Abstract: Techniques and apparatuses are described for deep neural network (DNN) processing for a user equipment-coordination set (UECS). A network entity selects (910) an end-to-end (E2E) machine-learning (ML) configuration that forms an E2E DNN for processing UECS communications. The network entity directs (915) each device of multiple devices participating in an UECS to form, using at least a portion of the E2E ML configuration, a respective sub-DNN of the E2E DNN that transfers the UECS communications through the E2E communication, where the multiple devices include at least one base station, a coordinating user equipment (UE), and at least one additional UE. The network entity receives (940) feedback associated with the UECS communications and identifies (945) an adjustment to the E2E ML configuration. The network entity then directs at least some of the multiple devices participating in an UECS to update the respective sub-DNN of the E2E DNN based on the adjustment.

Abstract: A method is described for generating a prediction of a disease classification error for a magnified, digital microscope slide image of a tissue sample. The image is composed of a multitude of patches or tiles of pixel image data. An out-of-focus degree per patch is computed using a machine learning out-of-focus classifier. Data representing expected disease classifier error statistics of a machine learning disease classifier for a plurality of out-of-focus degrees is retrieved. A mapping of the expected disease classifier error statistics to each of the patches of the digital microscope slide image based on the computed out-of-focus degree per patch is computed, thereby generating a disease classifier error prediction for each of the patches. The disease classifier error predictions thus generated are aggregated over all of the patches.

Abstract: A method for determining features of a trailer being towed by a vehicle includes initiating a calibration drive of the vehicle and capturing frames of image data via a vehicle camera, and, via processing by an image processor of frames of captured image data, determining features of the trailer being towed by the vehicle during the calibration drive. The features of the trailer are determined by determining features that have similar changes between a current frame of captured image data and a previous frame of captured image data captured during the calibration drive. The features are tracked over multiple frames of captured image data for different angular positions of the trailer relative to the vehicle to determine trailer length from the hitch ball of the vehicle to an axle of the trailer. The trailer angle is determined based on the determined trailer length.

Abstract: Embodiments discussed herein facilitate segmentation of histological primitives from stained histology of renal biopsies via deep learning and/or training deep learning model(s) to perform such segmentation. One example embodiment is configured to access a first histological image of a renal biopsy comprising a first type of histological primitives, wherein the first histological image is stained with a first type of stain; provide the first histological image to a first deep learning model trained based on the first type of histological primitive and the first type of stain; and receive a first output image from the first deep learning model, wherein the first type of histological primitives is segmented in the first output image.

Abstract: A computing-device implemented system and method for identifying an item in an x-ray image is described. The method includes training a machine learning algorithm with at least one training data set of x-ray images to generate at least one machine-learned model. The method further includes receiving at least one rendered x-ray image that includes an item, identifying the item using the at least one model, and generating an automated detection indication associated with the item.

Abstract: A method to optimize learning based upon ocular information of a subject includes providing a video camera for recording a close-up view of a subject's eye. A first electronic display shows a plurality of educational subject matter to the subject. A second electronic display shows an output to an instructor. Changes in ocular signals of the subject are processed through the use optimized algorithms. A cognitive state model determines a low to a high cognitive load experienced by the subject. The cognitive state model is evaluated based on the changes in the ocular signals for determining a probability of the low to the high cognitive load experienced by the subject. The probability of the low to the high cognitive load experienced by the subject is displayed to the instructor.

Abstract: The present disclosure is directed to an apparatus and method for data analysis for use in data classification via training of a recurrent neural network to identify features from limited reference sets. Based on a one-shot learning algorithm, the method includes selecting a subset of reference data and training a classifier with the selected data. This small subset of reference data can be iteratively tuned to enhance classification of the data according to the desired output of the method. The apparatus may be configured to allow a user to interactively select a subset of reference data which is used to train the classifier and to evaluate classifier performance.

Abstract: The invention is notably directed to a computer-implemented method for normalizing medical images, e.g., whole slide images. This method includes steps performed for each image of a first subset of images of a dataset. Actual quantities are estimated for each image, including actual stain vectors and, possibly, robust maximum stain concentrations (typically hematoxylin and eosin stain vectors and concentrations). The actual quantities estimated are assessed by comparing them to reference data based on reference quantities estimated for one or more images of a second subset of images of the dataset, where the second subset of images differ from the first subset of images. The reference quantities include reference stain vectors. For each image, either the actual quantities or the reference quantities for the dataset are selected as effective quantities, based on an outcome of the previous assessment of the actual quantities. Each image is then normalized.

Abstract: A method for use of machine learning in computer-assisted anatomical prediction. The method includes identifying with a processor parameters in a plurality of training images to generate a training dataset, the training dataset having data linking the parameters to respective training images, training at least one machine learning algorithm based on the parameters in the training dataset and validating the trained machine learning algorithm, identifying with the processor digitized points on a plurality of anatomical landmarks in a radiographic image of a person's skeleton displayed on a screen by determining anatomical relationships of adjacent bony structures as well as dimensions of at least a portion of a body of the skeleton in the displayed image using the validated machine learning algorithm and a scale factor for the displayed image, and making an anatomical prediction of the person's skeletal alignment based on the determined anatomical dimensions and a known morphological relationship.

