Team:TU Darmstadt/tech/hardware

ChiTUcare

MainPage

Digital Inline Holographic Microscopy - An iGEM Approach

Our project aims to construct a low cost Digital Inline Holography Microscope (DIHM). The DIHM features an ease-of-use, lens-less inline structure, and a state-of-art reconstruction algorithm from holograms to 3D visualization with micrometer resolution. The working principle of a DIHM starts with a point laser source, emanating spherical waves, illuminating the object to be observed, and forming a magnified diffraction pattern at the image sensor, followed by reconstruction algorithms. The holograms collected by the image sensor already contain the difference of intensity and phase shifts, compared with the reference part of the spherical wave. Thus, the inline structure without the need of a lens or beam splitter. Our project uses easily accessible hardware components: Huawei P9 lite flashlight as a light source, a 5 µm pinhole, a Pi-cam and a Raspberry Pi for taking pictures, and certain 3D printed parts to assemble the microscope. The open source library HoloPy is then deployed to reconstruct the 3D volumes from the holograms.

Introduction

Digital inline holography microscopy is a special technique to recover object properties from interference patterns. We will explain the working principle in greater detail later on, but first let us show the basic properties of our DIHM setup in a short video. The light coming from a bright LED passes a pinhole, which acts as a spatial filter and emits spherical waves. At small samples, e.g. yeast cells, the wavefront is scattered. Small fringes are created. The unscattered waves serves as reference template. The interference pattern is captured on a camera screen.

The resulting hologram contains the phase and the intensity information, encoding the 3D model. We present an 'easy-to-use' interface for a commonly used analyzation package in our software section. First, we invite to check-out what we could achieve and then how to get our own DIHM.

Achievements

low-cost microscope solution with micrometer resolution
robust and portable setup
easily attachable to smartphone
comparison of different 3D printable micromanipulators and presenting another solution
testing different light sources
3D printed parts can be reproduced with the most basic 3D printers
software solution for analyzing the recorded images
extracting the 3D information out of the data, thus enabling 3D view
development of an user interface for HoloPy

We took pictures from different samples and analyzed them with our software HoloPyGuy. We demonstrate the capture of a micrometer scale used in optical microscopy (first row), which is the prove of our resolving capability. Further on, we captured red blood cells (second row) and bubbles (third row) and analyzed their 3D structure. For a detailed explanation see our software section.

A reconstructed hologram with HoloPyGuy. — **Figure 1.** At first, a picture of the beam without sample serves as reference (left). A raw hologram has fringes around the objects of interest (middle). The reconstructed object is calculated by substracting the reference picture from the raw hologram and then analyzed with reverse propagation.

Get It

To build our own DIHM, you need just few compoments. In the following section, we want to present you how to get it and why we decided on the used parts.

Bill of Materials

If you want to rebuild our setup, a listing is shown in the table below. The prices can vary between countries as well as the exchange. Also, shipping and taxes are not allways included.
You can download the 3D printed parts for our setup here.

Part	Price	Link to a reseller (not related nor repsonsible for the offer or the validity of the link)
Raspberry Pi Starter Kit	50 € (58 $)	amazon
Raspberry Camera module V2	30 € (35 $)	amazon
5 µm Pinhole	68 € (79 $)	Thorlabs
Online Printing service for two parts	15 € (17.5 $)	Trinckle
total DIHM	163 € (190.5 $)

Micromanipulators

In order to align our DIHM setup, it is necessary to focus the brightest spot of the light source to the pinhole. Due to the tiny dimension of this setup, we need to manipulate the objects in the micrometer level. Therefore, we tested different micromanipulators. We started with a concept of BDan from Thingiverse, which lacks flexibility and proportion. Later on, we tried 3D printed flexible bearing micromanipulators, movable with a screw. They failed on accuracy and the ability to move in more than one dimension. Our last and final approach is to split the movement axes using screws for control, as demonstrated below:

In the figure shown above, the green stack functions as a frame to fixate the screws. Thereby, one end is attached with springs allowing more bounce and flexibility. The yellow and the red stacks enable the alignment control in the x and y direction, correspondingly. The camera is then placed onto the red stack to record holograms once the light source, the object to be observed, and the camera is properly aligned. We used these micromanipulators to study the necessary parameters in our project. Due to the compactness of them we stepped back from the micromanipulators in our final setup improving the robustness on the other hand.

