead of 3D stacks) by a thick light sheet. The optical thickness of the illuminated plane was modulated (by the use of multiple pixels wide projected lines, FIG. 2a-c) and optimized by assessing the resolv ability of a large fraction of visible bacteria in the field-of-view and their significant continuous trackability (before moving out of the plane), resulting in the final use of 4-pixels wide projected lines and 16x detection objective, yielding -12 microns axial FWHM of the image plane and -0.9 microns lateral resolution (as shown in FIG. 2a).

[0105] Next, we carefully considered the toxicity effects of high imaging speeds on pellicle biofilms, settling on a 2 Hz sampling rate that still allowed capturing of meaningful cellular dynamics over extended durations while minimizing the energy load. In addition, we also performed very long-term recordings, lasting up to 7 hours at 12-second intervals. The resulting large datasets were analyzed using the Ilastik framework for pixel classifications into three categories: non-resolvable tight clusters at the air-liquid interface, individual cells or small clusters, and background regions. Our findings from live imaging of sparsely labeled biofilms revealed the presence of three morphologically and dynamically distinct regions within Pseudomonas aeruginosa pellicles.

[0106] We obtained images mapping the dynamic layered architecture of sparsely labeled (2.5%) bacterial pellicle biofilms at the air-liquid interface. Some of these images showed Pseudomonas aeruginosa pellicle biofilms that were generated in cuvettes for live imaging of a thick optical cross-section through the depth, including images showing representative raw data and detection of clusters and cells. Color-coded projections across time were added to visualize the dynamic layered architecture of biofilm pellicles. Three distinct dynamic layers were apparent, showing disorganization to organized dynamics. The color-coded projections were performed across time, at three representative time points, showing examples of cellular dynamics.

[0107] Layer 1 comprised tightly clustered cells (non-resolvable in this experiment) located at the air- liquid interface, exhibiting highly dynamic shape changes. Layer 2 exhibited a heterogeneous porous organization characterized by chain-like arrangements of bacterial cells, allowing rapid cellular movements in specific regions while impeding motility in others. Layer 3 encompassed freely floating bacteria within the liquid, displaying varying degrees of motility, ranging from near immobility to high-speed projectile-like motions. Notably, we observed significant cell exchange between layers, particularly between layers 2 and 3, indicating intricate cellular interactions within the biofilm as a single system.

[0108] Overall, this combination of sparse (2.5%) fluorescent labeling and the use of modulated thick light sheets in pLSM for planar illumination and detection from the side allowed sampling of cellular dynamics through deep-layered architecture of pellicle biofilms directly in their native state at the air-liquid interface. It's crucial to note that sparse labeling was crucial for these experiments; resolving cells in fully labeled will require much higher detection NA and thinner light sheets. These findings contribute towards a betterunderstanding of the biofilm biology, with significant implications in clinical, geo-biological, and biotechnology contexts.

[0109] Tested Uses of the FIGS, la-b embodiment.

[0110] The FIGS, la-b embodiment was used to capture a set of videos in a diverse range of contexts, ten of which are described below. (1) Complete brain-wide visualization of TH+ neurons mapped with projected Light Sheet Microscopy (pLSM). This video showed volumetric rendering of TH+ neurons throughout an iDISCO+ cleared sample, showcasing the extensive brain-wide mapping capabilities of pLSM. (2) Whole-brain segmentation of TH+ neurons in the same sample imaged with pLSM as well as CLARITY -optimized light sheet microscopy (COLM) system. This example highlighted the high degree of agreement in the automated segmentation of whole-brain data acquired using pLSM and COLM. (3) Whole-brain rendering of CLARITY -cleared Thyl-eYFP transgenic mouse brain imaged using pLSM. This example showcased high-resolution imaging of CLARITY -cleared Thyl- eYFP transgenic mouse brain images using pLSM. (4) Small volume rendering of CLARITY -cleared Thyl-eYFP transgenic mouse brain imaged using pLSM. This representative small volume rendering showcased the high-resolution imaging capabilities of pLSM. (5) Multi-channel pLSM imaging of vasculature in neurotypical post-mortem human brain sample. The sample was labelled using antibodies against Podocalyxin / CD31 (cyan) and Acta2 (yellow), markers of endothelial and smooth muscle cells respectively. (6) High- resolution rendering of vasculature image volumes at different depths of an intact human brain sample. Small image volumes extracted from different depths of the sample were visualized. The automated segmentation and quantitative tracing of the vasculature diameters and connectivity were shown. (7) High-resolution imaging of Amyloid Precursor Protein (APP) in Alzheimer’s disease and neurotypical cortical samples. Volumetric rendering of APP immunostaining signals in Alzheimer’s disease and neurotypical post-mortem brain samples. (8) Volumetric rendering of the pLSM imaging of an ensemble of human iPSC- derived brain organoids. This video showed mounting and one-step pLSM imaging of an ensemble of 8 brain organoid samples labeled with anti-Sox2 antibody. (9) Volumetric rendering of human iPSC-derived vessel organoids imaged with pLSM. This video showed pLSM imaging of vessel organoids stained with anti-CD31 antibody. (10) Live imaging of sparsely labeled (2.5%) bacterial biofilm pellicle at the air-liquid interface. Cellular dynamics across the Pseudomonas aeruginosa pellicle biofilm were visible at the air-liquid interface.

[0111] Discussion

[0112] We have introduced an innovative LSFM imaging framework, called pLSM, that provides a highly compact and cost-effective solution for achieving high-resolution imaging, both for living samples as well as large cleared biological samples. By leveraging off-the-shelf commercial-grade components strategically, we have minimized the mechanical, optical, electronics, and computational footprint, leading to a substantially lower implementation cost. This positions pLSM as a scalable and compact alternative to conventional LSFM systems. We demonstrated the robustness and versatility of pLSM through multi-channel imaging and quantitative analyses of various biological samples, including intact mouse brains, pathological human brain tissue, vessel and brain organoids, and bacterial biofilms.

[0113] The pLSM framework also incorporates several advanced features. Firstly, the software-driven modulation of the light sheet thickness allows for easy adjustment of axial resolution, accommodating diverse experimental needs, such as iterative multi-scale imaging of very large samples. This is achieved through pixel binning for lateral and multi-pixel projected lines for axial resolution modulation. Notably, this feature has proven effective in sampling the cellular dynamics within sparsely labeled bacterial pellicle biofilms formed at the air-liquid interface, providing crucial insights into their layered architecture. Secondly, the implementation of a generic over-the-network control enables remote operation and, together with compact hardware footprint, facilitates easy adoption in new locations. Thirdly, the linearly adaptive light sheet offset correction approach, by utilizing only a few 3D predefined points, optimizes the sample imaging time and significantly advances the offset calibration strategy previously introduced in the COLM system. Importantly, this approach also ensures compatibility with cleared samples of different optical properties, such as iDISCO, and CLARITY cleared samples.

[0114] There are, however, some limitations to consider. The imaging speed in pLSM is partly constrained by the refresh rate of the projectors used. In the current implementation, we utilized projectors with a 60 Hz refresh rate, setting a maximum imaging speed of 60 frames per second. While this scanning speed suffices for most applications in structural imaging of large samples (typically imaged at 5-10 frames per second, Table 2), certain live imaging applications requiring very high imaging speeds (e.g., volumetric Ca2+or voltage sensor imaging) may not be fully compatible with pLSM. Additionally, while pLSM iscompatible with commonly used fluorophores (e.g., we demonstrated imaging of GFP, eYFP, mScarlet, Alexa 555 and Alexa647 dyes), the constrained excitation wavelengths may not be ideally suited for all possible fluorophores. Further, the multi-color imaging in pLSM is limited to three channels. Lastly, while the network control architecture provides advantages for ease of adoption and operation, fluctuations in network connection may limit applications in certain scenarios. It is worth noting that acquired images are not streamed via network but are saved in physically attached SSD disk, eliminating the risk of data loss during an imaging experiment.

[0115] Methods

[0116] pLSM implementation

[0117] pLSM setu . The pLSM system consists of two illumination arms and one detection arm (FIGS, la-b), mounted on 60 mm cage assembly system (Thorlabs) (Table 1 lists all components). Each illumination arm utilizes a MEMS LASER projector 10 (e.g., Nebra Anybeam), mounted on two dovetail translation stages (DT12) for precise adjustment along the light path. This projector incorporates TO-8 LASER diodes that provide blue (440- 460nm), green (515-530nm), and red (632-642nm) colors. Note that the pLSM system design is agnostic to the specific projector used as the control is implemented through universal display interfaces. A combination of two achromatic doublets (AC254-045-A) forms a Plbssl- style scan lens 20 with a focal length of approximately 22.5mm. An achromat doublet 32 (AC508-075-A) was paired with an air objective 40 (Mitutoyo lOx Plan Apo) for illumination. In the detection arm, the Mitutoyo lOx Plan Apo or ASI 16.67x / 0.4 NA detection objective 50 is used, along with a tube lens 55 (TTL200) and a CMOS camera 60 (GS3-U3-89S6M-C, FLIR) with a pixel dimension of 3.45 pm. A multi-band filter 52 (577 / 690nm, #87-251, Emund Optics or 432 / 515 / 595 / 730 nm, Semrock) was used as the emission filter.

