Abstract: Techniques and architecture are described that provide a multi-platform tool focused on displaying various levels of a topology of a network to a user while maintaining an original context of the user that was initially selected. Elements within the topology are dynamically added or removed from the user's view, e.g., the display, as the elements come in and out of focus. For example, in configurations, as the user traverses the topology, e.g., graph, the nodes that are not of interest generally become invisible. Thus, the user receives a clear view of how the selected component changed according to the user's actions and may achieve an easily maintained history of the current hops through the network topology.

Abstract: Systems and methods are provided for recording video of a driver in a vehicle and a surrounding visual field of the driver. The system includes a detachable body coupled to a windshield of a vehicle. The system further includes three or more cameras coupled to the detachable body. The three or more cameras are configured to capture surrounding views from the detachable body. One of the three or more cameras faces a driver seat of the vehicle in response to the detachable body being coupled to the windshield of the vehicle. The video may be processed by the system, the cloud, or a combination of them.

Abstract: The present disclosure provides devices, systems and methods for mapping of traffic boundaries.

Abstract: A laser eye surgery system that has a patient interface between the eye and the laser system relying on suction to hold the interface to the eye. The patient interface may be a liquid-filled interface, with liquid used as a transmission medium for the laser. During a laser procedure various inputs are monitored to detect a leak. The inputs may include a video feed of the eye looking for air bubbles in the liquid medium, the force sensors on the patient interface that detect patient movement, and vacuum sensors directly sensing the level of suction between the patient interface and the eye. The method may include combining three monitoring activities with a Bayesian algorithm that computes the probabilities of an imminent vacuum loss event.

Abstract: Systems and methods are provided for recording video of a driver in a vehicle and a surrounding visual field of the driver. The system includes a detachable body coupled to a windshield of a vehicle. The system further includes three or more cameras coupled to the detachable body. The three or more cameras are configured to capture surrounding views from the detachable body. One of the three or more cameras faces a driver seat of the vehicle in response to the detachable body being coupled to the windshield of the vehicle. The video may be processed by the system, the cloud, or a combination of them.

Abstract: Apparatus to treat an eye with an ophthalmic laser system comprises a patient interface having an annular retention structure to couple to an anterior surface of the eye. The retention structure is coupled to a suction line to couple the retention structure to the eye with suction. Liquid is added above the eye to act as a transmissive medium. A coupling sensor is coupled to the suction line to determine coupling of the retention structure to the eye. A separate pressure monitoring circuit having a much smaller volume than the suction line is connected to the annular retention structure to measure suction pressure therein. A system processor coupled to the monitoring pressure sensor includes instructions to interrupt firing of a laser when the pressure measured with a monitoring pressure sensor rises above a threshold amount.

Abstract: Systems and methods are provided for recording video of a driver in a vehicle and a surrounding visual field of the driver. The system includes a detachable body coupled to a windshield of a vehicle. The system further includes three or more cameras coupled to the detachable body. The three or more cameras are configured to capture surrounding views from the detachable body. One of the three or more cameras faces a driver seat of the vehicle in response to the detachable body being coupled to the windshield of the vehicle. The video may be processed by the system, the cloud, or a combination of them.

Abstract: Systems, devices and methods provide, implement, and use vision-based methods of sequence inference for a device affixed to a vehicle.

Abstract: Apparatus to treat an eye with an ophthalmic laser system comprises a patient interface having an annular retention structure to couple to an anterior surface of the eye. The retention structure is coupled to a suction line to couple the retention structure to the eye with suction. Liquid is added above the eye to act as a transmissive medium. A coupling sensor is coupled to the suction line to determine coupling of the retention structure to the eye. A separate pressure monitoring circuit having a much smaller volume than the suction line is connected to the annular retention structure to measure suction pressure therein. A system processor coupled to the monitoring pressure sensor includes instructions to interrupt firing of a laser when the pressure measured with a monitoring pressure sensor rises above a threshold amount.

Abstract: Some embodiments disclosed here provide for a method fragmenting a cataractous lens of a patient's eye using an ultra-short pulsed laser. The method can include determining, within a lens of a patient's eye, a high NA zone where a cone angle of a laser beam with a high numerical aperture is not shadowed by the iris, and a low NA zone radially closer to the iris where the cone angle of the laser beam with a low numerical aperture is not shadowed by the iris. Laser lens fragmentation is accomplished by delivering the laser beam with the high numerical aperture to the high NA zone, and the laser beam with the low numerical aperture to the low NA zone. This can result in a more effective fragmentation of a nucleus of the lens without exposing the retina to radiation above safety standards.

