Holographic Reconstruction with Bright-field Microscopy Contrast using Cross-Modality Deep Learning

Author(s):  
Yilin Luo ◽  
Yichen Wu ◽  
Gunvant Chaudhari ◽  
Yair Rivenson ◽  
Ayfer Calis ◽  
...  
2021 ◽  
Vol 11 (6) ◽  
pp. 2692
Author(s):  
Danny Salem ◽  
Yifeng Li ◽  
Pengcheng Xi ◽  
Hilary Phenix ◽  
Miroslava Cuperlovic-Culf ◽  
...  

Accurate and efficient segmentation of live-cell images is critical in maximizing data extraction and knowledge generation from high-throughput biology experiments. Despite recent development of deep-learning tools for biomedical imaging applications, great demand for automated segmentation tools for high-resolution live-cell microscopy images remains in order to accelerate the analysis. YeastNet dramatically improves the performance of the non-trainable classic algorithm, and performs considerably better than the current state-of-the-art yeast-cell segmentation tools. We have designed and trained a U-Net convolutional network (named YeastNet) to conduct semantic segmentation on bright-field microscopy images and generate segmentation masks for cell labeling and tracking. YeastNet enables accurate automatic segmentation and tracking of yeast cells in biomedical applications. YeastNet is freely provided with model weights as a Python package on GitHub.


2020 ◽  
Author(s):  
Danny Salem ◽  
Yifeng Li ◽  
Pengcheng Xi ◽  
Hilary Phenix ◽  
Miroslava Cuperlovic-Culf ◽  
...  

Accurate and efficient segmentation of live-cell images is critical in maximising data extraction and knowledge generation from high-throughput biology experiments. Despite recent development of deep learning tools for biomedical imaging applications, great demand for automated segmentation tools for high-resolution live-cell microscopy images remains in order to accelerate the analysis. YeastNet dramatically improves the performance of non-trainable classic algorithm, and performs considerably better than the current state-of-the-art yeast cell segmentation tools. We have designed and trained a U-Net convolutional network (named YeastNet) to conduct semantic segmentation on bright-field microscopy images and generate segmentation masks for cell labelling and tracking. YeastNet enables accurate automatic segmentation and tracking of yeast cells in biomedical applications. YeastNet is freely provided with model weights as a Python package on GitHub. https://github.com/kaernlab/YeastNet


2021 ◽  
Vol 9 (8) ◽  
pp. 848
Author(s):  
Elise E. B. LaDouceur ◽  
Linda A. Kuhnz ◽  
Christina Biggs ◽  
Alicia Bitondo ◽  
Megan Olhasso ◽  
...  

Sea pigs (Scotoplanes spp.) are deep-sea dwelling sea cucumbers of the phylum Echinodermata, class Holothuroidea, and order Elasipodida. Few reports are available on the microscopic anatomy of these deep-sea animals. This study describes the histologic findings of two, wild, male and female Scotoplanes sp. collected from Monterey Bay, California. Microscopic findings were similar to other holothuroids, with a few notable exceptions. Sea pigs were bilaterally symmetrical with six pairs of greatly enlarged tube feet arising from the lateral body wall and oriented ventrally for walking. Neither a rete mirabile nor respiratory tree was identified, and the large tube feet may function in respiration. Dorsal papillae protrude from the bivium and are histologically similar to tube feet with a large, muscular water vascular canal in the center. There were 10 buccal tentacles, the epidermis of which was highly folded. Only a single gonad was present in each animal; both male and female had histologic evidence of active gametogenesis. In the male, a presumed protozoal cyst was identified in the aboral intestinal mucosa, and was histologically similar to previous reports of coccidians. This work provides control histology for future investigations of sea pigs and related animals using bright field microscopy.


2017 ◽  
pp. 307-344
Author(s):  
Onkar D. Dhingra ◽  
James B. Sinclair

2021 ◽  
Vol 44 (4) ◽  
Author(s):  
Fabio Giavazzi ◽  
Antara Pal ◽  
Roberto Cerbino

Abstract Soft and biological materials are often composed of elementary constituents exhibiting an incessant roto-translational motion at the microscopic scale. Tracking this motion with a bright-field microscope becomes increasingly challenging when the particle size becomes smaller than the microscope resolution, a case which is frequently encountered. Here we demonstrate squared-gradient differential dynamic microscopy (SG-DDM) as a tool to successfully use bright-field microscopy to extract the roto-translational dynamics of small anisotropic colloidal particles, whose rotational motion cannot be tracked accurately in direct space. We provide analytical justification and experimental demonstration of the method by successful application to an aqueous suspension of peanut-shaped particles. Graphic abstract


2004 ◽  
Vol 56 (3) ◽  
pp. 325-332 ◽  
Author(s):  
L.A. Barros ◽  
R. Tortelly ◽  
R.M. Pinto ◽  
D.C. Gomes

Rabbits were infected per os with 10 Eustrongylides ignotus L4 and with 50 Contracaecum multipapillatum L3 per rabbit, recovered from naturally infected freshwater fishes (Hoplias malabaricus) in order to evaluate the patogenicity of these two nematode species in mammalian host. Two rabbits (20%) infected with E. ignotus died before the fourth day post-inoculation (one after 51 and the other after 78 hours). Six rabbits (60%) were inappetent until the fifth day following experimental inoculation. No clinical signs in rabbits inoculated with C. multipapillatum were observed; nevertheless, eight (80%) animals were positive for this nematode species. Rabbits inoculated with E. ignotus, had gastric congestion with hematoma of the gastric wall in 60% of the cases. Peritoneum was congested in 20% of the animals with the presence of peritoneal abscess in 10% of the cases. All inoculated animals showed hyperemia of the gastric mucosa with hemorrhagic gastritis due to infections with E. ignotus. In C. multipapillatum inoculated animals, the hyperemia was followed by disruption of the epithelial mucosa in the sites of parasite attachment. In the gastric mucosa, miscellaneous leukocitary infiltrates, with multifocal necrosis reaching the submucosa in the infections with C. multipapillatum were observed under bright field microscopy. Perforating lesions in several organs, mainly in the gastric wall, pancreas and liver, always in the presence of a mixed inflammatory process, intensely fibrous, with hemorrhage and necrosis were observed in animals infected with E. ignotus.


2019 ◽  
Vol 9 (1) ◽  
Author(s):  
C. Martins ◽  
A. P. Rodrigo ◽  
L. Cabrita ◽  
P. Henriques ◽  
A. J. Parola ◽  
...  

Abstract True green pigments in the animal kingdom are scarce and are almost invariably porphyrinoids. Endogenous porphyrins resulting from the breakdown of haem are usually known as “bile pigments”. The pigmentation of intertidal Polychaeta has long gained attention due to its variety and vivid patterning that often seems incompatible with camouflage, as it occurs with Eulalia viridis, one of the few truly green Polychaeta. The present study combined UV and bright-field microscopy with HPLC to address the presence and distribution of pigments in several organs. The results showed two major types of porphyrin-like pigments, yellowish and greenish in colour, that are chiefly stored as intraplasmatic granules. Whereas the proboscis holds yellow pigments, the skin harbours both types in highly specialised cells. In their turn, oocytes and intestine have mostly green pigments. Despite some inter-individual variation, the pigments tend to be stable after prolonged storage at −20 °C, which has important implications for future studies. The results show that, in a foraging predator of the intertidal where melanins are circumscribed to lining the nervous system, porphyrinoid pigments have a key role in protection against UV light, in sensing and even as chemical defence against foulants and predators, which represents a remarkable adaptive feature.


Sign in / Sign up

Export Citation Format

Share Document