[0118] To match refractive index, a custom-designed oil chamber is 3D printed using polylactic acid (PLA) and features windows for illumination and detection, sealed with cover glass (VWR). A refractive index matching oil with an RI of 1.454 (Cargille) is used to fill the chamber. Sample positioning is achieved using a motorized stage (PT3 Z8, Thorlabs) for fine adjustments during imaging, along with additional 3-degree-of-freedom dovetail translation stages (DT12XYZ, Thorlabs) and a rotation stage (QRP02, Thorlabs) for additionalflexibility during sample mounting. The sample is mounted in a quartz glass cuvette (FireflySci). The imaging workflow is implemented on Nvidia Jetson Nano board, which provides four high-speed USB 3.0, one HDMI, one DP, and one Ethernet ports. The HDMI ports of the two projectors 10 are connected to the board's HDMI and DP display ports using an HDMI- HDMI cable and an active DP to HDMI cable, respectively. The motorized stages are connected to motor control boxes (KDC 101). The projectors power cables, solid state hard drive, CMOS camera and the control cable for motor control boxes are connected to the USB ports of the Nvidia Jetson Nano board, with the help of a USB hub. The board is connected to the control computer via an Ethernet cable for remote control. Additional details regarding assembly, maintenance and operation are provided below in Supplementary notes 1-2.

[0119] pLSM control. The control software, implemented in Python, comprises several classes that handle different aspects of the system: (a) the Illumination Class controls the light sheet using the Xlib protocol to adjust the projector output, enabling precise control over the illumination settings; (b) the Camera Class utilizes the Spinnaker package to control the FLIR camera, allowing for parameter configuration and imaging trigger control; (c) the Stage Class is responsible for the precise control of Thorlabs motorized stages through APT communication protocol using thorlabs-apt-device Python interface; (d) the Widget Class is responsible for creating the software's graphical user interface (GUI) using the IPyWidgets package, providing an interactive control interface for the user; (e) the Experiment Class is responsible for setting all the experiment parameters, executing actions, and saving acquired images directly to the hard drive via USB 3.0. It coordinates the entire imaging process. Remote control of the software on the Nvidia Jetson Nano board is achieved through SSH protocol for secure remote access. The control GUI can be accessed through a web browser, providing convenience and flexibility in controlling the pLSM system remotely.

[0120] System PSF characterization. The point spread function of the system was measured by imaging 500nm sized fluorescent beads (17152-10, PolyScience), prepared by a l:0.5million dilution in 0.8% low melting agarose solidified in quartz cuvette. The point spread function was imaged by moving the beads in axial direction at 1pm step size.Different number of light sheet lines were used to test the tunable light sheet profiles. For the beads data analysis, images were first processed by ilastik to generate beads mask, which were then used to determine the center of mass coordinates for each bead. Beads from the 1original image were cropped and aligned based on the center of mass, followed by full width half maximum calculations.

[0121] Image processing. Median filter was applied to remove hot pixel noise. The image tiles were stitched using terastitcher or bigstitcher. The volume rendering was performed with Amira. For TH segmentation, suite WB pipeline was used as described previously. Briefly, the whole brain datasets were registered onto a local average reference atlas. The registered data was split into dense and sparse regions for multi-model segmentation. Finally, the detected cells were annotated with Allen brain atlas (ABA, ccfv3) regions for further comparison.

[0122] Fixed brain sample preparation

[0123] Mouse brain samples. For mouse dopamine and vasculature labeling, CD1 WT (CrkCDl(ICR), Charles river #22) mice were used. Thyl-eYFP (B6.Cg-Tg(Thyl- YFP)HJrs / J, JAX Strain 003782) mice were used for the CLARITY sample. All experiments were performed with 8-10 weeks old mice.

[0124] Neurotypical human brain samples. Brain tissue was obtained from the Quantitative Brain Biology Institute (brain QUANT) at the New York State Psychiatric Institute (NYSPI) and Columbia University Irving Medical Center. Brain tissue collection was conducted with IRB approval and informed consent obtained from next of kin who agreed to donate the brains and participate in psychological autopsy interviews. The right cerebral hemisphere, brain stem, and rostral 1.5 cm of the left cerebral hemisphere are sliced, placed on glass plates, rapidly frozen in environmentally safe, non-toxic Freon (1, 1,1,2- tetrafluorethane), and stored at -80°C. All brain specimens are tested for therapeutic levels of the most commonly prescribed or abused substances by state-of-the-art liquid chromatography / quadrupole mass spectroscopy at the Biomarkers Laboratory Resources (Core Lab) at the Irving Institute for Clinical and Translational Research, Columbia University. Routine neuropathological examination is performed on the fixed slices of the left hemisphere. If the subject was 45 years old or older, hippocampal formation and neocortex are stained for senile plaques and neurofibrillary tangles. Clinical and demographic information about all subjects is obtained through the Psychological Autopsy interviews with two or more informants, who were the next of kin to the deceased donor. For this study, we used a neurotypical control subject, male, 42 years old, white (Albanian), who diedaccidentally from electrocution, with negative toxicology for psychotropic drugs and alcohol, and time between demise and brain collection (post-mortem interval, PMI) less than 12 hours.

[0125] Alzheimer’s disease brain samples. De-identified Alzheimer’s disease cortical brain samples were obtained from Columbia University Alzheimer's Disease Research Center. The available phenotyping information includes - Age: 75, Gender: Woman, Site: BA9, frozen post-mortem Interval: 26:22.

[0126] Tissue fixation. The mouse brains were perfused with 4% PFA and extracted, followed by overnight fixation in 4% PFA at 4°C. Freshly frozen human brain samples were incubated in 4% PFA for one day at 4°C. All samples were washed and stored in IxPBS at 4°C.

[0127] iDISCO sample preparation. Standard iDISCO procedure was used, consisting of sample pretreatment, immunolabeling and clearing steps. For pretreatment, brain samples were dehydrated in 20%, 40%, 60%, 80%, 100%, 100% v / v methanol / FhO series for Ih each, followed by 66% v / v dichloromethane (DCM) / methanol incubation overnight. The samples were then washed by 100% methanol Ih twice and incubated in 5% H2O2 v / v in deionized water overnight at 4°C. Then rehydration was performed in 80%, 60%, 40%, 20% v / v methanol / FhO and PBS for Ih each. Before immunolabeling, the brain samples were first washed in PBS+0.2% Triton Ih for twice, followed by permeabilization in PBS+0.2% Triton at 37°C for 1 day. The samples were then blocked in 1% BSA in PBS for 1 day at 37°C. Antibody labelings were performed in PBS with 0.1% Triton / 1% BSA and corresponding primary antibodies. Anti-TH antibody (abl l3, Abeam) was used at 1:500 dilution. The mouse brain vasculature labeling was performed similar to previously reported procedure, by using a cocktail of antibodies: goat anti-podocalyxin (1:1500, AF1556, R&D Systems), goat anti-CD31 (1:300, AF3628, R&D Systems) and rabbit anti-Acta2 (1:1000, ab21027, Abeam). Human vasculature labeling was performed by using a cocktail of goat anti-podocalyxin (1:20, AF1658, R&D Systems), mouse anti-CD31 (1:50, BBA7, R&D Systems) and rabbit anti-Acta2 (1:1000, ab21027, Abeam). The brain samples were incubated in primary labeling solution for 10 days at 37°C, with one solution replacement at the end of day 5. The brains were washed in PBS + 0.1% Triton for 5 times at least Ih each and until the next day. For the mouse TH labeling, donkey anti- sheep Alexa 647 (1:1000, A-21448, Thermofisher) was used; for mouse vasculature labeling, donkey anti-goat Alexa 647 (1:500,A-21447, Thermofisher) and donkey anti-rabbit Alexa 647 (1:500, A-32795, Thermofisher) were used. For human vasculature labeling, donkey anti-goat Alexa 647 (1:500, A-21447, Thermofisher), donkey anti-mouse Alexa 647 (1:500, A-32787, Thermofisher) and donkey anti-rabbit Alexa 555 (1:500, A-31572, Thermofisher) were used. Secondary antibody incubation was also done for 10 days at 37°C, with one solution replacement at the end of day 5. For clearing, samples were dehydrated in 20%, 40%, 60%, 80%, 100%, 100% v / v methanol / FhO for Ih each step followed by incubation in 66% v / v DCM / methanol for 3h. The samples were then washed in DCM 15 min twice and put in dibenzyl ether (DBE, 33630, Sigma- Aldrich, RI=1.562) for refractive index matching until the next day for imaging.

