Abstract: A vacuum tube receiver can include a housing that has a proximal end and a distal end. The housing can include a hollow interior and a proximal opening for receiving a vacuum tube into the hollow interior. The distal end of the housing can form an adapter for coupling the vacuum tube receiver to an intravenous system. The vacuum tube receiver may also include one or more spikes that extend proximally into the hollow interior to form a blood flow path. The vacuum tube receiver can include a number of features to control the pressure and/or flowrate of blood into a vacuum tube.

Abstract: A catheter system may include a catheter adapter, a catheter extending distally from the distal end of the catheter adapter, and a needle hub coupled to the catheter adapter. The needle hub may be transparent, and a needle may be secured within the needle hub. The catheter system may also include a flow control plug coupled to the proximal end of the needle hub. In some embodiments, a flashback pathway may be disposed between an outer surface of the flow control plug and an inner surface of the needle hub. In some embodiments, the catheter system may include an inner barrel and an outer barrel, and the inner barrel and the needle hub may be configured to move proximally within the outer barrel to retract the needle. The flashback pathway may be disposed between an outer surface of the needle hub and an inner surface of the inner barrel.

Abstract: A method for improving geometry-based monocular structure from motion (SfM) by exploiting depth maps predicted by convolutional neural networks (CNNs) is presented. The method includes capturing a sequence of RGB images from an unlabeled monocular video stream obtained by a monocular camera, feeding the RGB images into a depth estimation/refinement module, outputting depth maps, feeding the depth maps and the RGB images to a pose estimation/refinement module, the depths maps and the RGB images collectively defining pseudo RGB-D images, outputting camera poses and point clouds, and constructing a 3D map of a surrounding environment displayed on a visualization device.

Abstract: A computer-implemented method for implementing a self-supervised visual odometry framework using long-term modeling includes, within a pose network of the self-supervised visual odometry framework including a plurality of pose encoders, a convolution long short-term memory (ConvLSTM) module having a first-layer ConvLSTM and a second-layer ConvLSTM, and a pose prediction layer, performing a first stage of training over a first image sequence using photometric loss, depth smoothness loss and pose cycle consistency loss, and performing a second stage of training to finetune the second-layer ConvLSTM over a second image sequence longer than the first image sequence.

Abstract: A computer-implemented method, system, and computer program product are provided for a stabilization system utilizing pose estimation in an aerial drone. The method includes receiving, by a pose estimation system, a plurality of images from one or more cameras. The method also includes predicting, by the pose estimation system, a pose from the score map and a combined feature map, the combined feature map correlated from a pair of the plurality of images. The method additionally includes moving, by a propulsion system, the aerial drone responsive to the pose.

Abstract: The probabilistic accumulation method is an independent and comprehensive approach for validating third-party catastrophe models. The method starts with a standard technique, namely, a limited scenario accumulation analysis, and extends the approach via a novel sampling methodology to evaluate the third-party models against a significantly larger data set. The sampling approach includes a technique for extending the fixed frequency and severity assumption used by the vendor catastrophe models. The result is a more complete set of loss estimates against which to evaluate the vendor model output.

Abstract: A computer-implemented method, system, and computer program product are provided for pose estimation. The method includes receiving, by a processor, a plurality of images from one or more cameras. The method also includes generating, by the processor with a feature extraction convolutional neural network (CNN), a feature map for each of the plurality of images. The method additionally includes estimating, by the processor with a feature weighting network, a score map from a pair of the feature maps. The method further includes predicting, by the processor with a pose estimation CNN, a pose from the score map and a combined feature map. The method also includes controlling an operation of a processor-based machine to change a state of the processor-based machine, responsive to the pose.

Abstract: A method for correcting blur effects is presented. The method includes generating a plurality of images from a camera, synthesizing blurred images from sharp image counterparts to generate training data to train a structure-and-motion-aware convolutional neural network (CNN), and predicting a camera motion and a depth map from a single blurred image by employing the structure-and-motion-aware CNN to remove blurring from the single blurred image.

Abstract: A method for estimating dense 3D geometric correspondences between two input point clouds by employing a 3D convolutional neural network (CNN) architecture is presented. The method includes, during a training phase, transforming the two input point clouds into truncated distance function voxel grid representations, feeding the truncated distance function voxel grid representations into individual feature extraction layers with tied weights, extracting low-level features from a first feature extraction layer, extracting high-level features from a second feature extraction layer, normalizing the extracted low-level features and high-level features, and applying deep supervision of multiple contrastive losses and multiple hard negative mining modules at the first and second feature extraction layers.

