Deep learning method for pinhole array color image reconstruction.

The following paper proposes a combination of a supervised encoder-decoder neural network with coded apertures. Coded apertures provide improved sensitivity and signal-to-noise ratio (SNR) in planar images. The unique array design of this method overcomes the spatial frequency cutoff found in standard multi-pinhole arrays. In this design, the pinholes were positioned to minimize loss in spatial frequencies. The large number of pinholes results in significant overlapping on the detector. To overcome the overlapping issue, reconstruction of the object from the obtained image is done using inverse filtering methods. However, traces of duplications remain leading to a decline in SNR, contrast, and resolution. The proposed technique addresses the challenge of image distortion caused by the lack of accuracy in the inverse filter methods, by using a deep neural network. In this work, the coded aperture is combined with a deep convolutional neural network (CNN) to remove noise caused by pinhole imaging and inverse filter limitations. Compared to only using Wiener filtering, the proposed method delivers higher SNR, contrast, and resolution. The imaging system is presented in detail with experimental results that illustrate its efficiency.

Reduction in Irradiation Dose in Aperture Coded Enhanced Computed Tomography Imager Using Super-Resolution Techniques.

One of the main concerns regarding medical imaging is the danger tissue's ionizing due to the applied radiation. Many medical procedures are based on this ionizing radiation (such as X-rays and Gamma radiation). This radiation allows the physician to perform diagnosis inside the human body. Still, the main concern is stochastic effects to the DNA, particularly the cause of cancer. The radiation dose endangers not only the patient but also the medical staff, who might be close to the patient and be exposed to this dangerous radiation in a daily manner. This paper presents a novel concept of radiation reduced Computed Tomography (CT) scans. The proposed concept includes two main methods: minification to enhance the energy concertation per pixel and subpixel resolution enhancement, using shifted images, to preserve resolution. The proposed process uses several pinhole masks as the base of the imaging modality. The proposed concept was validated numerically and experimentally and has demonstrated the capability of reducing the radiation efficiency by factor 4, being highly significant to the world of radiology and CT scans. This dose reduction allows a safer imaging process for the patient and the medical staff. This method simplifies the system and improves the obtained image quality. The proposed method can contribute additively to standard existing dose reduction or super-resolution techniques to achieve even better performance.

Gamma Radiation Imaging System via Variable and Time-Multiplexed Pinhole Arrays.

Biomedical planar imaging using gamma radiation is a very important screening tool for medical diagnostics. Since lens imaging is not available in gamma imaging, the current methods use lead collimator or pinhole techniques to perform imaging. However, due to ineffective utilization of the gamma radiation emitted from the patient's body and the radioactive dose limit in patients, poor image signal to noise ratio (SNR) and long image capturing time are evident. Furthermore, the resolution is related to the pinhole diameter, thus there is a tradeoff between SNR and resolution. Our objectives are to reduce the radioactive dose given to the patient and to preserve or improve SNR, resolution and capturing time while incorporating three-dimensional capabilities in existing gamma imaging systems. The proposed imaging system is based on super-resolved time-multiplexing methods using both variable and moving pinhole arrays. Simulations were performed both in MATLAB and GEANT4, and gamma single photon emission computed tomography (SPECT) experiments were conducted to support theory and simulations. The proposed method is able to reduce the radioactive dose and image capturing time and to improve SNR and resolution. The results and method enhance the gamma imaging capabilities that exist in current systems, while providing three-dimensional data on the object.

Plasma dispersion effect based super-resolved imaging in silicon.

We present here a new method for shaping a pulsed IR (λ = 1550nm) laser beam in silicon. The shaping is based on the plasma dispersion effect (PDE). The shaping is done by a second pulsed pump laser beam at 532nm (in either a Gaussian mode or a donut mode) which simultaneously and collinearly illuminates the silicon's surface with the IR beam. Following the PDE, and in proportion to its spatial intensity distribution, the 532nm laser beam shapes the point spread function (PSF) by controlling the lateral transmission of the IR probe beam. The use of this probe in a laser scanning microscope allows imaging and a wide range of contactless electrical measurements in silicon integrated circuits (IC) being under operation. We propose this shaping method to overcome the diffraction resolution limit in silicon microscopy on and deep under the silicon surface.

Improved Margins Detection of Regions Enriched with Gold Nanoparticles inside Biological Phantom.

Utilizing the surface plasmon resonance (SPR) effect of gold nanoparticles (GNPs) enables their use as contrast agents in a variety of biomedical applications for diagnostics and treatment. These applications use both the very strong scattering and absorption properties of the GNPs due to their SPR effects. Most imaging methods use the light-scattering properties of the GNPs. However, the illumination source is in the same wavelength of the GNPs' scattering wavelength, leading to background noise caused by light scattering from the tissue. In this paper we present a method to improve border detection of regions enriched with GNPs aiming for the real-time application of complete tumor resection by utilizing the absorption of specially targeted GNPs using photothermal imaging. Phantoms containing different concentrations of GNPs were irradiated with a continuous-wave laser and measured with a thermal imaging camera which detected the temperature field of the irradiated phantoms. By modulating the laser illumination, and use of a simple post processing, the border location was identified at an accuracy of better than 0.5 mm even when the surrounding area got heated. This work is a continuation of our previous research.