[0128] Thyl-eYFP CLARITY sample preparation. PFA fixed brain samples were incubated in 1% hydrogel monomers (HMs) consisting of 1% w / v acrylamide (#1610140, BioRad), 0.5% w / v bisacrylamide (#1610142, BioRad), 4% paraformaldehyde (15710-S, Electron Microscopy Sciences), IxPBS, deionized water, and 0.25% of thermal initiator (VA- 44, Fisher Scientific) at 4°C overnight followed by oxygen removal and replacement of nitrogen using vacuum pump. The air-tight sample was then polymerized at 37°C for 6 hours. Then the sample was undergone passively clearing by incubating in the SBC buffer at 37°C with gentle shaking. Sample transparency was assessed to determine clearing completion, which usually took about 4-6 weeks. Sample was washed in 0.2M boric acid and can be left in upon imaging. RapiClear (SunJin lab, RI=1.47) was used for RI matching before imaging.

[0129] Passive SHANEL sample preparation. SHANEL tissue clearing was adapted as previously described with minor modifications. Briefly, freshly frozen ~0.7cm thick human brain samples (neurotypical and Alzheimer’s Disease) were randomly selected and fixed with 4% PFA in PBS overnight at 4°C. After washing in PBS, the samples underwent ethanol / FhO dehydration series (50%, 70%, 100%, 100% v / v; 2 hours / gradient) followed by DCM / methanol (2:1 v / v) delipidation overnight at room temperature (RT). Samples were then rehydrated in ethanol / FEO series (100%, 70%, 50%, 0% v / v; 2 hours / gradient) and switched to incubation with 0.5 M acetic acid (A6283, Sigma- Alderich) overnight at RT. Subsequently, samples were incubated with 4 M guanidine hydrochloride (G3272, Sigma- Alderich), 0.5 M sodium acetate (A6283, Sigma- Alderich), 2% Triton X-100 in PBS (pH 6) for 4 hours at RT and immediately replaced with a short incubation in a solution containing 10% w / v CHAPs (28300, Thermofisher), 25% w / v N-methyldiethanolamine (471828, Sigma- Alderich) for 2 hours. Samples were then washed with PBS before blocking in 0.2% TritonX-100, 5% BSA in PBS overnight at 37 °C. Primary antibodies of interest — goat anti-human podocalyxin (1:300, AF1658, R&D Systems), goat anti-CD31 (1:300, BBA7, R&D Systems), mouse anti-Acta2 (1:1000, ab21027, Abeam), rabbit anti-APP [Y188] (1:500, ab32136, abeam) — were diluted in staining solution (2% BSA, 0.2% Tween-20, 2% DMSO, 10 mg / L Heparin in PBS) for five-day labeling at 37 °C. Samples were then washed with a washing buffer (0.2% Tween-20, lOmg / L Heparin in PBS) overnight at RT before proceeding to secondary antibody incubation (1:500 for all Alexa 555 / 647-conjugated antibodies) in the staining solution for 5 days at 37°C. Finally, samples were washed in PBS, underwent ethanol / H2O dehydration series, delipidated in DCM for 4 hours at RT, and incubated in BABB (2: 1 benzyl benzoate:benzyl alcohol) to RI match for tissue transparency, which was achieved as early as within 3 hours in the solution. The samples were vacuumed for 30min in vacuum chamber to remove bubbles before imaging.

[0130] Bacterial strains and pellicle preparation

[0131] Pseudomonas aeruginosa strains used in this study are UCBPP-PA14 (WT, LD0) and PA14 PpAi / 4 / 3-mScarlet (LD4764). Liquid cultures were grown in Lysogeny Broth (LB) at 37 °C with shaking at 250 rpm. Overnight pre cultures were diluted 1:100 in LB and grown to mid-exponential phase (OD at 500 nm - 0.5). OD values at 500 nm values were read in a Spectronic 20D+ spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and cultures were adjusted to the same OD. Adjusted cultures were then mixed in a 2.5:97.5 ratio of fluorescentmon-fluorescent cells. Two mL of mixed culture was added to a 10 mm optical glass Type 1FL Macro Fluorescence Cuvette (1FLG10, Fireflysci). The cuvette was covered with parafilm and grown at 25°C for 3 days. The live imaging was performed with pLSM using ASI 16.67X / 0.4 NA objective and 4 pixels thick light sheet. The raw data was segmented by Ilastik framework by classifying pixels into tight clusters at air-liquid interface, individual cells or small cluster, and background regions. The color-coded projections across time were generated by using ImageJ / Fiji.

[0132] Brain and vessel organoids

[0133] Culture of hiPSCs. The hiPSC line EdiO42A was purchased from CedarsSinai. The hiPSC line WTC-ZOl-mEGFP (AICS-0023) was a gift from the Gordana Vunjak- Novakovic lab. Both cell lines were maintained in six- well plates coated with Matrigel growth factor reduced basement membrane matrix (GFR-Matrigel, Corning, 354230) inmTeSR plus medium (STEMCELL Technologies, 100-0276). To coat the six- well plates, 1.5 mL of GFR-Matrigel (diluted at a 1:100 ratio with DMEM / F12 medium, Gibco, 11320033) was added per well and incubated at 37°C for Ih. To passage hiPSCs (60-70% confluency), cells were rinsed with 3-4 mL of DMEM / F12 medium per well, and then 2 mL of Accutase (Sigma- Aldrich, A6964) was added for 5-6 min at 37°C. Subsequently, 2 ml of mTeSR plus medium was added to neutralize the dissociation. Cells were collected after centrifugation at 1000 rpm for 3 min. Cells were resuspended with mTeSR plus with 10 pM Y27632 (Selleckchem, S1049) and evenly distributed on GFR-Matrigel coated wells.

[0134] Generation of brain organoids. The brain organoids were generated from the Edi042 hiPSC line based on a previously described method with minor modifications. On day 0, hiPSCs were incubated with Accutase at 37°C for 7 min and dissociated into single cells. Approximately 10,000 single cells were added to a 96-well U-shaped-bottom low-attachment plate per well (Thermo Scientific, 174925) in mTeSR plus medium with 10 pM Y27632, and incubated at 37°C with 5% CO2. From day 1 to day 6, the medium was replaced with neural induction medium (NIM) daily, containing DMEM / F12 medium, 20% Knockout Serum Replacement (KSR, Gibco, 10828028), 1% Minimum Essential Medium non-essential amino acids (MEM-NEAA, Gibco, 11140050), 0.1 mM 2-mercaptoethanol (Gibco, 21985023), 50 pg / mL Gentamycin (Gibco, 15750060), 10 pM SB431542 (Selleckchem, S1067) and 5 pM dorsomorphin (Medchemexpress, HY-13418). From day 7 to day 26, the medium was replaced with neural differentiation medium every other day, containing neurobasal medium (Gibco, 21103049), 2% B27 supplement minus vitamin A (Gibco, 12587010), 1% GlutaMAX (Gibco, 35050079), 50 pg / mL Gentamycin, 20 ng / mL EGF (ProSpec, CYT-217) and 20 ng / mL FGF2 (ProSpec, CYT-557). From day 27, the EGF and FGF2 were replaced with 20 ng / mL of brain-derived neurotrophic factor (BDNF, ProSpec, CYT-1081) and 20 ng / mL of neurotrophin-3 (NT3, ProSpec, CYT-257).

[0135] Generation of vessel organoids. The vessel organoids were generated from the WTC-ZOl-mEGFP line as previously described with minor modifications. On day 0, hiPSCs were dissociated into single cells with Accutase. The collected cells were resuspended in mTesR plus medium with 10 pM Y27632. 1,000 cells were reaggregated in a 96-well U-shaped-bottom low-attachment plate per well. On day 1, the medium was replaced with N2B27 medium, containing 48.5% (v / v) DMEM / F12 medium, 48.5% (v / v) neurobasal medium, 0.5% N2 supplement (Gibco, 17502048), 1% B27 supplement minus vitamin A,0.5% MEM-NEAA, 0.5% GlutaMAX, 0.1 mM 2-mercaptoethanol, 50 ug / mL Gentamycin, 12 M CHIR99021 (Tocris, 4423) and 30 ng / mL BMP4 (ProSpec, CYT-1093). On day 4, the medium was replaced with N2B27 medium + 100 ng / ml VEGFA (Peprotech, 100-20) + 2 pM forskolin. On day 6, the medium was replenished with N2B27 medium + 100 ng / ml VEGFA + 100 ng / ml FGF2 every other day.

[0136] Immunostaining. Brain organoids at day 37 and vessel organoids at day 14 were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, 15713) at 4°C overnight. The organoids were washed three times with phosphate-buffered saline (PBS, Corning 46013CM), blocked in 10% donkey serum for 2h and incubated with primary antibodies at 4°C overnight. After aspirating the primary antibodies, the organoids were washed with PBS Tween-20 buffer (PBST, Thermo Fisher, 28352) three times for 30 min and incubated with secondary antibodies at 4°C overnight. The organoids were washed three times with Tris-buffered saline Tween-20, (TBST, Thermo Fisher, 28360) and stained with 4',6-Diamidino-2-phenylindole dihydrochloride, 2-(4-Amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI, Millipore-Sigma, D9542) for 15 min.

[0137] Tissue clearing. Organoids were embedded in 1% agarose gel and lined up before tissue clearing. The stained organoids were cleared by FDISCO as previously described. Samples were dehydrated with tetrahydrofuran (Sigma, 676764) solutions at graded concentrations: 50% (v / v), 70% (v / v), 80% (v / v), 100% (v / v) and 100% (v / v), with Ih for each step and incubated at 4°C. Then the samples were transferred to dibenzyl ether (DBE, Sigma, 33630) and incubated for Ih at room temperature.

