Digital Inline Holographic Microscopy (DIHM) of Weakly-scattering Subjects
February 8th, 2014
The three-dimensional locations of weakly-scattering objects can be uniquely identified using digital inline holographic microscopy (DIHM), which involves a minor modification to a standard microscope. Our software uses a simple imaging heuristic coupled with Rayleigh-Sommerfeld back-propagation to yield the three-dimensional position and geometry of a microscopic phase object.
The overall goal of the following experiment is to reconstruct the 3D position in geometry of a chain of swimming bacterial cells. This is achieved by first growing bacterial cells to the correct concentration to provide a suitable sample for imaging. As a second step, the LED light source is set up and aligned, allowing holographic images to be obtained.
Next, the reconstruction software is used in order to obtain the 3D coordinates of a chain of bacterial cells. The results show the ability to reconstruct the 3D position and configuration of the bacterial cells based on holographic imaging. The main advantage of this technique over existing methods like confocal microscopy, is that label free volumetric imaging is possible at high frame rates.
To begin inoculate, 10 milliliters of KTY media with ice crystals from a frozen stock of streptococcus strain V 4 0 5 1 1 97 incubate the starter culture in a rotary shaker overnight at 35 degrees Celsius and 150 RPM. Once it has grown up, inoculate 10 milliliters of fresh KTY medium with 500 microliters from the saturated starter culture. Incubate for 3.5 hours at 35 degrees Celsius and 150 RPM until the cells reach an optical density at a wavelength of 600 nanometers of 0.8 to 1.0.
This represents approximately five times 10 to the eighth cells per milliliter. Then dilute the cells at one to 400 in fresh media to obtain the final concentration of approximately 1 million cells per milliliter. Next, fill a syringe with petroleum jelly and make a ring of grease one millimeter in height on a microscope slide.
Then place a drop of the sample solution in the center. Gently place a cover glass over the sample and press lightly on the glass to seal the edges, ensuring that the liquid is in contact with both the slide and the cover glass. The resulting sample chamber should end up 100 to 200 micrometers in depth.
Rotate a 60 x oil immersion lens into place and add a drop of oil. Then place the sample chamber on the microscope stage with the cover glass facing down. Make sure no air bubbles are formed in between the glass cover slip and the lens.
Focus on the bottom surface of the sample chamber and then focus the condenser. Next, switch off the standard illumination and place the LED head behind the condenser aperture of the microscope. Ensure the condenser is set to bright field mode and set the LED power supply to maximum output.
Then close the condenser aperture to its fullest extent and nudge the LED in its mount until the illumination is centered on the objective lens aperture. Switch on the computer and then redirect the light path to the microscope's camera port. Adjust the frame rate to 30 frames per second and the image size to five 12 by five 12 pixels in the image acquisitions software.
Check an image intensity histogram to ensure that no part of the image is saturated or underexposed. The frame exposure time should be as short as possible while retaining good contrast. If cells collect close to an interface, refocus the microscope so that bacterial cells are slightly defocused.
The cells should lie around 10 to 30 microns from the focal plane for optimal reconstruction. After refocusing, acquire a series of video frames and save as an uncompressed a VI file. Then launch the DIHM software available for free download to convert a video frame to a numerically refocused image stack.
Example frames are provided with the software to begin reconstruction. Input the frame of interest and its background as separate image files into the boxes. The background image fairly represents the background of the video in the absence of the hologram and will be used to suppress any fixed pattern noise that may interfere with the holographic localization and analysis.
Next, enter output stack parameters into the global settings boxes. First, enter the axial position of the first frame in start focus. Then enter the number of slices in the reconstructed stack in number of steps, and finally, enter the actual distance between each slice in the stack in step size.
Set the step size to be the same as the lateral pixel spacing in order to reconstruct objects with the correct proportions using the example data. This value should be 0.233 micrometers. Next, enter the camera's illumination, wavelength, and medium refractive index and the lateral sampling frequency.