Light Sources

As the resolution of the microscope highly depends on the implemented light source, we decided to test a variety of different light sources. We started with an upcycling idea: why not use an old X-Box One's DVD Pickup, the advantage being the little cost for a reliable laser source. DIyouware Bros gives a detailed instruction on how to use the pickup, and a customized PCB board as a micromanipulator to align the laser source at the desired position with micrometer precision, and to control the intensity of the laser. However, a fatal disadvantage was discovered during our tests with the experimental setup. A Fresnel lens is integrated onto the laser diode within the pickup, producing an undesired interference pattern, thus destroying the recorded holograms. Other approaches are using an LED, a collimated light source, and a pinhole, to produce point source eminating spherical waves. The disadvantages of common LEDs is the low intensity, and the little coherence time compared to the laser source. The brightness can be compensated with high power LEDs, such as the ones used in fishkeeping, or even better, smartphone flash lights. Imagine a DIHM adapter for a smartphone: nowadays, most smartphones have tremendously bright flash lights. They are common LEDs with a high intensity peak lying in the blue light spectrum making it easy to detect in our experimental setup. Thus, offering a proper solution for a portable easy to use microscope.

Smartphone illuminated background image. — **Figure 3.** At first, a picture of the HD-DVD sled. When closely looked upon the small rings inside of the lens can be observed. These work as a Fresnel lens and create a focus of about 4 µm. The second picture shows the resulting light intensity profile, which can not be used for holography (without pinhole). The third picture is taken with a smartphone flashlight and pinhole. Black spots are caused by dust on the detector.

Electronics

Our initial plan was to use an Xbox DVD Pickup as the laser source, as it provides 405 nm, 650 nm, and 780 nm laser point source with possible laser alignment and intensity control by using a customized PCB board. However, as we experimented with the pickup, we found out that the hologram was greatly destructed by the Fresnel lens within the pickup. Thus we opted out on the pickup and PCB idea. Instead, our setup now consists solely of a smartphone flashlight, a pinhole, a micromanipulator and stage for alignment, and a V2 Picam with a RapsberryPi 3 for the recorded holograms.

pisetup — **Figure 4.** A Raspberry Pi is connected to a Raspi Pi Cam, which features a webserver for retrieving the captured pictures online. The electronics setup can be powered with batteries or a smartphone charger, which enormously increases the flexibility.

As shown in the picture above, RaspberryPi 3 is connected to power, Pi cam, and the LAN cable for internet connection. A webcam server is then setup with local host configuration such that a live stream from the Picam can be accessed via a browser on another PC. The webcam server is also customized with a GUI to record images or videos, adjust camera settings such as ISO, resolution, and save the recorded data on the server for eventual downloading.

Assembling it

If you would like to copy our setup we welcome you to watch our short assembling video.

Working Principle

The working principle of digital holographic microscopy relies on the interference pattern, i.e. the hologram, which encodes 3D information in a 2D picture, and on the 3D reconstruction algorithm, to extract the 3D information from the recorded holograms. The interference pattern comes from the joint wavefronts of the object beam and the reference beam. As indicated in the picture below, a coherent collimated light source is split into two beams: The object beam passing through the lens and the object that is to be observed, and the reference beam pertaining the phase and coherence of the light source. The joint wavefronts form the hologram, which is recorded by an image sensor. The 3D reconstruction algorithm then functions as a digital lens, cutting the joint wavefronts in minuscule distances near the object, rendering a stack of cross-sectional wavefronts. Based on the stack of wavefronts, or interference patterns, the image of the object can be reconstructed in different depths, thus achieving an axial resolution for 3D imaging.

**Figure 5.** Simple DHM setup: From left to the right: A laser beam is split in object and reference beam with a beam splitter. A series of optics focus the beam on the sample and magnifies the interference pattern. A second beam splitter superposes reference and object beam on the camera screen (picture taken from [17]).

The inline flavor of the DIHM kicks in when neither lenses nor beam splitters are needed as in the typical DHM setup. Instead a point light source is used, which can be produced by passing a collimated light source through a pinhole. The point source creates a spherical wave, illuminates the object, and reaches the image sensor containing the joint wavefronts of both, the reference beam and the object beam. Due to the fact that the peripheral of the wavefront passes by and remains unaffected by the object it creates the reference beam. The wavefront passing through the object forms the object beam.

