Single-shot digital holography in Gabor mode offers cost-effective quantitative phase imaging but suffers from the fundamental twin image problem, where real and conjugate images are inherently superimposed, severely limiting phase reconstruction accuracy. Traditional iterative phase retrieval methods require computationally expensive multiple propagations, while off-axis holography demands complex optical setups with precise alignment. We present the first unsupervised diffusion model for automated phase image reconstruction from single-shot in-line holograms, eliminating both twin image artifacts and the need for expensive off-axis configurations. Our framework integrates cycle-consistency and denoising modules to enable training on unpaired hologram-phase image datasets, learning the mapping between low-cost in-line measurements and high-quality phase distributions without requiring labeled data pairs. Comprehensive evaluation on diverse biological specimens demonstrates that our approach significantly outperforms conventional unsupervised methods, achieving superior Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) values for both red blood cells and cancer cells. Critically, the model maintains exceptional performance even with limited training data, consistently outperforming supervised learning approaches under data-constrained conditions. The framework exhibits remarkable generalization capabilities, successfully reconstructing phase images from holograms captured at different propagation distances and processing various cancer cell types not included in training data. This computational breakthrough enables accurate, scalable, and hardware-efficient quantitative phase imaging, democratizing access to high-quality phase microscopy for resource-constrained environments while maintaining reconstruction fidelity comparable to complex off-axis systems.

MOTIVATION: Skeletal muscle cells (skMCs) combine together to create long, multi-nucleated structures called myotubes. By studying the size, length, and number of nuclei in these myotubes, we can gain a deeper understanding of skeletal muscle development. However, human experimenters may often derive unreliable results owing to the unusual shape of the myotube, which causes significant measurement variability. RESULTS: We propose a new method for quantitative analysis of the dexamethasone side effect on human-derived young and aged skeletal muscle by simultaneous myotube and nuclei segmentation using deep learning combined with post-processing techniques. The deep learning model outputs myotube semantic segmentation, nuclei semantic segmentation, and nuclei center, and post-processing applies a watershed algorithm to accurately distinguish overlapped nuclei and identify myotube branches through skeletonization. To evaluate the performance of the model, the myotube diameter and the number of nuclei were calculated from the generated segmented images and compared with the results calculated by human experimenters. In particular, the proposed model produced outstanding outcomes when comparing human-derived primary young and aged skMCs treated with dexamethasone. The proposed standardized and consistent automated image segmentation system for myotubes is expected to help streamline the drug-development process for skeletal muscle diseases. AVAILABILITY AND IMPLEMENTATION: The code and the data are available at https://github.com/tdn02007/QA-skMCs-Seg.

This study presents a novel approach to automatically perform instant phenotypic assessment of red blood cell (RBC) storage lesion in phase images obtained by digital holographic microscopy. The proposed model combines a generative adversarial network (GAN) with marker-controlled watershed segmentation scheme. The GAN model performed RBC segmentations and classifications to develop ageing markers, and the watershed segmentation was used to completely separate overlapping RBCs. Our approach achieved good segmentation and classification accuracy with a Dice's coefficient of 0.94 at a high throughput rate of about 152 cells per second. These results were compared with other deep neural network architectures. Moreover, our image-based deep learning models recognized the morphological changes that occur in RBCs during storage. Our deep learning-based classification results were in good agreement with previous findings on the changes in RBC markers (dominant shapes) affected by storage duration. We believe that our image-based deep learning models can be useful for automated assessment of RBC quality, storage lesions for safe transfusions, and diagnosis of RBC-related diseases.