Abstract: Apparatuses, systems, methods, and computer program products are presented for a propensity module based optimization. An apparatus comprises a processor and a memory that stores code executable by the processor to receive an electronic submission for a pass/fail interface, identify information from the electronic submission to suggest to a user for entering into an input field for the pass/fail interface prior to submitting the electronic submission to the pass/fail interface to reduce a likelihood that the electronic submission will be rejected at the pass/fail interface, determine the likelihood that the electronic submission will be accepted by the pass/fail interface, and submit the electronic submission to the pass/fail interface in response to the likelihood satisfying a threshold.

Abstract: Described are systems and methods for detecting objects using calibrated imaging devices and obfuscating, in real-time or near real time, portions of the video data to protect the privacy of operators represented in the video data. For example, a position of an operator within a fulfillment center may be determined or tracked in video data and the pixels representative of that operator may be obfuscated using pixilation and/or other techniques so that a reviewing agent that is viewing the video data cannot determine the identity of the operator. Such obfuscation may be performed in real-time or near real-time using automated processing. In addition, only portions of the video data may be obfuscated so that events (e.g., item picks, item place) and/or other objects represented in the video data are still viewable to the reviewing agent.

Abstract: A method for determining a trailer characteristic includes disposing a camera at a rear portion of a vehicle so as to have a field of view rearward, providing a control having an image processor, and hitching a tongue of a trailer to a hitch ball of the vehicle. Location of a portion of the trailer relative to the camera is determined via processing of captured image data. Responsive to the determination of the location of the portion of the trailer relative to the camera, a subregion of the imaging array of the camera is determined that includes the determined portion of the trailer, and processing at the control is enhanced at the determined subregion during processing of subsequent frames of captured image data to determine location of the portion of the trailer relative to the vehicle centerline in the subsequent frames of captured image data.

Abstract: A computer-implemented method to perform image-to-image translation. The method can include obtaining one or more machine-learned generator models. The one or more machine-learned generator models can be configured to receive an input image and a user-specified conditioning vector that parameterizes one or more desired values for one or more defined characteristics of an output image. The one or more machine-learned generator models can be configured to perform, based at least in part on the user-specified conditioning vector, one or more transformations on the input image to generate the output image with the one or more desired values for the one or more defined characteristics. The method can include receiving the input image and the user-specified conditioning vector. The method can include generating, using the machine-learned generator model, an output image having the one or more desired values for the one or more characteristics.

Abstract: Aneurysms are classified and quantitatively analyzed based on medical image data acquired from a subject. In general, one or more algorithms are implemented to automatically classify, or otherwise diagnose, and measure aneurysms and their change over time. These algorithms make use of artificial intelligence and deep learning to develop quantitative analytics that can be consolidated into diagnostic reports.

Abstract: The present invention provides a comprehensive detection device and method for a cancerous region, and belongs to the technical field of deep learning. In the present invention, a cancerous region detection network is trained for preprocessed and annotated CT image data to predict bounding box coordinates of a cancerous region and a corresponding cancer confidence score; a clinical analysis network is trained for preprocessed clinical data with a cancer risk level to predict a cancer probability value of a corresponding patient; and a predicted cancer probability value is weighted to a predicted cancer confidence score to realize a comprehensive determination of the cancerous region. The present invention can detect a cancerous region with high accuracy and high performance.

Abstract: A medical image segmentation method based on a U-Net, including: sending real segmentation image and original image to a generative adversarial network for data enhancement to generate a composite image with a label; then putting the composite image into original data set to obtain an expanded data set, and sending the expanded data set to improved multi-feature fusion segmentation network for training. A Dilated Convolution Module is added between the shallow and deep feature skip connections of the segmentation network to obtain receptive fields with different sizes, which enhances the fusion of detail information and deep semantics, improves the adaptability to the size of the segmentation target, and improves the medical image segmentation accuracy. The over-fitting problem that occurs when training the segmentation network is alleviated by using the expanded data set of the generative adversarial network.

Abstract: Scene filtering using motion estimation, including identifying, in camera data from an autonomous vehicle, based on motion relative to the autonomous vehicle, one or more pixels; filtering, from the camera data, the one or more pixels; and training, based on the filtered camera data, a neural network.

Abstract: Apparatuses, systems, and techniques to train and apply a first machine learning model to identify a plurality of regions of interest within an input image, and to train and apply a plurality of second machine learning models to identify one or more objects within each region of interest identified by the first machine learning model.

Abstract: Even if an existing learning dataset is limited, a new learning dataset with sufficient variation is generated. Therefore, for each of a plurality of learning data subsets, new input signals are generated from input signals of a plurality of pieces of learning data, and a plurality of pieces of new learning data that are respectively combinations of the new input signals and output signals of the corresponding learning data subset are generated. The input signals of the plurality of pieces of the learning data included in the corresponding learning data subset are divided into a first signal group and a second signal group, and the new input signals are generated by a learning device that is generated by performing learning by the first signal group set as an input signal set and the second signal group set as an output signal set.

Abstract: The invention discloses a deep network lung texture recognition method combined with multi-scale attention, which belongs to the field of image processing and computer vision. In order to accurately recognize the typical texture of diffuse lung disease in computed tomography (CT) images of the lung, a unique attention mechanism module and multi-scale feature fusion module were designed to construct a deep convolutional neural network combing multi-scale and attention, which achieves high-precision automatic recognition of typical textures of diffuse lung diseases. In addition, the proposed network structure is clear, easy to construct, and easy to implement.