[0138] Antibodies . The antibodies used in this study are listed as follows with the dilution ratios: goat-anti-human SOX2 (R&D, AF2018, 1:200), rabbit-anti-human CD31 (Abeam, ab28364, 1:200); Alexa Fluor 647 Donkey anti-goat IgG (Jackson Immunoresearch Lab, 705-606-147, 1:1000), Alexa Fluor 488 Donkey anti-rabbit IgG (Jackson Immunoresearch Lab, 711-546-152, 1:1000), Alexa Fluor 647 Donkey anti-rabbit IgG (Thermo Fisher, A32795, 1:1000).

[0139] Table 2. Specification comparison of pLSM with other accessible, open- source light-sheet microscopy platforms. It is important to note that the specifications listed are based on the published information and interpolation if not directly provided. Also, pricesfor other systems are considered as provided in respective publications, without any inflation adjustments. structural / objective lateral / cost eff. camera image pixel / pixel MP / sec / live axial $US magnif. pixel pixel image rate kUSD imaging reso- size size (MP) (MP / demo lution ( m) (pin) sec)(pm) pLSM Structural / Mitutoyo 1.74 / 6.13 10k- lOx 3.45 0.345 8.9 44.5 2.97 -Live lOx / O.28 15k (a) 4.45(air) ASI 16x / 0.91 / 5.6 24k- 16.7x 3.45 0.207 8.9 44.5 1.48 -0.4 (oil) 30k (a) 1.85Meso- Structural Olympus 2.7 / 5 162k 4x 6.5 1.625 4 20 0.12SPIM MVMPLA (0.63x-Version PO 6.3x)5 (b) lx / 0.25(air)Meso- Structural Mitutoyo 2.6 / 3.3 95k 5x 4.25 0.85 15 37.5 0.39SPIM 5x / 0.14Benchto (air)P (b) Mitutoyo 1.8 / 3.8 95k lOx 4.25 0.45 15 37.5 0.39 lOx / O.28 (air)Mitutoyo 1.5 / 4 95k 20x 4.25 0.213 15 37.5 0.3920x / 0.28 (air)Desc- Structural Thorlabs 4.2Z7.2 20k- lx 3.45 3.45 8.9 22.25 0.45-SPIM TL2X- 50k 1.11(c) SAP2x / 0.1 (air)Open- Structural / Varies N / A 30k- Varies Varies Varies Varies N / A N / ASPIM Live (water) 50k

[0140] Notes for Table 2: (a) The pixel rate for pLSM is estimated based on full frame image acquisition for structural imaging mode. For live imaging mode, the frame rate can reach up to 30Hz. (b) For mesoSPIM Version 5 and the mesoSPIM Benchtop systems, camera size references are drawn from the mesoSPIM Benchtop publication. The throughput for mesoSPIM Version 5 is estimated using a frame rate of 5Hz derived from the 200mssweep time when axially swept light-sheet mode (ASLM) is on, as reported in the corresponding paper, while the mesoSPIM Benchtop estimation uses a frame rate of 2.5Hz. (c) The pixel rate for descSPIM is deduced from the camera's reported 8.9MP size and an imaging frame rate of 5Hz, with an 8-bit pixel depth, which is then adjusted to a 16-bit equivalent for a more accurate comparison.

[0141] SECTION 1 - SUPPLEMENTARY NOTE 1

[0142] System Setup

[0143] pLSM system control setup

[0144] Begin setting up the pLSM system by gathering all necessary components listed in Table 1. Download the system control image from Github project page and clone it to an SD card using Win32DiskImager or similar software of your choice. Insert this SD card into the Nvidia Nano Board connected to a monitor, keyboard, mouse, and Ethernet. Power up the board and log into the Ubuntu control system. Use the ‘ifconfig’ command in the terminal to obtain the IP address. At this stage you may disconnect the monitor, keyboard, and mouse as the system is ready to be fully operated via remote network connection.

[0145] Next, connect the Thorlabs motor stages, the camera and projectors to Nano Board via a USB ports and HDMI port. Note that, serial numbers of the stages will be needed - if not known they can be found by connecting to a Windows PC and using Thorlabs APT software. For remote control of the pLSM imaging system, establish an SSH connection using Putty (Windows) or Terminal (Mac) with the IP address (identified above). Start the Jupyter notebook through Putty or Terminal using no browser mode, which will generate a remote-control link for opening in a browser on any machine. In this interface, Jupyter notebook and various .py class files are provided. In the stage.py file, edit the stage serial number and model name of the Thorlabs motor stages. Finally, open and run the microscope control Jupyter notebook that will automatically detect the camera, projectors, and motor stages. Once these components are detected and operational, the control GUI (see below for details) will be displayed and ready for use.

[0146] pLSM hardware assembly

[0147] The pLSM can be constructed using either a rail or a caged system, e.g., as depicted in FIG. la. In one embodiment, we assembled a caged system on a breadboard that features l / 4"-20 threaded holes. To build a 60mm cage system, we utilize various components from Thorlabs (as detailed in Table 1): mounting brackets (LCP01B, Thorlabs), cage plates (LCP09, Thorlabs), cage assembly rods (ER12 or other lengths, Thorlabs) and optical posts (TRI, Thorlabs).

[0148] Optical components mounting. The AnyBeam projector 10 features l / 4"-20 threaded holes, making it compatible for connection to an optical post (TRI, Thorlabs). This post is then attached to an angle bracket (DT12A, Thorlabs). The angle bracket is mounted on two-axis miniature dovetail translation stages (DT12, Thorlabs), allowing for precise adjustment of the projector's centre. These translation stages are subsequently mounted to the cage system via another angle bracket (DT12A, Thorlabs) and a 60mm cage mounting bracket (LCP01B, Thorlabs). For the Plbssl scan lens assembly 20, two achromatic doublets (AC254-045-A, Thorlabs) are mounted in an SMI lens tube (SM1L10, Thorlabs). The SMI lens tube is further connected to an SM2 tube using adapter (SM2A6, Thorlabs). Illumination tube lens 32 (AC508-075-A, Thorlabs) is directly fitted in SM2 tube. Detection tube lens 55 (TTL200) is mounted in adapter (SM2A20, Thorlabs) and then connected to SM2 tube. The detection objective lens 50, a Mitutoyo lOx Plan Apo, is attached to an adapter (SM2A6, Thorlabs) and then connected to an adjustable lens tube (SM2V05 or SM2V10, Thorlabs) for fine tuning. Camera 60 (GS3-U3-89S6M-C, FLIR) can be mounted on adapter (SM2A31, Thorlabs) which is connected to SM2 tube.

[0149] Illumination arm assembly. For assembling the illumination arm, begin by positioning the Plbssl scan lens 20 immediately in front of the projector 10, with the distance between them as close as possible (approximately 0 cm). Utilize software control to produce a small, centered laser pattern (for instance with settings of width=10, height=10, center=0, offset=0). Place an alignment target in front of the scan lens and adjust the lateral location and pitch angle of the projector so that its illumination center aligns with the optical axis. Next, change the software settings to generate a single pixel light sheet (for example width=l, height = 500).

[0150] Finely adjust the axial location of the scan lens to ensure that the light sheet's height remains consistent in front of the scan lens when the target is moved axially. With the light sheet still active, place the illumination tube lens approximately 6.5 cm in front of the scan lens. Adjust its position so that the width of the light sheet in front of the tube lens stays consistent when the target is moved axially. Continue to keep the light sheet on and position the illumination objective about 7.5 cm in front of the tube lens. Adjust its location so that the height of the light sheet in front of the objective remains the same when moving the target axially. Lastly, make adjustments to the above steps to ensure that the focus of the illumination objective approximately aligns with the detection axis.

[0151] This process is shown in FIGS. 5a-e, which shows the sequential assembly and alignment of the illumination arm. More specifically, FIG. 5a depicts the setup for aligning the illumination arm, showing the alignment target in position to assess the alignment accuracy of the components, and FIGS. 5b-e illustrate the step-by-step alignment of each optical component. The process involves iterative adjustments of the optical components’ positions in conjunction with specific illumination patterns projected through them. The alignment target, positioned at various points along the light path, aids in visualizing the correct pattern when it is moved axially, confirming accurate alignment. FIG. 5b shows the alignment of the projector center; FIG. 5c shows the alignment of the scan lens location; FIG. 5d shows the alignment of the tube lens location; and FIG. 5e shows the alignment of the objective location.

[0152] Detection arm assembly. Start by mounting the detection objective 50 so that its focal plane is approximately aligned with the illumination plane. If using an oil objective, first mount the oil chamber, then carefully insert the objective through the chamber hole, passing it through the latex seal. Following this, mount the detection tube lens 55 (TTL200). Finally, connect the lens tubes (SM2L20 and SM2L30, Thorlabs) on TTL200 adapter, and place the camera 60 (with adapter SM2A31) on the back of the tube. This will ensure that the camera chip is placed at the working distance of the tube lens 55.

