Abstract: A method of generating virtually-stained image annotations. The method includes providing a neural network executed by image processing software. The image processing software runs on a processor of a computing device. The method includes training the neural network with a plurality of chemically stained patterns of endogenous signals to identify virtual staining patterns. The method includes producing and annotating an image of a biological sample for a biomarker. The biological sample includes endogenous signals. The method includes identifying, using the trained neural network, the virtual staining patterns in the image of the biological sample. Lastly, the method includes overlaying the virtual staining patterns in the image of the biological sample with annotations using spatial matching to produce virtually-stained image annotations.

Abstract: An image reconstruction algorithm system for hazardous source mapping. The algorithm system can be used to automate and optimize the search path of a movable vehicle (such as a UAV), equipped with detection capability. The algorithm allows the vehicle to localize hazardous sources in multiple scenarios effectively. Hazard mapping is formulated as an inverse problem and solved either with a deconvolution or a reconstruction algorithm, according to the problem complexity. The algorithms can use the Maximum a Posteriori (MAP) and the least square regression algorithm, respectively. However, alternative algorithms can be used as set forth herein. The source mapping algorithms are able to provide a quantitative estimation of the hazard source magnitude. A non-negative version of the least square algorithm is used to reconstruct the map at each step of the navigation algorithm of the vehicle. The navigation algorithm correctly located single and multiples simulated hazardous sources.

Abstract: A method for performing color image reconstruction of a single super-resolved holographic sample image includes obtaining a plurality of sub-pixel shifted lower resolution hologram images of the sample using an image sensor by simultaneous illumination at multiple color channels. Super-resolved hologram intensity images for each color channel are digitally generated based on the lower resolution hologram images. The super-resolved hologram intensity images for each color channel are back propagated to an object plane with image processing software to generate a real and imaginary input images of the sample for each color channel. A trained deep neural network is provided and is executed by image processing software using one or more processors of a computing device and configured to receive the real input image and the imaginary input image of the sample for each color channel and generate a color output image of the sample.