Pennsylvania Patent of the Month – February 2024
In the dynamic realm of artificial intelligence and machine learning, Proscia Inc has emerged as a trailblazer with their newly patented Critical Component Detection technology. The software company is focused on accelerating pathology with state of the art technology. A data-driven approach could remove human-based limitations and lead to accelerated discovery and improved patient outcomes.
This inventive method is designed to train electronic neural network classifiers, specifically focusing on identifying a particular property in novel supra-images while mitigating spurious correlations with extraneous properties.
At the core of this innovation is the utilization of a training corpus, consisting of a multitude of supra-images, each containing several images with various components. The method employs a second electronic neural network classifier, equipped with an attention layer, to identify the presence of an extraneous property. The attention layer assigns attention weights to each component, allowing the network to discern the significance of each element.
What sets Proscia’s invention apart is the meticulous process of identifying supra-image threshold attention weights through a binary search. This method ensures precision, as each component is associated with a threshold weight, enabling the differentiation between positive and negative classifications.
The attention to detail extends to the components themselves, which can be either a 128-pixel-by-128-pixel square portion of an image or a feature vector corresponding to a portion of an image. This flexibility enhances the adaptability of the neural network in various applications.
The applications of this invention span diverse fields, with the training corpus incorporating biopsy supra-images, and the particular property being dermatopathology-related. The innovation holds promise for detecting malignancies, specific grades of malignancy, or categories of risk within the realm of dermatopathology.
Proscia’s invention exemplifies the fusion of deep learning and attention mechanisms, offering a sophisticated solution to challenges in digital pathology. As we navigate the intersection of machine learning and medical diagnostics, this groundbreaking method propels us toward a future where accuracy and efficiency converge to redefine the landscape of pathology research and development.
Are you developing new technology for an existing application? Did you know your development work could be eligible for the R&D Tax Credit and you can receive up to 14% back on your expenses? Even if your development isn’t successful your work may still qualify for R&D credits (i.e. you don’t need to have a patent to qualify). To find out more, please contact a Swanson Reed R&D Specialist today or check out our free online eligibility test.
Who We Are:
Swanson Reed is one of the U.S.’ largest Specialist R&D tax advisory firms. We manage all facets of the R&D tax credit program, from claim preparation and audit compliance to claim disputes.
Swanson Reed regularly hosts free webinars and provides free IRS CE and CPE credits for CPAs. For more information please visit us at www.swansonreed.com/webinars or contact your usual Swanson Reed representative.