Gokhan Bora Esmer

Off-axis holographic microscopy enables single-shot quantitative phase imaging, but its sensitivity is limited by mixed noise that degrades fringe visibility and biases phase recovery. We compare classical filters (Wiener/median/NLM/BM3D and variance-stabilized variants), a supervised Swin transformer, and a diffusion denoiser trained on synthetic data, using 100,000 physics-based holograms with realistic blur, speckle, shot, and read noise. On a 10,000-hologram test set, the best pipeline (diffusion followed by BM3D) achieves 36.86 dB/0.9964 intensity PSNR/SSIM and 46.48 dB/0.9772 phase PSNR/SSIM, reducing RMSE by 88.5% in intensity and 52.8% in phase, with similar improvements on experimental bead and cell holograms.

Off-axis holography enables single-shot phase retrieval but reduces spatial bandwidth, while in-line phase-shifting interferometry preserves bandwidth yet requires reference-path stepping and is sensitive to drift, limiting dynamic measurements. Moreover, viscoelastic mapping is rarely available from the same holographic measurement. We propose vibration-encoded in-line Mach-Zehnder holography for simultaneous thickness and viscoelasticity mapping of soft samples. Twelve holograms acquired over one vibration cycle are analyzed using Bessel-based harmonic inversion and robust regression to recover the static phase, modulation depth, and phase lag, yielding thickness and Kelvin-Voigt storage and loss modulus maps ( E ′, E ″). Simulations recover E ′ and E ′′ to within ∼2% across a wide E ′′/ E ′ range and achieve sub-micron thickness error over 20-45 μ m beads. Experiments on calibrated polyacrylamide beads show sub-micron thickness repeatability (median ∼0.57 μ m over 40 repeats) and stiffness estimates typically within 10% of ground truth, and we further demonstrate the approach on adherent MCF-7 cells.

Despite good axial resolution in holographic microscopy, lateral resolution remains constrained by optical limitations, pixel size, and noise. These restrictions hinder accurate reconstruction of fine structural and phase details. To address this, we evaluate three deep-learning super-resolution models (RCAN, SwinIR, and a conditional diffusion network) on 1,440 off-axis digital holograms of microbeads downsampled by 2 ×, 3 ×, and 4 ×. We compare their performance to bicubic spline interpolation using PSNR, SSIM, MSE, and phase-derived depth errors. RCAN and SwinIR yield the most accurate reconstructions, preserving structural and quantitative phase information, and offering guidance on model selection for phase-focused holography.

Significance: Measuring cell stiffness is essential in cellular biomechanics, particularly in understanding disease progression, including cancer metastasis and tissue mechanics. However, conventional techniques such as atomic force microscopy and optical stretching present limitations, including invasiveness, low throughput, and complex sample preparation. These factors restrict their applicability in dynamic and sensitive biological environments.Aim: This study introduces a noninvasive holographic sensor system for evaluating the stiffness of soft microscale samples.Approach: The proposed system integrates holographic imaging with acoustic stimulation using an off-axis Mach–Zehnder interferometer combined with bulk acoustic waves. This setup allows for label-free, high-throughput measurements while preserving sample integrity. The system was validated with polyacrylamide beads engineered to mimic cellular stiffness, ensuring precise and repeatable stiffness assessments.Results: Measurement errors caused by spatial variations were minimized through a structured imaging approach and a calibration strategy, improving uniformity across different regions. These corrections enhanced the consistency and reliability of stiffness assessments. Experimental validation demonstrated stable stiffness measurements regardless of sample size variations. Repeatability tests further confirmed the system’s robustness, producing consistent results across multiple trials.Conclusion: The findings highlight the potential of this holographic sensor system in advancing cell biomechanics research, cancer diagnostics, and mechanobiology. By offering a noninvasive, high-throughput alternative for mechanical property assessments in biological samples, this method contributes to improved characterization of cellular stiffness in biomedical applications.

In the textile industry, passive radio frequency identification (RFID) tag integration is increasing from warehouse tracking to production follow-up and healthcare applications. For instance, laundry management in hospitals is a complicated and massive system that should work without errors for patient safety. The importance of the proposed work is majorly on introducing the in silico design of a 30x30 mm2 textile-based chipless RFID (CRFID) tag with circular transmission lines using a conductive yarn for item identification using the CST Microwave Studio application. The tag antenna is designed with stainless steel/polyester twisted yarn integrated into the polypropylene tag fabric using the embroidery technique. The antenna is tested for its readability of the embedded code in the simulations with a range of 0-20 GHz frequency band, and its Radar Cross Section (RCS) responses, return losses (S1,1), E-field, H-field, and Farfield performances are obtained. As a result of these, 9 bits code capacity and 1-bit/cm2 code density are achieved. The UHF antenna design simulation results demonstrate the feasibility of using textile-based materials to create a CRFID tag that is lightweight, durable, and cost-effective. The tag can store and transmit data without a chip, making it ideal for healthcare applications. This research contributes to the advancement of CRFID technology and provides a practical solution for laundry applications. The washable and durable circular textile-based CRFID tag helps to track and monitor laundry data consistently and can improve the healthcare laundry industry. Furthermore, it can be used in various textile applications for tracking, asset management, and wearable and washable textiles.

Although, three-dimensional profile of a microscopic object can be obtained by using digital holography methods, they can be corrupted by speckle noise. We propose a method, based on deep learning, to suppress the noise.

Holographic three-dimensional television systems provide a natural 3D visualization. Fast calculation of the diffraction field from a three-dimensional object is essential to achieve video rate. In the literature, there are myriads of fast algorithms for diffraction field calculation from three-dimensional objects, but most of them omit the pixelated structure of the dynamic display devices which are used in the reconstruction process. In this chapter, the look-up table-based fast algorithm for diffraction field calculation from a three-dimensional object for a pixelated dynamic display device is presented. Real-time diffraction field calculations are obtained by running the algorithm in parallel on a graphical processing unit. Performance of the algorithm is evaluated in terms of computation time of the diffraction field and the normalized mean square error on the reconstructed object. To have optimization on the required memory space for the look-up table, two different sampling policies along the longitudinal axis are implemented. Uniform sampling policy along the longitudinal axis provides better error performance than nonuniform sampling policy. Furthermore, optical experiments are performed, and it is observed that both numerical and optical reconstructions are similar to each other. Hence, the proposed method provides successful results.