[0153] Alignment. As shown in FIG. 6, a 45-degree prism mirror with scratch will be used for fine optical alignment. A neutral-density (ND) filter (or emission filter) is used on detection arm. Using the control software, project a wide field of illumination (e.g. max width, max height settings, FIG. 6b.i). Adjust the mirror so that it reflects the illumination light into the detection arm. Adjust detection objective 50 such that the scratch on the mirroris in focus (mirror on detection focal plane). Gradually narrow down the width of the illumination light, simultaneously adjust the mirror's z-axis location to keep reflecting the illumination into the detection arm, effectively moving the mirror closer to the illumination axis (FIG. 6b ii-iii). You'll need to refocus the detection objective 50 to ensure the mirror scratch remains sharp and in focus. In parallel, adjust the location of the illumination objective(s) 40 to make the illumination pattern sharper (mirror on illumination focal plane). Iteratively repeat above steps until the mirror's center stays in focus under wide field illumination (FIG. 6b.i), and a clear, sharp light sheet is visible at the mirror's center with single pixel illumination (FIG. 6b. iv). This suggests that the detection and illumination focal planes intersect at the mirror center, indicating the precise interception of the illumination and detection axes.

[0154] FIG. 6a shows the initial alignment setup, where a prism mirror is positioned around the focal plane of the detection optics to serve as an alignment target. FIG. 6b depicts a series that illustrates how specific illumination patterns projected by the system are reflected by the mirror and captured as imaged patterns, guiding the fine adjustments of the optical components, as follows: FIG. 6b.i. With the system set to wide field illumination, the detection objective 50 is fine-tuned until the mirror's scratch comes into sharp focus. This step confirms that the mirror is on the detection focal plane. FIG. 6b.ii. As the illumination pattern is progressively narrowed, discrepancies in the imaged pattern may emerge, suggesting a misalignment between the mirror's center and the illumination axis. Correcting this involves adjusting the mirror's z-axis position to realign the illumination pattern with the center, followed by revisiting step i to re-focus the detection objective. FIG. 6b.iii. Through iterative refinements of steps i and ii, the narrow illumination pattern should align with the reflected pattern, indicating proper adjustment. At this stage, the illumination objective(s) 40 is adjusted to ensure the reflected light sheet pattern is sharp, signifying that the illumination focal plane intersects the mirror's center. FIG. 6b.iv. The final adjustment entails using a single line of illumination. The resulting sharp, centered reflection of the light sheet and the clearly focused scratch from the wide field illumination indicate that the mirror's center is at the convergence of both the illumination and detection axes, as well as their respective focal planes.

[0155] SECTION 1 - SUPPLEMENTARY NOTE 2

[0156] Experiment Setup

[0157] FIG. 7 depicts one example of a suitable graphical user interface (GUI) for the system. The GUIis composed of several panels: an image preview section, camera controls, stage controls, illumination adjustments, and experiment setup options. Panel a shows an image preview and parameter setup blocks. Camera settings, stage locations and illumination patterns can be controlled and the real time preview is shown above. Panel b shows settings for large FOV imaging for structural image of large samples with tiling. Three dimensional ranges, tiling overlap, imaging step sizes can be setup. Light sheet adaptation offset and anchor points locations can be setup below, panel c shows settings for light sheet scan experiment. This enables 3D scanning of light sheet while sample remain fixed, which is suitable for time-lapse 3D image with SPED settings. Panel d shows a wide field experiment setup. This enables time-lapse imaging of given number of frames with fixed light sheet and sample locations.

[0158] To begin with the experiment setup, secure the sample in the cuvette and attach it to the motor stage. Fill the chamber with immersion oil. Use the stage control panel to navigate the sample to the focal plane of the detection arm. Activate the camera and illumination panel to begin preview mode to locate the sample. Initially, command the projector(s) 10 to set the illumination light sheet to a higher thickness for easier sample detection. Adjust the stage to bring the sample into the camera's view. Then, command the projector(s) 10 to gradually reduce the light sheet thickness while simultaneously adjusting the sample's z-position and the light sheet's offset to maintain visibility of the sample in the camera view and align the light sheet closer to focus. The sample location process is complete once the light sheet reaches the desired thickness and the sample image is focused. Continue to move the sample to explore different areas in preview mode. Remember to continually fine-tune the light sheet offset when repositioning the sample because the light sheet optimal offset can vary at different sample locations (FIG. 2d).

[0159] Then we can move on to the experiment setup panel, which contains three experiment mode: large field of view (structural mode), wide field (live imaging mode) and light sheet scanning mode. The experiment will use the same imaging parameters from camera and illumination panels we used for preview.

[0160] The large field of view (FOV) experimental setup is designed for structural imaging of stacks of large, cleared samples with tiling. The setup parameters include the 3D ranges for sample space, z step size and tiling overlap. It is essential to note that due to the refractive index inhomogeneity often found in large samples, it can distort the light sheet path and alter the detection focal plane, leading to misalignment between the light sheet and the focal point of detection. To address this, we employ an adaptive light sheet technique (FIG. 2d). By varying the light sheet offset, we can correct for changes in the light sheet path, ensuring it aligns with the altered detection focal plane. To accurately determine the necessary light sheet offset across various spatial locations within the sample, a set of anchor points is established. The user manually navigates the sample through these points to identify the optimal light sheet offset at each location. An interpolation method is then applied to calculate the appropriate offsets for the remaining areas. During imaging, the light sheet adapts to these offsets in tandem with sample movement, ensuring consistent focus throughout the sample. This adaptive approach also accommodates samples prepared with different clearing techniques, which may exhibit slight variations in refractive index. pLSM is capable of imaging terabytes scale large, cleared samples, with estimated imaging time and data size in Table 3 for reference.

[0161] Table 3 I Estimation of pLSM imaging time and file size of 1cm3sample.This serves as a reference for the imaging time and data size. A z step size of 5pm is used for the estimation. Tiling overlap is not considered in this case.

[0162] For wide field experimental setup, it’s designed for time-lapse imaging of the sample with static light sheet or wide field illumination. A predetermined number of frames is required as parameter. Upon commencement of the experiment, the system begins realtime image acquisition at the configured frame rate until the desired number of frames is collected.

[0163] In light sheet scanning mode, the light sheet sweeps through a volume while the stage remains stationary, suitable for SPED (SPherical-aberration-assisted Extended Depth-of-field) style 3D imaging. This technique allows for 3D live imaging at an effective volume rate equal to the frame rate divided by the number of z-axis slices. A scanning range and scanning step number specify light sheet offsets for 3D volume scanning while keeping the sample stationary. The number of scanning volumes denotes how many 3D volumes will be captured in a single trial. The scanning duration and interval define the overall time frame and the time gaps between consecutive experimental trials.

[0164] SECTION 2

[0165] The embodiments depicted in FIGS. 8a and 8b are variations of the FIGS, la-b embodiment (with like reference numerals representing like elements), and take advantage of the concepts described above in Section 1 to advantageously provide a fully automated, portable projected Light Sheet Microscope (pLSM) for high-resolution imaging of large fixed and living samples that can provide the following features and benefits: (1) fully automated smart microscopy of very large pathological and normal biological samples (2) compact and highly simplified mechanical, electronics, optical and software footprint; (3) orders of magnitude lower cost due to the adoption of consumer-grade components, without compromising the imaging quality; and (4) network-based remote operation architecture for easy deployment and integration of data. Potential applications for these embodiments include high-content 3D Histopathology of cancer biopsies and other diseases, neuroscience research, bacterial biofilms live imaging among others.

[0166] Both the FIG. 8a embodiment and the FIG. 8b embodiment are pLSM systems that include three modules: illumination optics to generate light sheets for illumination a plane in sample, detection optics to take the image of the illuminated plane, and the sample mounting, movement and oil chamber.

[0167] These embodiments employ standard wide-field detection optics, although with highly optimized cost due to the successfully testing and integration of cheap consumergrade camera chips: These embodiments replace the sCMOS camera used in conventional LSM systems (which costs $15-20K) with a consumer-grade camera chip (such as Sony CMOS chips), and can still provide sufficient sensitivity to be used for microscopy applications.

[0168] Both the FIG. 8a embodiment and the FIG. 8b embodiment use up to 3 cameras (and associated dichroic splitters) for simultaneous multi-color imaging of biological samples. The detection objectives can range in price depending on the desired properties. We found that the pLSM system is compatible with various detection objectives, and that using a cheap Air objective from machine vision (optionally with a spherical aberration correction lens) provides sufficient image quality for many applications. In addition, we utilize a mechanism for ease of changing detection objectives.

[0169] Conventional LSM illumination arms typically use a costly LASER engine, galvo scanners for scanning the beam, a f-theta scan lens, tube lens and objective, and several associated controller electronics. But like the FIGS, la-b embodiment described above, the FIG. 8a and 8b embodiments can generate the same or better illumination light sheets at a very small fraction of cost by replacing the LASER engine, galvo scanners and all the associated electronics with an inexpensive pocket projector 10 (e.g., a MEMS LASER projector such as the Nebra Anybeam). This results in a highly compact and portable illumination arm with 3 wavelengths (corresponding to red, green, blue) for imaging. Note that the LASER Diode or other light source used in pocket projectors are customizable. In fact, multiple interchangeable projectors can be utilized to choose the one with the most favorable illumination wavelengths for a given experimental setup. This switch is as simple as pulling out the projector 10 from the mounting slot and inserting another one.

