Abstract: Shapley values (SVs) have become an important tool to further the goal of explainability of machine learning (ML) models. However, the computational load of exact SV calculations increases exponentially with the number of attributes. Hence, the calculation of SVs for models incorporating large numbers of interpretable attributes is problematic. Molecular diagnostic tests typically seek to leverage information from hundreds or thousands of attributes, often using training sets with fewer instances. Methods are described for evaluate SVs using Monte Carlo sampling or exact calculation in polynomial time (i.e., reasonably quickly and efficiently) using the architecture of a ML model designed for robust molecular test generation, and without requiring classifier retraining.

Abstract: An accelerometer system comprising: a first magnetic assembly comprising a first pole piece and a first magnet; a second magnetic assembly comprising a second pole piece and a second magnet; a proof mass between the first magnetic assembly and the second magnetic assembly; a first coil disposed around the first pole piece; and a second coil disposed around the second pole piece; and processing circuitry configured to: cause a first current to flow through the first coil to apply a first Lorentz force to the proof mass; cause a second current to flow through the second coil to apply a second Lorentz force to the proof mass; and limit a maximum amplitude of the first current and the second current for at least a threshold period of time.

Abstract: A tissue anchor (200, 258, 300) includes a shaft (122), a tissue-coupling element (128), and a flexible elongate tension member (202). The tissue-coupling element (128) includes a wire (150), which is shaped as an open loop (154) having more than one turn when the tissue anchor (200, 258, 300) is unconstrained. The tension member (202) includes a distal portion (204) that is fixed to a site (206) on the open loop (154), a proximal portion (208), which has a longitudinal segment (209) that runs alongside at least a portion (210) of the shaft (122), and a crossing portion (212), which (i) is disposed between the distal and the proximal portions (204, 208) along the tension member (202), and (ii) crosses at least a portion of the open loop (154) when the tissue anchor (200, 258, 300) is unconstrained.

Abstract: A tissue anchor system is provided that includes a first tissue anchor, a second tissue anchor that is separate and distinct from the first tissue anchor, and one or more tethers, which are configured to couple the first tissue anchor to the second tissue anchor. When the first tissue anchor is unconstrained, a head thereof is coaxial with an axis of a shaft thereof, and a tissue-coupling element thereof extends from a distal end of the shaft, is generally orthogonal to the axis, and is shaped such that if the tissue-coupling element were to be projected onto a plane that is perpendicular to the axis, at least 80% of an area of a projection of the tissue-coupling element on the plane would fall within a first angle of 180 degrees in the plane having a vertex at the axis. Other embodiments are also described.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A tissue anchor system is provided that includes a first tissue anchor, a second tissue anchor that is separate and distinct from the first tissue anchor, and one or more tethers, which are configured to couple the first tissue anchor to the second tissue anchor. When the first tissue anchor is unconstrained, a head thereof is coaxial with an axis of a shaft thereof, and a tissue-coupling element thereof extends from a distal end of the shaft, is generally orthogonal to the axis, and is shaped such that if the tissue-coupling element were to be projected onto a plane that is perpendicular to the axis, at least 80% of an area of a projection of the tissue-coupling element on the plane would fall within a first angle of 180 degrees in the plane having a vertex at the axis. Other embodiments are also described.

Abstract: An emissions control device for treating or removing pollutants from an exhaust gas produced by an internal combustion engine is disclosed. The emissions control device comprises a solid foam layer disposed on a substrate, wherein the solid foam layer disposed on a substrate has an open cell structure and comprises a particulate material which is a catalytic material comprising a catalytically active metal supported on a support material.

Abstract: A method of coating a substrate with a foam is described. The method comprises: (a) introducing a foam into a substrate comprising a plurality of channels through open ends of the channels at a first end of the substrate; and (b) applying at least one of (i) a vacuum to open ends of the channels at a second end of the substrate and (ii) a pressure to the open ends of the channels at the first end of the substrate; wherein the foam comprises a particulate material, and wherein the foam is particle stabilized.

Abstract: A process quantifies a concentration of a targeted molecule in a liquid sample by pulsing signal and reference beams from their own sources, then spatially combining the pulsed beams into a single radiation beam which passes into the liquid sample and then detecting pulsed output beams after the single radiation beam passes out of the liquid sample. The pulsed outputs of the signal and reference beams are processed to obtain a value over a preselected period of time and, if an interference beam is used, it is processed with the reference beam to obtain a calibration curve adjustment representative of optical interference represented by at least one interfering molecule concentration which is used to calculate the concentration level of the targeted particle in the liquid sample. Two detectors, which may have an optical co-axial configuration, can be used for detection of pulsed beams.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A tissue anchor is provided that includes a head connected to a shaft, and a tissue-coupling element extending from the shaft. The shaft includes a seal that is configured to form a blood-tight seal between the shaft and a heart wall, and to promote hemostasis. When the tissue anchor is unconstrained, the head is coaxial with an axis of the shaft, and the tissue-coupling element is generally orthogonal to the axis and is shaped such that if the tissue-coupling element were to be projected onto a plane that is perpendicular to the axis, at least 80% of an area of a projection of the tissue-coupling element on the plane would fall within a first angle of 180 degrees in the plane having a vertex at the axis. Other embodiments are also described.

