Publications by authors named "Gong Rao"
Article Synopsis
- Computer-assisted diagnosis is essential for enhancing cervical cancer screening, but existing algorithms struggle with whole slide image analysis and vary in performance based on staining and imaging techniques.
- The authors created a new method that combines low- and high-resolution WSIs for better lesion cell recognition, alongside a recurrent neural network model to assess lesion severity.
- Their system has been trained on a large dataset and achieved high specificity (93.5%) and sensitivity (95.1%), outperforming average cytopathologist performance, and can analyze a one giga-pixel WSI in approximately 1.5 minutes after deployment.
Article Synopsis
- Researchers developed a weakly supervised learning method using a deep CNN to trace 3D neuronal structures from noisy optical microscopy images without needing extensive manual annotations.
- This approach uses existing automatic tracing methods to create pseudo-labels for the images and improves prediction through iterative training, focusing on weak neurites by analyzing their features.
- Testing on various public datasets showed that the method effectively identifies weak neuronal structures and performs comparably to fully supervised methods that require manual labeling.
Resin embedding is widely used and facilitates microscopic imaging of biological tissues. In contrast, quenching of fluorescence during embedding process hinders the application of resin embedding for imaging of fluorescence-labeled samples. For samples expressing fluorescent proteins, it has been demonstrated that the weakened fluorescence could be recovered by reactivating the fluorophore with alkaline buffer.
View Article and Find Full Text PDFHigh-resolution three-dimensional biomolecule distribution information of large samples is essential to understanding their biological structure and function. Here, we proposed a method combining large sample resin embedding with iDISCO immunofluorescence staining to acquire the profile of biomolecules with high spatial resolution. We evaluated the compatibility of plastic embedding with an iDISCO staining technique and found that the fluorophores and the neuronal fine structures could be well preserved in the Lowicryl HM20 resin, and that numerous antibodies and fluorescent tracers worked well upon Lowicryl HM20 resin embedding.
View Article and Find Full Text PDFResin embedding has been widely applied to fixing biological tissues for sectioning and imaging, but has long been regarded as incompatible with green fluorescent protein (GFP) labeled sample because it reduces fluorescence. Recently, it has been reported that resin-embedded GFP-labeled brain tissue can be imaged with high resolution. In this protocol, we describe an optimized protocol for resin embedding and chemical reactivation of fluorescent protein labeled mouse brain, we have used mice as experiment model, but the protocol should be applied to other species.
View Article and Find Full Text PDF