[0170] Using a pocket projector 10 also allows unique software-driven modulation of light sheet thickness to control the achievable axial resolution. This is achieved by controlling the pixel- width of projected lines to generate illumination light sheet. Further downstream, a pair of standard achromat lenses as Plbssl f-theta scan lens 20 are used in the FIG. 8a and 8b embodiments to replace use of an expensive f-theta scan lens that are typically used to flatten the light sheet.

[0171] FIGS. 8a and 8b depict two alternative hardware configurations: more specifically, FIG. 8a depicts a more compact implementation that relies only on the projector 10 and scan lens 20 in the illumination arm, and does not have a tube lens or an objective pair. To compensate for lack of de-magnification that would ordinarily be provided by using a tube lens and an objective lens, the scan angle range parameter in the projector is reduced. In contrast, the FIG. 8b embodiment includes a tube lens 32 and an illumination objective 40 (and is therefore more similar to the FIGS, la-b embodiment described above).

[0172] In both the FIG. 8a and FIG. 8b embodiments, an Electrically Tunable Lens (ETL) 34 is included at the illustrated position. The incorporation of the ETL is optional to allow automated matching of the Light sheet FOV with the detection FOV (i.e. the thinnest portion of the light sheet locates to the center of the detection FOV) to allow for correction in a sample position-dependent manner.

[0173] The light sheet in pLSM is generated by having the projector 10 project a line of defined pixel width (e.g., 1 pixel wide, 2 pixels wide, 3 pixels wide, etc.). This is controlled natively by X library in the control software. More specifically, the controller depicted in FIG. 1c commands the projector 10 to display a line by sending appropriate commands to an input port (e.g., an HDMI port) of the projector 10. When the line that is being displayed is projected out of the projector 10, the result is a sheet of light. The controller can vary the thickness of the sheet of light by sending appropriate commands to an input port of the projector 10 to vary the thickness of the line that is being displayed by the projector. For example, if the controller commands the projector 10 to display a line that is 1 pixel wide, the result will be a thin sheet of light. And if the controller commands the projector 10 to display a line that is 2 or more pixels wide, the result will be correspondingly thicker sheets of light. This property also allows easy modulation of the light sheet thickness using software (instead of requiring a hardware adjustment), and yields a flatter light sheet profile. This allows easy modulation of the axial resolution of the data acquired.

[0174] pLSM facilitates a fully automated procedure for efficient multi-resolution scanning of very large volumes defined by coordinates of two opposite comers of a cuboid. Firstly, the light sheet offsets and the light sheet axial position parameters (by controlling ETL) are optimized at a few locations distributed uniformly (e.g., a uniform grid of defined size) or defined by a criterion. These determined optimal parameters are used to generate a parameter look-up table for the entire volume by interpolation, which is then utilized to acquire high quality images throughout. Second, the automated scanning procedure determines (in 3D) which parts of the defined volume need to be imaged and at what resolution (as shown in FIG. 8c). This is achieved by first performing a rapid scan of the defined volume at low lateral resolution (by camera pixels binning) and low axial resolution. The low axial resolution is achieved by modulating light sheet thickness (i.e. using thicker light sheets in the Z direction) and continuous sample motion for a defined distance (e.g. 100 microns) while keeping the light sheet and camera exposure on. The resulting low-resolutiondata is used to generate a map of which range of z-planes and tiles throughout the volume contain interesting features based on pre-selected criteria, including existence of significant signal and existence of enough contrast and sharpness. Then, any plane or tile that includes an interesting feature is imaged at a higher resolution (e.g., using the full resolution of the camera) using a thinner light sheet.

[0175] Both the FIG. 8a embodiment and the FIG. 8b embodiment are light sheet microscopes for obtaining images of a sample that include a projector 10, a scan lens 20, a first set of optical components 34 or 32-40, and a first objective lens 50. The projector 10 is configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector 10. The projector can be implemented, e.g., using a MEMS laser projector 10. The scan lens 20 (which can be implemented, e.g., using a Plbssl lens) is positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light.

[0176] The first set of optical components 34 or 32-40 is positioned to route the flattened first sheet of illumination light into the sample, so that the flattened first sheet of illumination light travels through the sample. And the first objective lens 50 is positioned to accept light (e.g., fluorescent light) from a depth within the sample that corresponds to the flattened first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens 50 and towards at least one first camera 60. In the embodiments depicted in both FIG. 8a and FIG. 8b, the objective lens 50 is mounted so that its optical axis is perpendicular to the flattened first sheet of illumination light that travels through the sample.

[0177] In both the FIG. 8a and the FIG. 8b embodiments, the projector 10 can have an input port and can be configured to change characteristics of the projected image based on data that arrives at the input port. And a controller (FIG. 1c) is configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector. This controller can be further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector 10. The first data causes the projector to output a thick first line of light (e.g., with a width of at least 3 pixels). The controller is further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of theprojector. The second data causes the projector to output a thin first line of light (e.g., with a width of one pixel).

[0178] Both the FIG. 8a and the FIG. 8b embodiments can further comprise a movable stage (not shown) that is controllable by the controller. The controller is further configured to obtain multiple low resolution images of respective thick slices of the sample by sending the first data to the input port of the projector 10, and instructing the at least one first camera to capture respective images of the sample while the movable stage is set to each of a plurality of different respective locations in a Z direction. Optionally, in these embodiments, each pixel in the low resolution images includes at least 4 native-resolution pixels of the at least one first camera that have been binned together. Optionally, in these embodiments, the controller can be further configured to identify a region of interest from the low resolution images, and to subsequently obtain high resolution images of a plurality of thinner slices of the sample that correspond to the region of interest.

[0179] In both the FIG. 8a and the FIG. 8b embodiments, the first set of optical components 34 or 32-40 can include an electrically tunable lens 34 positioned between the scan lens and the sample. In these embodiments, the controller can be configured to (a) determine, based on feedback obtained using the at least one first camera, at least one axial position parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust the electrically tunable lens 34 based on the at least one axial position parameter. Alternatively and / or additionally, the controller can be configured to (a) determine, based on feedback obtained using the at least one first camera, at least one light sheet offset parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust a projected line offset based on the at least one light sheet offset parameter.

[0180] In the FIG. 8b embodiment, the first set of optical components includes a tube lens 32 and a second objective lens 40 (i.e., an illumination objective) arranged so that the flattened first sheet of illumination light that is output by the scan lens 20 will (a) enter the tube lens 32, (b) exit the tube lens 32 and enter a back aperture of the second objective lens 40, and (c) exit the second objective lens 40 and enter the sample. Optionally, these embodiments can include an electrically tunable lens 34 positioned between the tube lens and the second objective lens.

[0181] The sheet of illumination light that travels through the sample (as described in this Section 2) can be projected into the sample from only a single side of the sample, in which case only a single illumination arm will be necessary. Accordingly, in these embodiments, the second illumination arm that is depicted in both FIG. 8a and 8b can be omitted. But especially for large samples, projecting the sheet of illumination light into the sample from only a single side of the sample may not provide sufficient penetration of the light sheet before the light sheet degrades. It can therefore be beneficial to include a second illumination arm (e.g., as depicted in both FIG. 8a and 8b) that projects a sheet of illumination light into the sample from the opposite side of the sample, especially for larger samples. The second illumination arm is typically (but not necessarily) identical to the first illumination arm. When two illumination arms are provided, the light sheets emanating from those illumination arm should be aligned.

[0182] When the image projected by each of the projectors in the embodiments described above in connection with FIG. 8a and 8b includes only a single line of light, a projection of that single line of light will form a single sheet of light that is projected out of the projector 10. The scan lens 20 will then flatten this single sheet of light, and the first set of optical components 34 or 32-40 will route the flattened sheet of illumination light into the sample so that it travels through the sample.

[0183] But notably, because each of the projectors can project a wide variety of different images (in response to different video data that arrives at the projector’s input port), each of the projectors 10 can be configured to project a plurality of parallel lines of light, in which case a projection of those plurality of lines will form a plurality of sheets of light that are projected out of the projector. The plurality of sheets of light that are projected out of the projector 10 enter the scan lens 20, and the scan lens 20 will output a plurality of flattened sheets of illumination light. The first set of optical components 34 or 32-40 will route these flattened sheets of illumination light into the sample, so that the flattened sheets of illumination light travel through the sample.

[0184] Respective objective lenses 50 are positioned to accept light from respective depths within the sample that corresponds to the respective flattened sheets of illumination light that travel through the sample, and route that light through the back aperture of respective objective lenses and towards respective cameras 60.

