The University of Texas System

Tufts University

Aaron Benjamin Greenblatt

Sos S Agaian

Karen A Panetta

Clara M Mosquera-Lopez

A method and system for creating hypercomplex representations of data includes, in one exemplary embodiment, at least one set of training data with associated labels or desired response values, transforming the data and labels into hypercomplex values, methods for defining hypercomplex graphs of functions, training algorithms to minimize the cost of an error function over the parameters in the graph, and methods for reading hierarchical data representations from the resulting graph. Another exemplary embodiment learns hierarchical representations from unlabeled data. The method and system, in another exemplary embodiment, may be employed for biometric identity verification by combining multimodal data collected using many sensors, including, data, for example, such as anatomical characteristics, behavioral characteristics, demographic indicators, artificial characteristics. In other exemplary embodiments, the system and method may learn hypercomplex function approximations in one environment and transfer the learning to other target environments. Other exemplary applications of the hypercomplex deep learning framework include: image segmentation; image quality evaluation; image steganalysis; face recognition; event embedding in natural language processing; machine translation between languages; object recognition; medical applications such as breast cancer mass classification; multispectral imaging; audio processing; color image filtering; and clothing identification.

Hypercomplex deep learning methods, architectures, and apparatus for multimodal small, medium, and large-scale data representation, analysis, and applications

Described herein are systems and methods for performing multi-stage detection and classification of cancer regions from digitized images of biopsy slides. Novel methods for processing the digitized images to improve feature extraction and structure identification are disclosed, including but not limited to the use of quaternions, logarithmic mappings of color channels, and application of wavelets to logarithmic color channel mappings. The extracted features are utilized in improved machine learning algorithms that are further optimized to analyze multiple color channels in multiple dimensions. The improved machine learning algorithms include techniques for accelerating the training of the algorithms, making their application to biopsy detection and classification practical for the first time. The performance of the described systems and methods are further improved by the disclosure of a novel multistage machine learning scheme, in which additional classifiers are utilized to choose among the classes proposed by other classifiers in close cases.

Systems and methods for quantitative analysis of histopathology images using multiclassifier ensemble schemes

Methods and apparatus for restoration of a digital image. In one embodiment, a method for repairing a defect in a digital image to provide a restored image comprises determining a plurality of pixel locations to form a neighborhood relating to the defect and whether or not the neighborhood has a well-defined, dark border along its edge. Should the neighborhood not have dark border, one embodiment of the method entails processing the neighborhood to bring the neighborhood approximately to uniform darkness, processing the neighborhood to match surroundings in the digital image, copying an edge of a neighborhood in the digital image into the processed neighborhood, processing pixels of the edge to repair the copied edge pixels, and outputting the restored image for display to a user. Should the neighborhood have a dark, well-defined border, one embodiment of the method entails processing the neighborhood as to locally enhance the neighborhood and match its surroundings in the digital image; processing the neighborhood's edge such that the edge also matches its surroundings in both the defect and the digital image; processing the neighborhood to invert its pixel values and then perform the last two steps once again; processing the neighborhood to increase its contrast and then perform the last three steps once again; processing the neighborhood to bring the neighborhood to a more uniform darkness; processing the uniform darkness neighborhood to match surroundings in the digital image; and outputting the restored image for display to a user.

Growing community of inventors

Stoughton, MA, United States of America

Aaron Benjamin Greenblatt