In the pixel per micron box, the camera's lateral sampling frequency can be found by dividing the objective lens magnification by the camera sensor pixel size. Using the example data, the number would be 60 x divided by 14 micrometers equals 4.29 pixels per micron. Then if the center of the object is dark in the hologram, press the flip Z gradient button.
This will apply a filter that extracts objects below the focal plane in the original sample. Then turn on the band pass filter, which is responsible for suppressing the noisy pixel contribution to the image. It is applied to each image slice immediately after generation.
Next, check the intermediate output switch on off state and make sure they're on. When these switches are on, it will cause two videos to be created. The first output is the refocus stack with a file name ending in stack a V.The second stack is the actual gradient operation with a file name ending in gradient a VI.Ideally, this second stack will contain the object of interest highlighted as a bright object against a dark background.
After setting all parameters, press run in the main window to execute the commands. The selected frame will appear in the main box of the software. Then use the magnifying glass tool found on the toolbar to the left of the image to zoom in and out.
Use the rectangle tool to select a region of interest or ROI and capture as many of the holograms fringes as possible inside the rectangle to optimize the reconstruction. Finally, press the process button to generate the two separate image stacks. Inspect the resulting stacks using Image J available at this website.
If the object of interest can be clearly seen in the gradient image stack, proceed to the rendering section to extract the object coordinates. To begin rendering, press the feature extraction button to enable the X, Y, Z coordinate location function. Then enter a path and extension for the output coordinates file, select POV ray style in the output coordinate style box.
This causes the program to write an object file that can be visualized using the free POV ray ray tracing software package, which can be obtained from this website, reprocess the images to extract the objects coordinates. The program will provide a series of X, Y, Z coordinates in the selected ROI written to the file name in the output cords box. Extraction of object coordinates takes significantly longer than stacks generation.
Make sure that the sample POV ray file supplied with the reconstruction code is present in the same folder as the output coordinates File. Edit the sample dot pov file and replace the file name in inverted commas on the line. Include my file NC with the name of the data file generated by the reconstruction code.
Then customize the camera position, lighting, texture, and rendering options in POV ray. When finished, click the run button to render a bitmap image. This is an image of a streptococcus bacteria chain that was momentarily oriented.
End on to the focal plane. Objects in this configuration are difficult to reconstruct and typically yield a blob close to the focal plane. Shown here is a properly rendered image representative of most frames in this series in which the chain is not oriented directly along the optical axis.
The shape and position of the object are faithfully reproduced in the rendering as seen in the panel to the right. Once this technique has been mastered, it can be done in minutes.
View the full transcript and gain access to thousands of scientific videos
Overview
This study demonstrates the use of digital inline holographic microscopy (DIHM) to reconstruct the three-dimensional positions and geometries of swimming bacterial cells. The technique allows for label-free volumetric imaging at high frame rates, providing a significant advantage over traditional methods.
Key Study Components
Area of Science
- Neuroscience
- Biophysics
- Imaging Techniques
Background
- Digital inline holographic microscopy (DIHM) is a modified microscopy technique.
- It enables the unique identification of weakly-scattering objects in three dimensions.
- Holographic imaging can capture dynamic processes in real-time.
- Existing methods like confocal microscopy often require labeling, which can alter the behavior of biological samples.
Purpose of Study
- To reconstruct the 3D position and geometry of swimming bacterial cells.
- To demonstrate the effectiveness of DIHM in capturing live cell dynamics.
- To compare the advantages of DIHM over traditional imaging methods.
Methods Used
- Preparation of bacterial cell cultures at appropriate concentrations.
- Alignment of an LED light source for optimal holographic imaging.
- Utilization of reconstruction software to obtain 3D coordinates.
- Analysis of the resulting holographic images to assess cell configuration.
Main Results
- The 3D positions of bacterial cells were successfully reconstructed.
- The technique demonstrated high frame rate imaging without the need for labels.
- Results indicated a clear advantage over confocal microscopy in terms of imaging speed and sample integrity.
- Holographic imaging proved effective for dynamic biological processes.
Conclusions
- DIHM is a powerful tool for studying live bacterial cell dynamics.
- The method allows for detailed 3D imaging without altering the sample.
