Abstract: Systems and methods for chest condition determination can leverage one or more machine-learned models to process radiograph data to determine risk data (e.g., a preliminary diagnosis). For example, systems and methods can utilize a pathology model to process a chest x-ray to generate a tuberculosis diagnosis. The one or more machine-learned models can segment the lungs, can detect features in the data, and can pool the segmentation and located features to determine the diagnosis.

Abstract: The present disclosure provides systems and methods for training and/or employing machine-learned models (e.g., artificial neural networks) to diagnose chest conditions such as, as examples, pneumothorax, opacity, nodules or masses, and/or fractures based on chest radiographs. For example, one or more machine-learned models can receive and process a chest radiograph to generate an output. The output can indicate, for each of one or more chest conditions, whether the chest radiograph depicts the chest conditions (e.g., with some measure of confidence). The output of the machine-learned models can be provided to a medical professional and/or patient for use in providing treatment to the patient (e.g., to treat a detected condition).

Abstract: Endorsement abuse detection via social interactions is described, including generating an endorsement log comprising an endorser identifier, a target identifier, endorsement data, and a token indicative of an endorsement event; analyzing the endorsement log with respect to a user aggregator and a target aggregator, to generate an online endorsement abuse candidate; generating an endorsement overlap graph for the online endorsement abuse candidate based on the endorsement log, the user aggregator and the target aggregator, wherein nodes of the endorsement overlap graph represent ones of the online endorsement abuse candidate, and edges represent a number of shared endorsements; and determining whether the endorsement event is authorized based an analysis of the endorsement overlap graph and an orthogonal signal at least one of transmitted and received by the ones of the online endorsement abuse candidate.

Abstract: Endorsement abuse detection via social interactions is described, including generating an endorsement log comprising an endorser identifier, a target identifier, endorsement data, and a token indicative of an endorsement event; analyzing the endorsement log with respect to a user aggregator and a target aggregator, to generate an online endorsement abuse candidate; generating an endorsement overlap graph for the online endorsement abuse candidate based on the endorsement log, the user aggregator and the target aggregator, wherein nodes of the endorsement overlap graph represent ones of the online endorsement abuse candidate, and edges represent a number of shared endorsements; and determining whether the endorsement event is authorized based an analysis of the endorsement overlap graph and an orthogonal signal at least one of transmitted and received by the ones of the online endorsement abuse candidate.