A Photonic 1 × 4 Power Splitter Based on Multimode Interference in Silicon-Gallium-Nitride Slot Waveguide Structures.

In this paper, a design for a 1 × 4 optical power splitter based on the multimode interference (MMI) coupler in a silicon (Si)-gallium nitride (GaN) slot waveguide structure is presented-to our knowledge, for the first time. Si and GaN were found as suitable materials for the slot waveguide structure. Numerical optimizations were carried out on the device parameters using the full vectorial-beam propagation method (FV-BPM). Simulation results show that the proposed device can be useful to divide optical signal energy uniformly in the C-band range (1530-1565 nm) into four output ports with low insertion losses (0.07 dB).

Targeted Magnetic Nanoparticles for Mechanical Lysis of Tumor Cells by Low-Amplitude Alternating Magnetic Field.

Currently available cancer therapies can cause damage to healthy tissue. We developed a unique method for specific mechanical lysis of cancer cells using superparamagnetic iron oxide nanoparticle rotation under a weak alternating magnetic field. Iron oxide core nanoparticles were coated with cetuximab, an anti-epidermal growth factor receptor antibody, for specific tumor targeting. Nude mice bearing a head and neck tumor were treated with cetuximab-coated magnetic nanoparticles (MNPs) and then received a 30 min treatment with a weak external alternating magnetic field (4 Hz) applied on alternating days (total of seven treatments, over 14 days). This treatment, compared to a pure antibody, exhibited a superior cell death effect over time. Furthermore, necrosis in the tumor site was detected by magnetic resonance (MR) images. Thermal camera images of head and neck squamous cell carcinoma cultures demonstrated that cell death occurred purely by a mechanical mechanism.

Decoupling and tuning the light absorption and scattering resonances in metallic composite nanostructures.

Utilizing the localized surface plasmon resonance (LSPR) effect of metallic nanoparticles enables their usage as contrast agents in a variety of applications for medical diagnostics and treatment. Those applications can use both the very strong absorption and scattering properties of the metallic nanoparticle due to their LSPR effects. There are certain applications where domination of the scattering over absorption or vice versa would be an advantage. However, the scattering and absorption resonance peaks have practically the same spectral location for solid noble metal nanoparticles at a certain domination of one over the other. In this paper we present gold nanoparticles coated with silicon that switches the order between the scattering and the absorption magnitude at the resonance peak by up to 34% in scattering-absorption ratio and tune the plasmon resonance over the spectrum by up to 56nm. This is obtained by modifying the refractive index of the silicon coating of the nanoparticle by illuminating it with a pumping light due to the plasma dispersion effect in silicon.

Cellular superresolved imaging of multiple markers using temporally flickering nanoparticles.

In this paper we present a technique aimed for simultaneous detection of multiple types of gold nanoparticles (GNPs) within a biological sample, using lock-in detection. We image the sample using a number of modulated laser beams that correspond to the number of GNP species that label a given sample. The final image where the GNPs are spatially separated is obtained computationally. The proposed method enables the simultaneous superresolved imaging of different areas of interest within biological sample and also the spatial separation of GNPs at sub-diffraction distances, making it a useful tool in the study of intracellular trafficking pathways in living cells.

Superresolved labeling nanoscopy based on temporally flickering nanoparticles and the K-factor image deshadowing.

Localization microscopy provides valuable insights into cellular structures and is a rapidly developing field. The precision is mainly limited by additive noise and the requirement for single molecule imaging that dictates a low density of activated emitters in the field of view. In this paper we present a technique aimed for noise reduction and improved localization accuracy. The method has two steps; the first is the imaging of gold nanoparticles that labels targets of interest inside biological cells using a lock-in technique that enables the separation of the signal from the wide spread spectral noise. The second step is the application of the K-factor nonlinear image decomposition algorithm on the obtained image, which improves the localization accuracy that can reach 5nm and enables the localization of overlapping particles at minimal distances that are closer by 65% than conventional methods.

Cellular imaging using temporally flickering nanoparticles.

Utilizing the surface plasmon resonance effect in gold nanoparticles enables their use as contrast agents in a variety of applications for compound cellular imaging. However, most techniques suffer from poor signal to noise ratio (SNR) statistics due to high shot noise that is associated with low photon count in addition to high background noise. We demonstrate an effective way to improve the SNR, in particular when the inspected signal is indistinguishable in the given noisy environment. We excite the temporal flickering of the scattered light from gold nanoparticle that labels a biological sample. By preforming temporal spectral analysis of the received spatial image and by inspecting the proper spectral component corresponding to the modulation frequency, we separate the signal from the wide spread spectral noise (lock-in amplification).