Abstract: Methods of dividing a neural network into chunks of operations executable in a hardware pass of hardware to execute a neural network. The layers of the neural network are divisible into layer groups that comprise a sequence of layers executable in the same hardware pass of the hardware. Each layer group is divisible into chunks of operations executable in a hardware pass of the hardware. The chunks for a layer group are defined by split parameters. A layer group loss function is obtained that represents a performance metric associated with executing a layer group on the hardware as a function of the split parameters and neural network architecture parameters for the layer group.

Abstract: A computer-implemented method of object enhancement in endoscopy images includes capturing an image of an object within a surgical operative site via an imaging device, determining a size of the object based on the captured image of the object, displaying the captured image of the object, and displaying on the displayed captured image of the object a representation of the determined size of the object.

Abstract: Techniques are disclosed for allowing remote participation in collaborative video review based on joint state data, for a video collaboration session, maintained by a video collaboration service. A user at a participant client device may provide one or more annotations, such as a drawing annotation, for the video data via a client application. The client application transmits computer-readable instructions for re-creating the drawing annotation to the service, which distributes the drawing instructions to the other participant client devices. Using the drawing instructions, the client applications at the client devices are configured to re-create the drawing annotation on the associated video frame displayed at the client devices. The joint state data communicated to client devices for a given session may include co-presence data that efficiently increases communication among the participants in a session.

Abstract: A knuckle and support structure for a reverse buckling rupture disc is described.

Abstract: A system and method for a fluid sampling tool. The fluid sampling tool may include a probe section. The probe section may include one or more probes, one or more stabilizers, and a housing that houses a bi directional piston pump. The method may include disposing a fluid sampling tool into a wellbore at a first depth, pressing the one or more probes into a surface of the wellbore, drawing a reservoir fluid from the wellbore through the one or more probes, placing the reservoir fluid into the housing, isolating the housing from the one or more modules of the fluid sampling tool with one or more shut in valves, depressurizing the housing with the bi directional piston pump, and measuring the asphaltene precipitation of the reservoir fluid within the housing.

Abstract: A door sill protector includes a base rotatably mounted to a door sill via a hinge. The hinge allows the base to be rotated between a stowed position and a deployed position. A plurality of support feet are disposed on an underside of the base. Each of the support feet is pivotable for pivoting between an extended orientation and a retracted orientation. A linkage bar is mechanically coupled to the support feet such that the support feet collectively pivot in unison between the extended orientation and the retracted orientation.

Abstract: A doorway side guide system includes a base mechanically coupled with a bottom portion of a doorway. A mounting member is rotatably coupled to the base via a hinge for rotating between a deployed position and a stowed position. A contact member is mechanically coupled with the mounting member via a pivot such that the contact member is configured to rotate about the pivot. The contact member extends outside of the doorway in the deployed position for guiding cargo through the doorway and for protecting a doorframe of the doorway.

Abstract: A method and system for determining a bed permeability. As disclosed, a fluid sampling tool may be disposed into a wellbore at a first location. The method may further include taking a drawdown and build up measurement with the fluid sampling tool, measuring a relative dip angle from the fluid sampling tool, calculating a bed anisotropy from the drawdown and build up measurement and the relative dip angle, calculating a bed mobility from the bed anisotropy, and calculating a bed permeability from the bed mobility and a viscosity.

Abstract: Systems and methods are provided for monitoring a user having a chronic gastrointestinal indication. A questionnaire regarding a plurality of conditions is provided on a repeating basis. Each such condition at least partly arises from the gastrointestinal indication. Questionnaire questions are each associated with a corresponding condition and provide an affordance that allows the user to select between low and high values in accordance with user association with the condition. Responses are stored in a data store associated with the user. A user requested report comprising a graphical quality of life measure of the user is provided based upon temporal questionnaire answers. Questionnaire information is communicated to a remote device for medical practitioner evaluation and computation of a temporal overall quality of life score based on a plurality of component quality of life scores, each of which is associated with a condition in the plurality of conditions.

Abstract: A computer-implemented method of selecting a number format for use in configuring a hardware implementation of a bidirectional recurrent neural network (BRNN) for operation on a sequence of inputs. A received BRNN representation is implemented as a test neural network equivalent to the BRNN over a sequence of input tensors, each step of the test neural network being for operation on (a) an input tensor of the sequence, (b) a corresponding backward state tensor generated in respect of a subsequent input tensor of the sequence, and (c) a corresponding forward state tensor generated in respect of a preceding input tensor of the sequence. The test neural network includes a forward recurrent neural network (RNN) for operation on the forward state tensors over the input tensors of the sequence; and a backward RNN for operation on the backward state tensors over the input tensors of the sequence.

Abstract: A method of implementing in hardware a bidirectional recurrent neural network (BRNN) for operation on a sequence of inputs, each step of the BRNN being for operation on (a) an input of the sequence, (b) corresponding backward state generated in respect of a subsequent input of the sequence, and (c) corresponding forward state generated in respect of a preceding input of the sequence. A representation of the BRNN is transformed into a derivative neural network equivalent to the BRNN over the sequence of inputs. The derivative neural network includes a forward recurrent neural network (RNN) for operation on the forward state over the inputs of the sequence, and a backward recurrent neural network (RNN) for operation on the backward state over the inputs of the sequence. The derivative neural network is implemented in hardware so as to perform the BRNN on the sequence of inputs.