Abstract: A neural network is configured to process image data captured by a vehicle-mounted camera. The neural network includes common processing layers (a trunk) and separate, parallelizable processing layers (branches). An object detection branch of the neural network is trained to detect objects that may be visible from the vehicle-mounted camera, such as cars, trucks, and traffic signs. A curve determination branch is trained to detect and parameterize salient curves, such as lane boundaries and road boundaries. The curve determination branch itself is configured with a trunk and branch architecture, having both common and separate processing layers. A first branch computes a likelihood that a curve is present in a given location of the image data and a second branch further localizes the curve within the given location if such a curve is present. Training of the different branches of the neural network may be decoupled.

Abstract: Systems and methods for curating video and other driving-related data for use in driver coaching, which may include selecting or ranking driving behaviors for coaching, selecting or ranking drivers for coaching, selecting or ranking video and other data to be used in coaching, preparing for, scheduling, and summarizing coaching sessions, matching the format of coaching to the behavior or person being coached, preventing unsafe driving situations, influencing job dispatch decisions based on safety scores, and/or reducing data bandwidth usage based on a determined coaching effectiveness of video data.

Abstract: Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle. Combinations of co-occurring driving events may be detected and used to warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Abstract: Systems and methods are provided for recording video of a driver in a vehicle and a surrounding visual field of the driver. The system includes a detachable body coupled to a windshield of a vehicle. The system further includes three or more cameras coupled to the detachable body. The three or more cameras are configured to capture surrounding views from the detachable body. One of the three or more cameras faces a driver seat of the vehicle in response to the detachable body being coupled to the windshield of the vehicle. The video may be processed by the system, the cloud, or a combination of them.

Abstract: Systems and methods are provided for recording video of a driver in a vehicle and a surrounding visual field of the driver. The system includes a detachable body coupled to a windshield of a vehicle. The system further includes three or more cameras coupled to the detachable body. The three or more cameras are configured to capture surrounding views from the detachable body. One of the three or more cameras faces a driver seat of the vehicle in response to the detachable body being coupled to the windshield of the vehicle. The video may be processed by the system, the cloud, or a combination of them.

Abstract: A laser eye surgery system produces a treatment beam that includes a plurality of laser pulses. An optical coherence tomography (OCT) subsystem produces a source beam used to locate one or more structures of an eye. The OCT subsystem is used to sense the distance between a camera objective on the underside of the laser eye surgery system and the patient's eye. Control electronics compare the sensed distance with a pre-determined target distance, and reposition a movable patient support toward or away the camera objective until the sensed distance is at the pre-determined target distance. A subsequent measurement dependent upon the spacing between the camera objective and the patient's eye is performed, such as determining the astigmatic axis by observing the reflection of a plurality of point source LEDs arranged in concentric rings off the eye.

Abstract: Systems and methods are provided for determining cause of atypical traffic events and/or encouraging good driving behavior. The systems and methods may involve a camera sensor and/or inertial sensors to detect traffic events, as well analytical methods that may attribute a cause to the traffic event. The systems and methods include a processor that is configured to determine an occurrence of an atypical traffic event at or near a monitored vehicle; and determine a cause of the atypical traffic event based on data collected at the monitored vehicle, wherein the cause of the atypical traffic event is at least one of: a driver or control system of the monitored vehicle; a second driver or second control system of a second vehicle; and a road condition.

Abstract: Systems and methods are provided for detecting a driving action that mitigates risk.

Abstract: A laser eye surgery system that has a patient interface between the eye and the laser system relying on suction to hold the interface to the eye, the patient interface using liquid used as a transmission medium for the laser. During a laser procedure sensors monitor the level of liquid within the patient interface and send a signal to control electronics if the level drops below a threshold value. The sensor may be mounted on the inside of the patient interface, within a fluid chamber. Alternatively, a gas flow meter may be added to a suction circuit for the patient interface that detects abnormal suction levels indicating low fluid level.

Abstract: A neural network is configured to process image data captured by a vehicle-mounted camera. The neural network includes common processing layers (a trunk) and separate, parallelizable processing layers (branches). An object detection branch of the neural network is trained to detect objects that may be visible from the vehicle-mounted camera, such as cars, trucks, and traffic signs. A curve determination branch is trained to detect and parameterize salient curves, such as lane boundaries and road boundaries. The curve determination branch itself is configured with a trunk and branch architecture, having both common and separate processing layers. A first branch computes a likelihood that a curve is present in a given location of the image data and a second branch further localizes the curve within the given location if such a curve is present. Training of the different branches of the neural network may be decoupled.