Abstract: Techniques for generating counterfactuals in connection with machine learning models. The techniques include applying a trained machine learning model to an input to obtain a first outcome; determining whether the first outcome has a value in a set of one or more target values; when it is determined that the first outcome does not have a value in the set of one or more target values, generating a counterfactual input at least in part by applying a trained neural network model to the input to obtain a corresponding output, the corresponding output indicating changes to be made to one or more values of one or more attributes of the input to obtain the counterfactual input, and generating feedback based on the counterfactual input.

Abstract: A personal self-propelled vehicle for travelling on snow using ski comprises a closed-loop flexible ribbon configured to envelope the ski over its length so as to pass under at least a portion of a sliding surface, two rollers and configured for installation on longitudinally opposite ends of the ski and to interact with the ribbon. The roller is a driving roller and is engaged with a motor comprising a control means and with the ribbon to transfer force for translational movement thereof. The roller is a guiding roller. The ribbon canvas has multiple via openings divided by webs and substantially constitutes a lattice. Dimensions and positions of the openings and parameters of the webs between them are selected so the sliding surface of the ski abuts on snow through the openings during motion of the vehicle, while the webs sink in snow under load of the vehicle bearing the user and provide repulsion from snow packed by the ski, thus assuring the ski sliding in longitudinal direction.

Abstract: Techniques for generating counterfactuals in connection with machine learning models. The techniques include applying a trained machine learning model to an input to obtain a first outcome; determining whether the first outcome has a value in a set of one or more target values; when it is determined that the first outcome does not have a value in the set of one or more target values, generating a counterfactual input at least in part by applying a trained neural network model to the input to obtain a corresponding output, the corresponding output indicating changes to be made to one or more values of one or more attributes of the input to obtain the counterfactual input, and generating feedback based on the counterfactual input.

Abstract: A personal self-propelled vehicle for travelling on snow using ski comprises a closed-loop flexible ribbon configured to envelope the ski over its length so as to pass under at least a portion of a sliding surface, two rollers and configured for installation on longitudinally opposite ends of the ski and to interact with the ribbon. The roller is a driving roller and is engaged with a motor comprising a control means and with the ribbon to transfer force for translational movement thereof. The roller is a guiding roller. The ribbon canvas has multiple via openings divided by webs and substantially constitutes a lattice. Dimensions and positions of the openings and parameters of the webs between them are selected so the sliding surface of the ski abuts on snow through the openings during motion of the vehicle, while the webs sink in snow under load of the vehicle bearing the user and provide repulsion from snow packed by the ski, thus assuring the ski sliding in longitudinal direction.

Abstract: A device and method for producing a portion of platelet rich plasma from a fluid sample obtained from a patient. The device comprises three intercommunicated portions, each portion having an internal chamber defined therein. The fluid sample is first brought into the base portion where it is separated into red blood cells and plasma. The plasma portion of the sample is then withdrawn into the second portion where it is further separated into platelet poor plasma disposed over platelet rich plasma. The portion of platelet poor plasma is then withdrawn from the second chamber, leaving only platelet rich plasma within the second chamber. The base portion also includes a membrane located on its distal end which helps maintain a vacuum within the device when the fluid sample is being initially inserted into base chamber.

Abstract: A device and method for producing a portion of platelet rich plasma from a fluid sample obtained from a patient. The device comprises three intercommunicated portions, each portion having an internal chamber defined therein. The fluid sample is first brought into the base portion where it is separated into red blood cells and plasma. The plasma portion of the sample is then withdrawn into the second portion where it is further separated into platelet poor plasma disposed over platelet rich plasma. The portion of platelet poor plasma is then withdrawn from the second chamber, leaving only platelet rich plasma within the second chamber. The base portion also includes a membrane located on its distal end which helps maintain a vacuum within the device when the fluid sample is being initially inserted into base chamber.

Abstract: An inflatable oral airway (IOA) device for maintaining the patency of an airway of a patient without substantial extension into the pharynx or larynx of the patient. The IOA is inserted into the patient's oro-pharynx after the patient is sedated or under general anesthesia. The IOA comprises a core that is substantially shaped the same as a conventional or normal human airway. The IOA has an inverted orientation, i.e. the bottom or ventral surface of the core rests on the upper surface of the patient's tongue. The IOA also comprises a cuff and a hinge balloon which may be inflated in tandem or in sequence to open and support the soft tissues of the mouth and pharynx in four different directions as well as articulate and increase the internal volume of the IOA itself.