Abstract: A targeted radiopharmaceutical of chemical Formula I, below, is disclosed wherein Q+3 is a trivalent radioactive isotope ion; M is a proton (H+), an ammonium ion or an alkali metal ion; “g” is a number that is 1 to about 12; the boxed mAb MNPR-101 represents the chemically-bonded humanized mAb MNPR-101; and Y? is an optional anion present in an amount needed to balance the ionic charge. A pharmaceutical composition that comprises a theranostic effective amount of a Formula I targeted radiopharmaceutical dissolved or dispersed in a pharmaceutically acceptable diluent is also disclosed, as are a method for treating and/or diagnosing a mammalian host having a disease, disorder or condition characterized by undesired angiogenesis, tumor growth and/or tumor metastasis. A targeted pro-radiopharmaceutical construct similar to that of Formula I but without the radioisotope (Formula III) is also contemplated.

Abstract: A targeted radiopharmaceutical comprising a targeting species chemically-bonded to a PCTA-chelated Q+3 trivalent radioactive ion of Formula I is disclosed. Six of R1 through R7 are H and the seventh is a reacted functionality, Z, that forms the chemical bond with the targeting species, T. “g” is a number whose average value is 1 to about 12. X1, X2, and X3, are substituent groups that can coordinate to the Q+3 ion and/or help neutralize the ionic charge. Anion Y? is optionally present to balance the ionic charge. A pharmaceutical composition comprising a theranostic effective amount of a targeted radiopharmaceutical of Formula I in a pharmaceutically acceptable diluent is also contemplated, as are a method for treating and/or diagnosing a mammalian host having a disease, disorder or condition characterized by undesired angiogenesis, tumor growth and/or tumor metastasis.

Abstract: A system and method for a fluid sampling tool. The fluid sampling tool may include a probe section. The probe section may include one or more probes, one or more stabilizers, and a housing that houses a bi directional piston pump. The method may include disposing a fluid sampling tool into a wellbore at a first depth, pressing the one or more probes into a surface of the wellbore, drawing a reservoir fluid from the wellbore through the one or more probes, placing the reservoir fluid into the housing, isolating the housing from the one or more modules of the fluid sampling tool with one or more shut in valves, depressurizing the housing with the bi directional piston pump, and measuring the asphaltene precipitation of the reservoir fluid within the housing.

Abstract: An exemplary lighting system utilizes intelligent system elements, such as lighting devices, user interfaces for lighting control or the like and possibly sensors, and utilizes network communication amongst such intelligent system elements. Some processing functions performed within the system are implemented on a distributed processing basis, by two or more of the intelligent elements of the lighting system. Distributed processing, for example, may enable use of available processor and/or memory resources of a number of intelligent system elements to process a particular job. Another distributed processing approach might entail programming to configure two or more of the intelligent system elements to implement multiple instances of a server functionality with respect to client functionalities implemented on intelligent system elements.

Abstract: Systems and methods are provided for identifying one or more health care applications in which a search request is received from a user at a remote client device. The request comprises an alphanumeric query and filtering criteria. A plurality of applications is searched to identify applications that satisfy the filtering criteria and further match the alphanumeric query thereby identifying a set of matching applications. Each respective application in the plurality of applications is (i) for a clinical indication and (ii) includes an evidence score generated by health care providers. The search query response is formatted for display by sorting the matching applications by sorting criteria thereby forming a sorted list of matching applications. At least a portion of the sorting criteria is the evidence score. The search query response, including the sorted list of matching applications, is communicated to the remote client device.

Abstract: A lighting system utilizes intelligent system elements, such as lighting devices, user interfaces for lighting control or the like and possibly sensors. The system also has a data communication network. Some number of the intelligent lighting system elements, including at least two of the lighting devices, also support wireless communication with non-lighting-system devices at the premises. Each such element has a communication interface system configured to provide a short range, low power wireless data communication link for use by non-lighting-system devices at the premises in proximity to the respective intelligent system element. Also, in such an element, the processor is configured to control communications via the communication interface system so as to provide access to the data network and through the data network to the wide area network outside the premises for non-lighting related communications of the non-lighting-system devices.

Abstract: Implants or systems of implants and methods apply a selected force vector or a selected combination of force vectors within or across the right atrium, which allow tricuspid valve leaflets to better coapt. The implants or systems of implants and methods make possible rapid deployment, facile endovascular delivery, and full intra-atrial retrievability. The implants or systems of implants and methods also make use of strong fluoroscopic landmarks. The implants or systems of implants and methods make use of an adjustable implant. The implants or systems of implants and methods may utilize a bridge stop to secure the implant.

Abstract: Implants or systems of implants and methods apply a selected force vector or a selected combination of force vectors within or across the right atrium, which allow tricuspid valve leaflets to better coapt. The implants or systems of implants and methods make possible rapid deployment, facile endovascular delivery, and full intra-atrial retrievability. The implants or systems of implants and methods also make use of strong fluoroscopic landmarks. The implants or systems of implants and methods make use of an adjustable implant. The implants or systems of implants and methods may utilize a bridge stop to secure the implant.

Abstract: Water-dispersible pellets are provided that includes spent lime particles present from 80 to 99.5 total weight percent. A binder component is also present from 20 to 0.5 total weight percent. The pellets have a mean pellet domain size. The spent lime particles, and the binder component present in a form such that contact with water causes pellet dispersion into more than 100 pieces. A process of forming spent lime pellets is also provided that include spent lime being dewatered to a free-flowing powder. A binder is then added under conditions to induce agglomeration of the free-flowing powder to the form the pellets. A process of treating a target crop is also provided based on such pellets.