Abstract: Among others, the present invention provides methods and compositions for controlling blood glucose or weight, including a combination of (i) a therapeutically effective amount of berberine (BBR), or a pharmaceutically acceptable salt, acid, ester, analog or derivative thereof, including dihydroberberine (DHB); and (ii) vitamin B12. The combination effectively ameliorates BBR or DHB side effects, including VB12 reduction, gastrointestinal discomfort, and slowed bowel movements, by promoting transcobalamin II receptors synthesis, decreasing homocysteine levels and maintaining intestinal balance, including short chain fatty acids content, gut microbiota diversity and richness.

Abstract: A robotic loading/unloading system is disclosed. In various embodiments, sensor data is received via a communication interface. The sensor data is used to determine a position and orientation of an extendable conveyor relative to a robotic loader comprising one or more robotic arms mounted on a robotically controlled rover. The determined position and orientation of the extendable conveyor relative to the robotic loader are used to control one or both of the extendable conveyor and the robotic loader to place the extendable conveyor and robotic loader to position a distal end of the extendable conveyor within reach of the one or more robotic arms at a location within a work area from which one or more pick or placement locations within the work area are within reach of at least one of the one or more robotic arms.

Abstract: Described are vapor deposition methods for depositing metal films or layers onto a substrate, wherein the metal is molybdenum or tungsten; the methods involve organometallic precursor compounds that contain the metal and one or more carbon-containing ligands, and include depositing a metal layer formed from the metal of the precursor, onto a substrate, followed by introducing oxidizer to the formed metal layer.

Abstract: Methods, systems, and techniques for automatically assessing credit risk of a counterparty are disclosed. A credit risk assessment method comprises: quantifying a relevance of news topics to one or more credit risk keywords; determining one or more relevant news topics that are relevant to the one or more credit risk keywords; obtaining a plurality of news articles each tagged with one or more news topics and having an associated sentiment; determining a relevant news article as a news article tagged with the one or more relevant news topics; and assessing a credit risk of a counterparty based at least in part on the sentiment of the relevant news article.

Abstract: The present disclosure pertains to compositions comprising anti-VEGF proteins and methods for producing such compositions.

Abstract: A computer vision processor of a camera generates hyperzooms for persons or vehicles from image frames captured by the camera. The hyperzooms include a first hyperzoom associated with the persons or vehicles. The computer vision processor tracks traffic patterns of the persons or vehicles while obviating network usage by the camera by predicting positions of the persons or vehicles using a Kalman Filter from the first hyperzoom. The persons or vehicles are detected in the second hyperzoom. The positions of the persons or vehicles are updated based on detecting the persons or vehicles in the second hyperzoom. The first hyperzoom is removed from the camera. Tracks of the persons or vehicles are generated based on the updated positions. The second hyperzoom is removed from the camera. Track metadata is generated from the tracks for storing in a key-value database located on a non-transitory computer-readable storage medium of the camera.

Abstract: A system and method of predicting a medical diagnosis is disclosed. The method includes receiving claims data, clinical data and demographic data and detecting a prediction target for the diagnosis. If present, inputting the prediction indicator and the clinical data into a machine learning model to predict diagnosis risk, to create a diagnosis risk score; determining a care seeking propensity score, from the demographic data; weighting the diagnosis risk score by the care seeking propensity score to create a weighted diagnosis risk score; determining whether the weighted diagnosis risk score indicates a likelihood of the medical diagnosis; and, in response, transmitting a recommendation for further evaluation. The machine learning model may be trained using historical claims data, clinical data, and demographic data and may be trained to detect correlation between medical diagnosis signals identified from the training data, and a positive result from a screening mechanism the medical diagnosis.

Abstract: A computer vision processor of a camera extracts attributes of persons or vehicles from hyperzooms generated from image frames. The hyperzooms represent traffic patterns. The extracting is performed using a feature extractor of an on-camera convolutional neural network (CNN) including an inverted residual structure. The attributes include at least colors of clothing of the persons or colors of the vehicles. Mobile semantic segmentation models of the CNN are generated using the hyperzooms and the attributes. Attribute analytics are generated by executing the mobile semantic segmentation models while obviating network usage by the camera. The attribute analytics are stored in a key-value database located on a memory card of the camera. A query is received from the server instance specifying one or more of the attributes. The attribute analytics are filtered using the one or more of the attributes to obtain a portion of the traffic patterns.