Abstract: A method is provided in an Advanced Driver-Assistance System (ADAS). The method extracts, from an input video stream including a plurality of images using a multi-task Convolutional Neural Network (CNN), shared features across different perception tasks. The perception tasks include object detection and other perception tasks. The method concurrently solves, using the multi-task CNN, the different perception tasks in a single pass by concurrently processing corresponding ones of the shared features by respective different branches of the multi-task CNN to provide a plurality of different perception task outputs. Each respective different branch corresponds to a respective one of the different perception tasks. The method forms a parametric representation of a driving scene as at least one top-view map responsive to the plurality of different perception task outputs.

Abstract: Systems and methods for detecting traffic scenarios include an image capturing device which captures two or more images of an area of a traffic environment with each image having a different view of vehicles and a road in the traffic environment. A hierarchical feature extractor concurrently extracts features at multiple neural network layers from each of the images, with the features including geometric features and semantic features, and for estimating correspondences between semantic features for each of the images and refining the estimated correspondences with correspondences between the geometric features of each of the images to generate refined correspondence estimates. A traffic localization module uses the refined correspondence estimates to determine locations of vehicles in the environment in three dimensions to automatically determine a traffic scenario according to the locations of vehicles. A notification device generates a notification of the traffic scenario.

Abstract: A blood sample collection system, may include a housing, the housing including a blood collection port; and a baffle chamber; a needle formed through the housing and into the blood collection port; and a baffle formed within the baffle chamber to counteract a vacuum within a blood collection tube when pierced by the needle.

Abstract: An intravenous therapy system, may include a processor; a data storage device; and a handheld ultrasound probe to detect structures within a patient's body, the handheld ultrasound probe including a video display device physically and operatively coupled to the handheld to display the structures within the patient's body and a magnetic field detector to detect the presence of a vascular access device (VAD) and provide closed-loop feedback to guide the VAD into a blood vessel within the patient's body detected by the ultrasound probe.

Abstract: An intravenous (IV) device, may include a needle; a catheter coaxially formed around the needle; a fluidic path fluidically coupled to a hollow formed through the needle and catheter; and a printed circuit board (PCB), including: a network interface device; a plurality of sensors operatively coupled to the fluidic path to measure a biological characteristic related to a plurality of patient vitals.

Abstract: Systems and methods for camera self-calibration are provided. The method includes receiving real uncalibrated images, and estimating, using a camera self-calibration network, multiple predicted camera parameters corresponding to the real uncalibrated images. Deep supervision is implemented based on a dependence order between the plurality of predicted camera parameters to place supervision signals across multiple layers according to the dependence order. The method also includes determining calibrated images using the real uncalibrated images and the predicted camera parameters.

Abstract: A catheter system may include a catheter adapter, which may include a distal end and a proximal end. The catheter system may include a catheter, which may include a distal end, a proximal end, a catheter lumen extending through the distal end of the catheter and the proximal end of the catheter, and an inner surface forming the catheter lumen. The catheter may extend distally from the distal end of the catheter adapter. The distal end of the catheter may include one or more holes. The distal end of the catheter may include one or more channels. The holes and/or the channels may facilitate visualization of blood flashback indicating the catheter is disposed within a vein of a patient.

Abstract: A patency checking system, may include a fluid path to be coupled to an intravenous (IV) device at a distal end of the fluid path; a patency port formed along a length of the fluid path; and a plunger fluidically coupled to the fluid path at the patency port to selectively draw blood into the fluid path to confirm patency of the fluid path.

Abstract: An intravenous (IV) therapy selection system, may include a processor; a memory device to maintain an electronic medical record database comprising a plurality of medical records related to a plurality of patients; a clinician training module to provide training services and credentials to a plurality of clinicians; an IV therapy recommendation module to provide, via a display device, an IV device recommendation based on data descriptive of a first patient and a first clinician; and a placement location recommendation module to provide, via the display device, a recommendation on where to introduce the IV device into the first patient's body based on data descriptive of the first patient and the first clinician.

Abstract: An intravenous therapy system may include a housing to house an actuation hub, the actuation hub comprising; a needle hub operatively coupled to a needle to insert the needle into a patient's body; and a catheter hub operatively coupled to a catheter to insert the catheter into a patient's body, the catheter being formed coaxially with an outer surface of the needle; and an inflatable bladder formed along a length of the catheter to maintain the catheter in the patient's body.

Abstract: Systems and methods for correspondence estimation and flexible ground modeling include communicating two-dimensional (2D) images of an environment to a correspondence estimation module, including a first image and a second image captured by an image capturing device. First features, including geometric features and semantic features, are hierarchically extract from the first image with a first convolutional neural network (CNN) according to activation map weights, and second features, including geometric features and semantic features, are hierarchically extracted from the second image with a second CNN according to the activation map weights. Correspondences between the first features and the second features are estimated, including hierarchical fusing of geometric correspondences and semantic correspondences. A 3-dimensional (3D) model of a terrain is estimated using the estimated correspondences belonging to the terrain surface.