[0185] FIG. 9 is a detail of the light sheets that are generated when the FIG. 8b embodiment is used, and the projectors in the FIG. 8B embodiment are controlled so that the image projected by the projectors 10 each include two lines of light (i.e., a first line of light and a second line of light). The projector 10 in the left illumination arm (shown in FIG. 8b) will project an image of two lines of light, so that a projection of these two lines of light form two sheets of light that are projected out of the projector 10. The left scan lens 20 (shown in FIG. 8b) accepts both of the sheets of light that are projected out of the projector, and outputs two flattened sheets of illumination light. The first set of optical components 34 or 32-40 route the two flattened sheets of illumination light (labeled LS#1 and LS#2 in FIG. 9) into the sample, so that the two flattened sheets of illumination light travel through the sample from left to right in FIG. 9.

[0186] Especially when the sample is relatively large, it can be beneficial to include a second illumination arm. When, the second illumination arm is present, the projector 10 in the right illumination arm (shown in FIG. 8b) will project an image of two lines of light, so that a projection of these two lines of light form two sheets of light that are projected out of the projector. The right scan lens 20 (shown in FIG. 8b) accepts both of the sheets of light that are projected out of the projector 10, and outputs two flattened sheets of illumination light. The first set of optical components route the two flattened sheets of illumination light (labeled LS#1 and LS#2 in FIG. 9) into the sample, so that the two flattened sheets of illumination light travel through the sample from right to left in FIG. 9. These two sheets of illumination light are aligned with their counterparts arriving from the left side of the sample.

[0187] The upper detection objective lens 50 is positioned to accept light from a depth within the sample that corresponds to the upper flattened sheet of illumination light LS#1 that travels through the sample, and route the accepted light through a back aperture of the upper detection objective lens 50 and towards at least one first camera 60. The lower detection objective lens 50 is positioned to accept light from a depth within the sample that corresponds to the lower flattened sheet of illumination light LS#2 that travels through the sample, and route the accepted light through a back aperture of the lower detection objective 50 lens and towards at least one second camera 60.

[0188] The sample can be translated and / or rotated in order to cause the flattened sheets of illumination light to travel through different portions of the sample, thereby making it possible to image the entire sample. Notably, using two flattened sheets of illuminationlight as depicted in FIG. 9 enables two slices of the sample to be imaged in parallel, which can cut the overall time needed to image the entire sample in half.

[0189] This provides a speed advantage that can be particularly beneficial in the context of surgery to remove a tumor. For example, if a surgeon is removing a tumor and wants to make sure that the entire tumor has been removed, the surgeon can take samples of the outer periphery of the removed tissue, perform a rapid clearing / labeling procedure on that tissue, and subsequently image those samples using the techniques described above in connection with FIG. 9. The speed of imaging of such samples can be increased further by using relatively thick light sheets to provide low resolution imaging, and only switching to higher resolution imaging (using thinner light sheets) when the low resolution images reveal that cancer cells might be present (e.g., as described above in connection with FIG. 8A and 8B).

[0190] SECTION 3

[0191] In the embodiments described above in connection with FIGS. 1A, IB, 8 A, and 8B, the sheet(s) of illumination light enter the sample from one or two lateral sides of the sample. And if we assume that the upper surface of the sample is flat, those light sheets will be parallel to the upper surface of the sample. In these embodiments, the objective lens of the detection arm is positioned above the upper surface of the sample, and it captures light that emanates upwards from the sample in a direction that is perpendicular to the upper surface of the sample.

[0192] In alternative embodiments (not shown), when the lower surface of the sample is supported by a transparent slide, the detection arm can be inverted with respect to the positions depicted in FIGS. 1A, IB, 8A, and 8B. More specifically, in the inverted configuration, the sheet(s) of illumination light enter the sample from one or two lateral sides of the sample, and those light sheets are parallel to the lower surface of the sample. In these embodiments, the objective lens of the detection arm is positioned below the lower surface of the sample, and it captures light that emanates downward from the sample in a direction that is perpendicular to the lower surface of the sample.

[0193] Note, however, that the approaches for forming light sheets described above in sections 1 and 2 and the configurations of components for forming light sheets described in those sections are not limited to situations in which the sheets of illumination light enterthrough the lateral sides of the sample. To the contrary, similar approaches can be used to implement embodiments in which (a) one or more sheets of illumination light enter the sample through the upper surface of the sample at an oblique angle, and the objective lens of an upper detection arm is positioned above the upper surface of the sample, and / or (b) one or more sheets of illumination light enter the sample through the lower surface of the sample at an oblique angle, and the objective lens of a lower detection arm is positioned below the lower surface of the sample. Examples of such embodiments are described below in connection with FIGS. 10-14.

[0194] FIG. 10 depicts an embodiment in which a sheet of illumination light enters the sample through the upper surface of the sample at an oblique angle, and the objective lens 50 of the detection arm is positioned above the upper surface of the sample. The light sheet (which is projected into the sample) is generated using a projector 10 in a manner similar to the embodiments described above in connection with FIG. 1A, IB, 8A, or 8B, except that the illumination arm is oriented with respect to the sample so that the sheet of illumination light exits the illumination objective 40 and enters through the upper surface of the sample. Note that in this embodiment, the flattened sheet of illumination light enters the upper surface of the sample with an angle of incidence of 30-60°, and the optical axis of the objective lens 50 in the detection arm is perpendicular to the flattened sheet of illumination light. The detection objective 50 in this embodiment can be either an oil immersion objective or an air objective with a cap containing a glass coverslip or an appropriate lens for spherical aberration correction.

[0195] FIG. 11 depicts another embodiment that is the inverted version of the FIG. 10 embodiment. In this FIG. 11 embodiment, the sheet of illumination light enters the sample through the lower surface of the sample at an oblique angle, and the objective lens of the detection arm is positioned below the lower surface of the sample. The light sheet (which is projected into the sample) is generated using a projector 10 in a manner similar to the embodiments described above in connection with FIG. 1A, IB, 8A, or 8B, except that the illumination arm is oriented with respect to the sample so that the sheet of illumination light exits the illumination objective 40 and enters through the lower surface of the sample. Note that in this embodiment, the flattened sheet of illumination light enters the lower surface of the sample with an angle of incidence of 30-60°, and the optical axis of the objective lens 50 in the detection arm is perpendicular to the flattened sheet of illumination light.

[0196] FIG. 12 depicts another embodiment in which two sheets of illumination light enter the sample through the upper surface of the sample at an oblique angle, and the objective lens of the detection arm is positioned above the upper surface of the sample. The geometry of the two illumination arms and the detection arm in this embodiment is similar to the geometry described in US patent 11,506,877, which is incorporated herein by reference in its entirety. But the initial portion (i.e., everything prior to the back aperture of the objective lens 40 in the illumination arm) of how each of the light sheets (which are projected into the sample) is generated in this embodiment is similar to the way that the light sheets are generated in the embodiments described above in connection with FIG. 1A, IB, 8A, or 8B. Note that in this embodiment, the flattened sheets of illumination light enter the upper surface of the sample with an angle of incidence of 30-60°, and the optical axis of the objective lens 50 in the detection arm is perpendicular to upper surface of the sample. Note also that while the FIG. 12 embodiment depicts two illumination arms, in alternative embodiments (not shown) one of those illumination arms is omitted.

[0197] FIG. 13 depicts another embodiment that is the inverted version of the FIG. 12 embodiment. In this FIG. 13 embodiment, two sheets of illumination light enter the sample through the lower surface of the sample at an oblique angle, and the objective lens 50 of the detection arm is positioned below the lower surface of the sample. Here again, the initial portion of how each of the light sheets (which are projected into the sample) is generated using a projector 10 in a manner similar to the embodiments described above in connection with FIG. 1A, IB, 8A, or 8B. And here again, while two illumination arms are depicted, one of those illumination arms can be omitted in alternative embodiments (not shown).

[0198] Finally, the illumination arms in the FIGS. 1A, IB, 8A, 8B, and 10-13 embodiments all include a projector 10, a scan lens 20, and a first set of optical components 32-40 positioned to route the flattened first sheet of illumination light into the sample (e.g., just an ETL 34 as depicted in FIG. 8A; or a tube lens 32 and an illumination objective 40 plus an optional ETL 34, as depicted in FIGS. 1A, IB, 8B, and 10-13) positioned between the projector 10 and the sample. But alternative embodiments can be implemented that do not include a scan lens in the illumination arm. For example, in a variation of the FIG. 10 embodiment depicted in FIG. 14, the output of the projector 10 is fed directly into the back aperture of the objective lens 40 in the illumination arm. And the embodiments of FIGS. IB, 8B, and 11-13 can be similarly varied.

[0199] In these embodiments, a projector 10 is configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector 10. A first set of optical components (which includes only the illumination objective 40) is positioned to route the first sheet of illumination light into the sample, so that the first sheet of illumination light travels through the sample. And a first objective lens 50 is positioned to accept light from a depth within the sample that corresponds to the first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens 50 and towards a camera 60.

[0200] CONCLUSION

[0201] While the present invention has been disclosed with reference to certain embodiments, numerous modifications, alterations, and changes to the described embodiments are possible without departing from the sphere and scope of the present invention, as defined in the appended claims. Accordingly, it is intended that the present invention not be limited to the described embodiments, but that it has the full scope defined by the language of the following claims, and equivalents thereof.