Abstract: A computer vision processor of a camera generates hyperzooms for persons or vehicles from image frames captured by the camera. The hyperzooms include a first hyperzoom associated with the persons or vehicles. The computer vision processor tracks traffic patterns of the persons or vehicles while obviating network usage by the camera by predicting positions of the persons or vehicles using a Kalman Filter from the first hyperzoom. The persons or vehicles are detected in the second hyperzoom. The positions of the persons or vehicles are updated based on detecting the persons or vehicles in the second hyperzoom. The first hyperzoom is removed from the camera. Tracks of the persons or vehicles are generated based on the updated positions. The second hyperzoom is removed from the camera. Track metadata is generated from the tracks for storing in a key-value database located on a non-transitory computer-readable storage medium of the camera.

Abstract: A computer vision processor of a camera generates hyperzooms for persons or vehicles from image frames captured by the camera. The hyperzooms include a first hyperzoom associated with the persons or vehicles. The computer vision processor tracks traffic patterns of the persons or vehicles while obviating network usage by the camera by predicting positions of the persons or vehicles using a Kalman Filter from the first hyperzoom. The persons or vehicles are detected in the second hyperzoom. The positions of the persons or vehicles are updated based on detecting the persons or vehicles in the second hyperzoom. The first hyperzoom is removed from the camera. Tracks of the persons or vehicles are generated based on the updated positions. The second hyperzoom is removed from the camera. Track metadata is generated from the tracks for storing in a key-value database located on a non-transitory computer-readable storage medium of the camera.

Abstract: A computer vision processor of a camera extracts attributes of persons or vehicles from hyperzooms generated from image frames. The hyperzooms represent traffic patterns. The extracting is performed using a feature extractor of an on-camera convolutional neural network (CNN) including an inverted residual structure. The attributes include at least colors of clothing of the persons or colors of the vehicles. Mobile semantic segmentation models of the CNN are generated using the hyperzooms and the attributes. Attribute analytics are generated by executing the mobile semantic segmentation models while obviating network usage by the camera. The attribute analytics are stored in a key-value database located on a memory card of the camera. A query is received from the server instance specifying one or more of the attributes. The attribute analytics are filtered using the one or more of the attributes to obtain a portion of the traffic patterns.

Abstract: A computer vision processor of a camera generates hyperzooms for persons or vehicles from image frames captured by the camera. The hyperzooms include a first hyperzoom associated with the persons or vehicles. The computer vision processor tracks traffic patterns of the persons or vehicles while obviating network usage by the camera by predicting positions of the persons or vehicles using a Kalman Filter from the first hyperzoom. The persons or vehicles are detected in the second hyperzoom. The positions of the persons or vehicles are updated based on detecting the persons or vehicles in the second hyperzoom. The first hyperzoom is removed from the camera. Tracks of the persons or vehicles are generated based on the updated positions. The second hyperzoom is removed from the camera. Track metadata is generated from the tracks for storing in a key-value database located on a non-transitory computer-readable storage medium of the camera.

Abstract: A computer vision processor of a camera extracts attributes of persons or vehicles from hyperzooms generated from image frames. The hyperzooms represent traffic patterns. The extracting is performed using a feature extractor of an on-camera convolutional neural network (CNN) including an inverted residual structure. The attributes include at least colors of clothing of the persons or colors of the vehicles. Mobile semantic segmentation models of the CNN are generated using the hyperzooms and the attributes. Attribute analytics are generated by executing the mobile semantic segmentation models while obviating network usage by the camera. The attribute analytics are stored in a key-value database located on a memory card of the camera. A query is received from the server instance specifying one or more of the attributes. The attribute analytics are filtered using the one or more of the attributes to obtain a portion of the traffic patterns.

