Abstract
Singlet lenses are free from precise assembling, aligning, and testing, which are helpful for the development of portable and low-cost microscopes. However, balancing the spectrum dispersion or chromatic aberrations using a singlet lens made of one material is difficult. Here, a novel method combining singlet lens microscopy and computational imaging, which is based on deep learning image-style-transfer algorithms, is proposed to overcome this problem in clinical pathological slide microscopy. In this manuscript, a singlet aspheric lens is used, which has a high cut-off frequency and linear signal properties. Enhanced by a trained deep learning network, it is easy to transfer the monochromatic gray-scale microscopy picture to a colorful microscopy picture, with only one single-shot recording by a monochromatic CMOS image sensor. By experiments, data analysis, and discussions, it is proved that our proposed virtual colorization microscope imaging method is effective for H&E stained tumor tissue slides in singlet microscopy. It is believable that the computational virtual colorization method for singlet microscopes would promote the low-cost and portable singlet microscopy development in medical pathological label staining observing (e.g., H&E staining, Gram staining, and fluorescent labeling) biomedical research.
Highlights
Commercial microscope objective lenses, cell-phone camera lenses, and other imaging lenses are cheap due to the mass industrial production
Singlet lenses are free from precise assembling, aligning, and testing, which are helpful for the development of portable and low-cost microscopes
Data analysis, and discussions, it is proved that our proposed virtual colorization microscope imaging method is effective for hematoxylin and eosin (H&E) stained tumor tissue slides in singlet microscopy
Summary
Scitation.org/journal/app eliminate spherical aberrations well, while it cannot overcome the off-axial aberrations at a large field of view (FOV). Colorfully imaging these metalenses and non-rotational symmetric freeform surfaces, made of one material, with single-shot recording is difficult In this manuscript, we propose a method, combining deep learning virtual colorization and designed singlet lens, to achieve large FOV singlet colorful microscopy. By computational image-style-transfer methods based on deep learning, we could achieve colorful microscopy with a singleshot digital recording. 1(c) and 1(a), the grayscale singlet microscopy image would be translated into a colorful image after convoluted by the trained virtual colorization deep neural network. Our singlet aspheric lens has a high cut-off frequency under a chromatic wavelength illumination, which is up to 350 lp/mm These good linear signal properties help greatly for the computational imaging algorithms. The deep training costs ∼20 h, while for practical usage, the virtually colorizing time is ∼7 ms
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have