- This technique could be applied to various biological research areas.
Frequently Asked Questions
Executive Industry Relevance
Digital inline holographic microscopy (DIHM) enables label-free volumetric imaging of weakly-scattering biological samples at high frame rates, addressing a key limitation in live-cell imaging where traditional methods require labeling or lack 3D spatial resolution. This capability supports early discovery workflows by providing quantitative, model-free measurements of cellular position and geometry in native conditions, reducing mechanistic ambiguity in target validation and phenotypic screening. The technique’s compatibility with standard imaging hardware and scalable sample volumes positions it as a reusable platform for de-risking therapeutic hypotheses in discovery biology.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Scientific Value: Enables unambiguous, model-free measurement of object location and shape in microscopic samples, supporting therapeutic hypothesis interrogation without exogenous labels.
- Operational Value: Uses standard CMOS/CCD cameras and LED or laser illumination, allowing integration into existing microscopy workflows without specialized reagents.
- Predictive Value: Provides quantitative 3D coordinates of cellular structures, facilitating functional target validation and biological de-risking through direct observation of native dynamics.
Screening & Assay Development
- Scientific Value: Captures interference patterns that contain depth information for 3D reconstruction, enabling label-free assessment of cellular responses in complex microenvironments.
- Operational Value: Defines accessible sample volume via camera sensor size and illumination coherence, supporting standardization across platforms with volumes up to 5 mm³ using laser illumination.
- Predictive Value: Employs Rayleigh-Sommerfeld back propagation and Gouy phase anomaly heuristics to generate reproducible, quantitative outputs for assay readiness and scalability.
Translational & Preclinical Research
- Scientific Value: Supports continuity from discovery through preclinical validation by enabling real-time observation of dynamic biological processes in disease-relevant systems.
- Operational Value: Allows high-frame-rate imaging without sample alteration, preserving native behavior for longitudinal tracking in preclinical models.
- Predictive Value: Delivers mechanistic de-risking by reducing ambiguity in cellular geometry and positioning, informing risk-adjusted advancement decisions.
Pipeline & Workflow Integration
DIHM fits within the discovery continuum from hypothesis testing in early biology to lead identification and preclinical validation, particularly where label-free, volumetric, and dynamic imaging are required to compare experimental conditions.
- Discovery Biology: Supports hypothesis testing and pathway clarification by providing model-free 3D localization of weakly-scattering objects like bacterial cells or subcellular structures.
- Screening: Enables assay readiness through quantitative, refocused image stacks that allow comparison of cellular configurations across conditions.
- Analytics: Generates X, Y, Z coordinate outputs via feature extraction, enabling statistical analysis of spatial distributions and motion patterns.
- Translational Research: Connects to preclinical continuity by preserving sample integrity during longitudinal imaging of dynamic processes.
- Enterprise Reuse: Functions as a reusable imaging capability rather than a single-use technique, leveraging free reconstruction software and standard hardware.
Operational & Enterprise Impact
- Scientific Value: Predictive confidence through label-free, model-free 3D reconstruction of object geometry and position.
- Operational Value: Standardization and reproducibility via numerical refocusion and fixed-pattern noise suppression using background subtraction.
- Strategic Value: Better go/no-go decisions by reducing late-stage biological risk through early, high-fidelity observation of native cellular dynamics.
- Portfolio Impact: Risk-adjusted prioritization via quantitative spatial data that informs target confidence and lead selection.
Implementation Considerations
- Requires expertise in holographic image acquisition and numerical reconstruction using Rayleigh-Sommerfeld back propagation.
- Needs instrumentation capable of plane-wave illumination and CMOS/CCD image capture at defined frame rates and resolutions.
- Demands cross-team standardization of illumination coherence, step size, and lateral sampling frequency for reproducible 3D reconstruction.
- Requires adaptation of output stack parameters (start focus, number of steps, step size) to match experimental scale and magnification.
- Practical limitation: optimal reconstruction requires objects to lie 10–30 microns from the focal plane, necessitating precise sample preparation and focusing.