**Figure 6.** Simple DIHM setup: From left to the right: camera, sample, pinhole and light source (LED).

In the above picture, a laser source of wavelength $λ$ emanates from a pinhole, forming a spherical wave. A small object is typically placed a few thousand wavelengths from the source, reaching the image sensor much further away such that magnification is achieved. Small object means that the object should only block a small fraction of the spherical cone wavefront recorded on the image sensor. Otherwise classical diffraction dominates the image, where 3D reconstruction based on holography would no longer work due to the lack of the reference beam.

The 3D reconstruction algorithm uses the Kirchhoff-Helmholtz transform to reconstruct the wavefront one at a time, on several planes at various distances near the object. When the stacks of reconstructions are made, a 3D image with depth information can be built.

References

[1]	Shiraki, A., Taniguchi, Y., Shimobaba, T., Masuda, N., Ito, T. (2012) Handheld and low-cost digital holographic microscopy. arXiv:1211.0336
[2]	Cotte, Y., Toy, F., Jourdain, P., Pavillon, N., Boss, D., Magistretti, P., Marquet, P., Depeursinge (2013) Marker-free phase nanoscopy Nature Photonics, 7 (2):113 DOI: 10.1038/nphoton.2012.329
[3]	Giuliano, C. B., Zhang, R., Wilson, L. G. (2014) Digital Inline Microscopy (DIHM) of Weakly-scattering Subjects Journal of Visualized Experiments, DOI:10.3791/50488
[4]	Molaei, M., Sheng, J. (2014) Imaging bacterial 3D motion using digital inline holographic microscopy and correlation-based de-noising algorithm Optics Express, DOI: 10.1364/OE.22.032119
[5]	Braat, J., Dirksen, P., Janssen, A. J. E. M. (2003) Diffractive Read-Out of Optical Discs, Optical Imaging Springer Verlag
[6]	Deng, Y., Chu, D., (2017) Coherence properties of different light sources and their effect on the image sharpness and speckle of holographic displays, Scientific Report, DOI: 10.1038/s41598-017-06215-x
[7]	Jericho, M. H., Kreuzer, H.J., (2011), Point Source Digital In-Line Holographic Microscopy, Chapter 1, Coherent Light Microscopy, Springer Series in Surface Sciences 46, 46 DOI: 10.1007/978-3-642-15813-1_1
[8]	Rostykus, M., Moser, C. (2017) Compact lensless off-axis transmission digital holographic microscope, Optics Express, DOI: 10.1364/OE.25.016652
[9]	Reichert, C. C., Herkommer, A., Claus, D. (2016) Das Smartphone als Mikroskop, AT-Fachverlag GmbH, www.biophotonik.de
[10]	Moon, I., Daneshpanah, M., Anand, A., Javidi, B. (2011) Cell Identification Computational 3-D Holographic Microscopy, Optics & Photonics, 22 (6),
[11]	Greenbaum, A., Luo, W., Su, T., Göröcs, Z., Xue, L., Isikman S., Coskun, A., Mudanyali, O., Ozcan, A. (2012) Imaging without lenses: achievments and remaining challenges of wide-field on-chip microscopy, Nature America, DOI:10.1038/nmeth.2114
[12]	beniroquai (2017) Blog, https://beniroquai.wordpress.com/2016/01/20/holoscope-linsenloses-holographisches-mikroskop/, last visited: 10/15/2017
[13]	BDan (2015) micromanipulator, Thingiverse, https://www.thingiverse.com/thing:923865/#files, last visited: 10/15/2017
[14]	"Do-it-yourself" project for steering HD-DVD pickup homepage: http://www.diyouware.com/ last visited: 10/15/2017
[15]	"HoloPy, a python framework for analyzing digital holographs, manoharan lab, Harvard https://github.com/manoharan-lab/holopy last visited: 10/18/2017
[16]	"HoloPy, documentation, https://holopy.readthedocs.io/en/latest/users/index.html last visited: 10/18/2017
[17]	User:Egelberg, OpticalSetupDHM, CC BY-SA 3.0
[18]	Schmid B., Schindelin, J., Cardona, A. et al. (2010) A high-level 3D visualization API for Java and ImageJ, BMC Bioinformatics 11(1): 1, DOI: 10.1038/nmeth.3392
[19]	Schindelin, J., Arganda-Carreras, I., Frise, E. et al. (2012) Fiji: an open-source platform for biological-image analysis, Nature methods 9(7): 676-682, DOI: 10.1038/nmeth.2019