Abstract: A computer vision processor of a camera generates hyperzooms for persons or vehicles from image frames captured by the camera. The hyperzooms include a first hyperzoom associated with the persons or vehicles. The computer vision processor tracks traffic patterns of the persons or vehicles while obviating network usage by the camera by predicting positions of the persons or vehicles using a Kalman Filter from the first hyperzoom. The persons or vehicles are detected in the second hyperzoom. The positions of the persons or vehicles are updated based on detecting the persons or vehicles in the second hyperzoom. The first hyperzoom is removed from the camera. Tracks of the persons or vehicles are generated based on the updated positions. The second hyperzoom is removed from the camera. Track metadata is generated from the tracks for storing in a key-value database located on a non-transitory computer-readable storage medium of the camera.

Abstract: A computer vision processor of a camera extracts attributes of persons or vehicles from hyperzooms generated from image frames. The hyperzooms represent traffic patterns. The extracting is performed using a feature extractor of an on-camera convolutional neural network (CNN) including an inverted residual structure. The attributes include at least colors of clothing of the persons or colors of the vehicles. Mobile semantic segmentation models of the CNN are generated using the hyperzooms and the attributes. Attribute analytics are generated by executing the mobile semantic segmentation models while obviating network usage by the camera. The attribute analytics are stored in a key-value database located on a memory card of the camera. A query is received from the server instance specifying one or more of the attributes. The attribute analytics are filtered using the one or more of the attributes to obtain a portion of the traffic patterns.

Abstract: A computer vision processor of a camera extracts attributes of persons or vehicles from hyperzooms generated from image frames. The hyperzooms represent traffic patterns. The extracting is performed using a feature extractor of an on-camera convolutional neural network (CNN) including an inverted residual structure. The attributes include at least colors of clothing of the persons or colors of the vehicles. Mobile semantic segmentation models of the CNN are generated using the hyperzooms and the attributes. Attribute analytics are generated by executing the mobile semantic segmentation models while obviating network usage by the camera. The attribute analytics are stored in a key-value database located on a memory card of the camera. A query is received from the server instance specifying one or more of the attributes. The attribute analytics are filtered using the one or more of the attributes to obtain a portion of the traffic patterns.

Abstract: This invention includes an autonomous driving system for automobiles, comprising: one or more common electronic communication ports of autonomous driving (communication ports) that are built-in on the automobiles; and one or more universal autonomous driving portable controllers (portable controllers) that are to be plugged-in to the said communication ports that are built-in on the automobiles. The interfaces of the communication ports and the portable controllers are both standardized such that the portable controllers can be plugged-in universally to all of the automobiles that are equipped with the built-in communication ports. The communication ports comprise electronic communication of all relevant electronic control units (ECUs) and feedback information of the automobiles, dedicated for the said portable controllers to communicate with and to control the automobiles.

Abstract: In one aspect there is a method for recommending one or more providers to a member. The method may comprise at a server, receiving, from a member computing device via a provider search interface, a provider request including a member identifier and member specified criteria. Retrieving one or more provider identifiers for association with the member specified criteria. Retrieving one or more member characteristics for association with the member identifier. For each provider identifier, retrieving one or more provider characteristics for association with the provider identifier. Analyzing the member characteristics and the provider characteristics to generate an ordered list of the providers, wherein the ordered list of the providers is based upon a relevancy rank of each provider that is specific to the member, wherein the relevancy rank is a function of at least one of: a provider cost factor, a provider quality factor, or a weighted combination thereof.

Abstract: A method for updating a vehicle reference speed for a vehicle is disclosed. The method includes increasing brake pressure to at least one wheel of the plurality of wheels and determining, via an electronic control unit, that a first wheel speed of the at least one wheel is less than a second wheel speed of another wheel of the plurality of wheels. The method also includes estimating the vehicle reference speed of the vehicle based upon the first wheel speed of the at least one wheel.

Abstract: Near normal incidence MUX/DEMUX designs may include an optical demultiplexer coupled to a photonics die, where the optical demultiplexer comprises an input fiber, thin film filters at a first surface of a substrate, a first mirror at the first surface of the substrate, and a second mirror at a second surface of the substrate. The optical demultiplexer may receive an input optical signal comprising a plurality of wavelength optical signals, reflect the input optical signal from the first mirror to the second mirror, reflect the input optical signal from the second mirror to a first of the thin film filters, communicate an optical signal at a first wavelength to the photonics die while reflecting others to the second mirror, reflect the other signals to a second of the plurality of thin film filters, and communicate an optical signal at a second wavelength to the photonics die.