Claims

WHAT IS CLAIMED IS:

1. A light sheet microscope for obtaining images of a sample, the light sheet microscope comprising: a projector configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector, wherein the projector has an input port and is configured to change characteristics of the projected image based on data that arrives at the input port; a first set of optical components positioned to route the first sheet of illumination light into the sample, so that the first sheet of illumination light travels through the sample; a first objective lens positioned to accept light from a depth within the sample that corresponds to the first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens and towards at least one first camera; and a controller configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector.

2. The light sheet microscope of claim 1, wherein the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to output a thick first line of light, and wherein the controller is further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light.

3. The light sheet microscope of claim 2, wherein the thin first line of light has a width of one pixel, and wherein the thick first line of light has a width of at least three pixels.

4. The light sheet microscope of claim 2, further comprising a movable stage that is controllable by the controller, wherein the controller is further configured to obtain multiple low resolution images of respective thick slices of the sample by sending the first data to the input port of theprojector, and instructing the at least one first camera to capture respective images of the sample while the movable stage is set to each of a plurality of different respective locations in a Z direction.

5. The light sheet microscope of claim 4, wherein each pixel in the low resolution images includes at least 4 native-resolution pixels of the at least one first camera that have been binned together.

6. The light sheet microscope of claim 5, wherein the controller is further configured to identify a region of interest from the low resolution images, and to subsequently obtain high resolution images of a plurality of thinner slices of the sample that correspond to the region of interest.

7. The light sheet microscope of claim 1, wherein the first set of optical components comprises a second objective lens positioned between an output of the projector and the sample.

8. The light sheet microscope of claim 1, wherein the first set of optical components includes only a second objective lens that is positioned between an output of the projector and the sample.

9. The light sheet microscope of claim 1, wherein the first set of optical components comprises: a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; and a second objective lens positioned to route the flattened first sheet of illumination light into the sample.

10. The light sheet microscope of claim 1, wherein the first set of optical components comprises: a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; a second objective lens positioned to route the flattened first sheet of illumination light into the sample; anda tube lens positioned between the scan lens and the second objective lens.

11. The light sheet microscope of claim 10, wherein the first set of optical components further comprises an electrically tunable lens positioned between the tube lens and the second objective lens.

12. The light sheet microscope of claim 1, wherein the first set of optical components comprises: a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; and an electrically tunable lens positioned to route the flattened first sheet of illumination light into the sample.

13. The light sheet microscope of claim 1, wherein the projector comprises a MEMS laser projector.

14. The light sheet microscope of claim 1, further comprising the at least one first camera.

15. The light sheet microscope of claim 1, wherein the first objective lens is mounted so that its optical axis is perpendicular to the first sheet of illumination light that travels through the sample.

16. The light sheet microscope of claim 1, wherein the image projected by the projector includes a second line of light, so that a projection of the second line of light forms a second sheet of light that is projected out of the projector, wherein the first set of optical components is positioned to route the second sheet of illumination light into the sample, so that the second sheet of illumination light travels through the sample, and wherein the light sheet microscope further comprises an additional objective lens positioned to accept light from a depth within the sample that corresponds to the second sheet of illumination light that travels through the sample, and route that light through a back aperture of the additional objective lens and towards at least one second camera.

17. The light sheet microscope of claim 16, wherein the second line of light is parallel to the first line of light, and wherein the second sheet of illumination light is parallel to the first sheet of illumination light.

18. The light sheet microscope of claim 1, wherein the first sheet of illumination light is oriented parallel to an upper surface of the sample, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample.

19. The light sheet microscope of claim 1, wherein the first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°.

20. The light sheet microscope of claim 19, wherein an optical axis of the first objective lens is perpendicular to the first sheet of illumination light.

21. The light sheet microscope of claim 19, wherein an optical axis of the first objective lens is perpendicular to the upper surface of the sample or the lower surface of the sample.

22. A light sheet microscope for obtaining images of a sample, the light sheet microscope comprising: a projector configured to project an image of a first line of light, so that a projection of the first line of light forms a first sheet of light that is projected out of the projector; a scan lens positioned to accept the first sheet of light that is projected out of the projector, and to output a flattened first sheet of illumination light; a first set of optical components positioned to route the flattened first sheet of illumination light into the sample, so that the flattened first sheet of illumination light travels through the sample; and a first objective lens positioned to accept light from a depth within the sample that corresponds to the flattened first sheet of illumination light that travels through the sample, and route the accepted light through a back aperture of the first objective lens and towards at least one first camera.

23. The light sheet microscope of claim 22, wherein the projector has an input port and is configured to change characteristics of the projected image based on data that arrives at the input port, and wherein the light sheet microscope further comprises a controller configured to vary a thickness of the first line of light by sending corresponding data to the input port of the projector.

24. The light sheet microscope of claim 23, wherein the controller is further configured to operate the light sheet microscope in a low resolution mode during a first time by sending first data to the input port of the projector, wherein the first data causes the projector to output a thick first line of light, and wherein the controller is further configured to operate the light sheet microscope in a high resolution mode during a second time by sending second data to the input port of the projector, wherein the second data causes the projector to output a thin first line of light.

25. The light sheet microscope of claim 24, wherein the thin first line of light has a width of one pixel, and wherein the thick first line of light has a width of at least three pixels.

26. The light sheet microscope of claim 24, further comprising a movable stage that is controllable by the controller, wherein the controller is further configured to obtain multiple low resolution images of respective thick slices of the sample by sending the first data to the input port of the projector, and instructing the at least one first camera to capture respective images of the sample while the movable stage is set to each of a plurality of different respective locations in a Z direction.

27. The light sheet microscope of claim 26, wherein each pixel in the low resolution images includes at least 4 native-resolution pixels of the at least one first camera that have been binned together.

28. The light sheet microscope of claim 27, wherein the controller is further configured to identify a region of interest from the low resolution images, and to subsequentlyobtain high resolution images of a plurality of thinner slices of the sample that correspond to the region of interest.

29. The light sheet microscope of claim 22, wherein the first set of optical components comprises an electrically tunable lens positioned between the scan lens and the sample.

30. The light sheet microscope of claim 29, further comprising a controller configured to (a) determine, based on feedback obtained using the at least one first camera, at least one axial position parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust the electrically tunable lens based on the at least one axial position parameter.

31. The light sheet microscope of claim 29, further comprising a controller configured to (a) determine, based on feedback obtained using the at least one first camera, at least one light sheet offset parameter of the flattened first sheet of illumination light that travels through the sample, and (b) adjust a projected line offset based on the at least one light sheet offset parameter.

32. The light sheet microscope of claim 29, wherein the scan lens comprises a Plbssl lens.

33. The light sheet microscope of claim 22, wherein the first set of optical components comprises a tube lens and an illumination objective lens arranged so that the flattened first sheet of illumination light that is output by the scan lens will (a) enter the tube lens, (b) exit the tube lens and enter a back aperture of the illumination objective lens, and (c) exit the illumination objective lens and enter the sample.

34. The light sheet microscope of claim 33, further comprising an electrically tunable lens positioned between the tube lens and the illumination objective lens.

35. The light sheet microscope of claim 34, wherein the scan lens comprises a Plbssl lens.

36. The light sheet microscope of claim 22, wherein the projector comprises a MEMS laser projector.

37. The light sheet microscope of claim 22, further comprising the at least one first camera.

38. The light sheet microscope of claim 22, wherein the first objective lens is mounted so that its optical axis is perpendicular to the flattened first sheet of illumination light that travels through the sample.

39. The light sheet microscope of claim 22, wherein the image projected by the projector includes a second line of light, so that a projection of the second line of light forms a second sheet of light that is projected out of the projector, wherein the scan lens is positioned to accept the second sheet of light that is projected out of the projector, and to output a flattened second sheet of illumination light, wherein the first set of optical components is positioned to route the flattened second sheet of illumination light into the sample, so that the flattened second sheet of illumination light travels through the sample, and wherein the light sheet microscope further comprises a second objective lens positioned to accept light from a depth within the sample that corresponds to the flattened second sheet of illumination light that travels through the sample, and route that light through a back aperture of the second objective lens and towards at least one second camera.

40. The light sheet microscope of claim 39, wherein the second line of light is parallel to the first line of light, and wherein the flattened second sheet of illumination light is parallel to the flattened first sheet of illumination light.

41. The light sheet microscope of claim 22, wherein the flattened first sheet of illumination light is oriented parallel to an upper surface of the sample, and an optical axis of the first objective lens is perpendicular to the upper surface of the sample.

42. The light sheet microscope of claim 22, wherein the flattened first sheet of illumination light enters an upper surface of the sample or a lower surface of the sample with an angle of incidence of 30-60°.

43. The light sheet microscope of claim 42, wherein an optical axis of the first objective lens is perpendicular to the flattened first sheet of illumination light.

44. The light sheet microscope of claim 42, wherein an optical axis of the first objective lens is perpendicular to the upper surface of the sample or the lower surface of the sample.

Images