Abstract: There is provided a user device including a transceiver, a processor, and a memory. The memory stores a device management application (DMA) arranged to disable at least one function of the user device in accordance with an operative device policy state of the user device, and a device policy schedule comprising a queue of device policy states each having an associated respective set of policy data. Responsive to receiving, from a remote system via the transceiver, first synchronisation data indicating a first device policy state in the queue of device policy states, the DMA is arranged to update the operative device policy state of the user device to the indicated first device policy state.

Abstract: A tissue anchor is provided for delivery by a deployment tool in a constrained state, the tissue anchor including a shaft; a tissue-coupling element, which includes a wire including a shape-memory alloy; and a flexible elongate tension member, which is distinct from the wire. The flexible elongate tension member includes a distal portion that is fixed to a site on the wire such that, when the tissue anchor is not constrained by the deployment tool, the flexible elongate tension member applies, to the tissue-coupling element, tension that constrains lateral expansion of the tissue-coupling element away from a central longitudinal axis of the shaft, by preventing the tissue-coupling element from automatically assuming a predetermined shape provided by the shape-memory alloy of the wire. Other embodiments are also described.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: A tissue anchor includes a shaft, a tissue-coupling element, and a flexible elongate tension member. The tissue-coupling element includes a wire, which is shaped as an open loop having more than one turn when the tissue anchor is unconstrained. The tension member includes a distal portion that is fixed to a site on the open loop, a proximal portion, which has a longitudinal segment that runs alongside at least a portion of the shaft, and a crossing portion, which (i) is disposed between the distal and the proximal portions along the tension member, and (ii) crosses at least a portion of the open loop when the tissue anchor is unconstrained. The tissue anchor is configured to allow relative axial motion between the at least a portion of the shaft and the longitudinal segment of the proximal portion of the tension member when the tissue anchor is unconstrained.

Abstract: A system for servicing cable includes a field-end assembly and a live-end assembly, each mounted on a support structure having at least one drive mechanism operable to cause relative linear movement between the field-end assembly and the live-end assembly. A drilling-and-shorting assembly can create a short circuit in a field-end of the cable, and a continuity tester can test the integrity of the short-circuit. The field end of the cable can then be ejected from the system, and an end-cap-cradle assembly can position an end cap on a live end of an electrical cable and test its installation.

Abstract: A method for predicting an unfavorable outcome for a patient admitted to a hospital, e.g., with a COVID-19 infection is described. Attributes from an electronic health record for the patient are obtained including at least findings obtained at admission, basic patient characteristics, and laboratory data. The attributes are supplied to a classifier implemented in a programmed computer which is trained to predict a risk of the unfavorable outcome. The classifier is arranged as a hierarchical combination of (a) an initial binary classifier stratifying the patient into either a high risk group or a low risk group, and (b) child classifiers further classifying the patient in a lowest risk group or a highest risk group depending how the initial binary classifier stratified the patient as either a member of the high risk or low risk group.

Abstract: Shapley values (SVs) have become an important tool to further the goal of explainability of machine learning (ML) models. However, the computational load of exact SV calculations increases exponentially with the number of attributes. Hence, the calculation of SVs for models incorporating large numbers of interpretable attributes is problematic. Molecular diagnostic tests typically seek to leverage information from hundreds or thousands of attributes, often using training sets with fewer instances. Methods are described for evaluate SVs using Monte Carlo sampling or exact calculation in polynomial time (i.e., reasonably quickly and efficiently) using the architecture of a ML model designed for robust molecular test generation, and without requiring classifier retraining.

Abstract: A method is provided including introducing, during a transcatheter procedure, a tissue anchor into a cardiac chamber of a heart of a subject, while a tissue-coupling element of the tissue anchor is constrained by a deployment tool. The tissue-coupling element is delivered distally through a cardiac wall. The tissue anchor is at least partially released from the deployment tool such that the tissue-coupling element is unconstrained by the deployment tool. After the tissue-coupling element is delivered through the cardiac wall and the tissue anchor is at least partially released from the deployment tool, whether the tissue-coupling element overlies a coronary blood vessel is ascertained, and, if the tissue-coupling element overlies the coronary blood vessel, the tissue anchor is rotated until the tissue-coupling element no longer overlies the coronary blood vessel. Thereafter, the tissue-coupling element is proximally pulled by applying tension to the tissue anchor. Other embodiments are also described.

Abstract: Flavour components for use in oral care compositions containing a metal salt, compositions comprising the flavour components, and methods of making and using the same. The flavour components comprise a taste-masking agent comprising cinnamic aldehyde, eugenol and eucalyptol and one or more flavouring agents selected from L-menthol, N-ethyl-p-enthane-3-carboxamide, anethole, peppermint oil, spearmint oil and corn mint oil. In particular the metal salt is zinc citrate.