Compact, cost-effective and field-portable microscope prototype based on MISHELF microscopy

Abstract

We report on a reduced cost, portable and compact prototype design of lensless holographic microscope with an illumination/detection scheme based on wavelength multiplexing, working with single hologram acquisition and using a fast convergence algorithm for image processing. All together, MISHELF (initials coming from Multi-Illumination Single-Holographic-Exposure Lensless Fresnel) microscopy allows the recording of three Fresnel domain diffraction patterns in a single camera snap-shot incoming from illuminating the sample with three coherent lights at once. Previous implementations have proposed an illumination/detection procedure based on a tuned (illumination wavelengths centered at the maximum sensitivity of the camera detection channels) configuration but here we report on a detuned (non-centered ones) scheme resulting in prototype miniaturization and cost reduction. Thus, MISHELF microscopy in combination with a novel and fast iterative algorithm allows high-resolution (μm range) phase-retrieved (twin image elimination) quantitative phase imaging of dynamic events (video rate recording speed). The performance of this microscope prototype is validated through experiments using both amplitude (USAF resolution test) and complex (live swine sperm cells and flowing microbeads) samples. The proposed method becomes in an alternative instrument improving some capabilities of existing lensless microscopes.

Portable deep learning singlet microscope

“…Lensless or singlet lens imaging setup is the simplest imaging system, which is very attractive in portable microscopy and field practices . Although diffraction‐based lensless microscopy works well, its imaging mechanism results in short working distance which means the bio‐sample must be very close to the CCD/CMOS image sensors .…”

3D-printable portable open-source platform for low-cost lens-less holographic cellular imaging

“…Electron holography today is a mature research field and widely in use, allowing for molecular imaging [9,10]. Digital holography within the visible wavelength regime has attracted considerable attention where a broad spectrum of light sources has been employed, ranging from gas [11,12], solid state [13] and semiconductor [1,14,15,16] lasers over light emitting diodes (LEDs) [17,18,19,20] to halogen lamps [21]. Recently, ultra-broadband digital holography with sunlight illumination has been demonstrated, employing a new reconstruction algorithm [22].…”

“…Moreover, lens-based DOAH has recently been employed in in-line industrial production control [40] and for imaging of semiconductor structures [41]. In life-sciences, DIHM has the advantage of working label-free, which enables non-invasive, in-vivo study of biological samples, for example RBCs [42,43,44], parasited mouse RBCs [45], sperm cells [15], diarrhea parasites [17], and several aquatic organisms [38]. Moreover, as information about the whole sample volume is encoded in one single hologram, processing speed is substantially increased as compared to microscopic scanning techniques such as confocal or fluorescence microscopy.…”

“…Furthermore, by acquiring and analyzing sequences of images, DIHM enables long term cell evolution studies [32] and micro-particle tracking [49]. Although differently complex, portable and low-cost DIHM solutions have recently been reported [50,18,15,51], critically important details neccessary for their realization in a laboratory are mostly disclosed. Such important details include details on employed of light sources, distances between objects and light source or detector, methods of reconstruction, non obvious limitations or design or construction files thus making it challenging to set up a DIHM without too much prior expertise and avoids access to low-cost micron spatial resolution imaging.…”

Untitled

“…Subsequent studies show similar results. Ibáñez-Sanz et al reported a discriminatory accuracy value of 0.63 for CRC risk prediction model combining some modifiable risk factors (alcohol, obesity, physical activity, red meat and vegetable consumption, and nonsteroidal anti-inflammatory drug use), family history of CRC, and a GRS based on 21 susceptibility SNPs [7]. More recently, Weigl et al [8] derived a GRS based only in a significant number of common variants (48 SNPs) previously associated with CRC.…”