Deep learning accelerates whole slide imaging for next-generation digital pathology applications

Digital staining facilitates biomedical microscopy

Traditional staining of biological specimens for microscopic imaging entails time-consuming, laborious, and costly procedures, in addition to producing inconsistent labeling and causing irreversible sample damage. In recent years, computational "virtual" staining using deep learning techniques has evolved into a robust and comprehensive application for streamlining the staining process without typical histochemical staining-related drawbacks. Such virtual staining techniques can also be combined with neural networks designed to correct various microscopy aberrations, such as out-of-focus or motion blur artifacts, and improve upon diffracted-limited resolution. Here, we highlight how such methods lead to a host of new opportunities that can significantly improve both sample preparation and imaging in biomedical microscopy.

Pre-Implantation Bovine Embryo Evaluation-From Optics to Omics and Beyond

Approximately 80% of the ~1.5 million bovine embryos transferred in 2021 were in vitro produced. However, only ~27% of the transferred IVP embryos will result in live births. The ~73% pregnancy failures are partly due to transferring poor-quality embryos, a result of erroneous stereomicroscopy-based morphological evaluation, the current method of choice for pre-transfer embryo evaluation. Numerous microscopic (e.g., differential interference contrast, electron, fluorescent, time-lapse, and artificial-intelligence-based microscopy) and non-microscopic (e.g., genomics, transcriptomics, epigenomics, proteomics, metabolomics, and nuclear magnetic resonance) methodologies have been tested to find an embryo evaluation technique that is superior to morphologic evaluation. Many of these research tools can accurately determine embryo quality/viability; however, most are invasive, expensive, laborious, technically sophisticated, and/or time-consuming, making them futile in the context of in-field embryo evaluation. However accurate they may be, using complex methods, such as RNA sequencing, SNP chips, mass spectrometry, and multiphoton microscopy, at thousands of embryo production/collection facilities is impractical. Therefore, future research is warranted to innovate field-friendly, simple benchtop tests using findings already available, particularly from omics-based research methodologies. Time-lapse monitoring and artificial-intelligence-based automated image analysis also have the potential for accurate embryo evaluation; however, further research is warranted to innovate economically feasible options for in-field applications.

Wide-Field and Real-Time Super-Resolution Optical Imaging By Titanium Dioxide Nanoparticle-Assembled Solid Immersion Lens

Super-resolution optical imaging techniques can break the optical diffraction limit, thus providing unique opportunities to visualize the microscopic world at the nanoscale. Although near-field optical microscopy techniques have been proven to achieve significantly improved imaging resolution, most near-field approaches still suffer from a narrow field of view (FOV) or difficulty in obtaining wide-field images in real time, which may limit their widespread and diverse applications. Here, the authors experimentally demonstrate an optical microscope magnification and image enhancement approach by using a submillimeter-sized solid immersion lens (SIL) assembled by densely-packed 15 nm TiO₂ nanoparticles through a silicone oil two-step dehydration method. This TiO₂ nanoparticle-assembled SIL can achieve both high transparency and high refractive index, as well as sufficient mechanical strength and easy-to-handle size, thus providing a fast, wide-field, real-time, non-destructive, and low-cost solution for improving the quality of optical microscopic observation of a variety of samples, including nanomaterials, cancer cells, and living cells or bacteria under conventional optical microscopes. This study provides an attractive alternative to simplify the fabrication and applications of high-performance SILs.