Abstract: In pose estimation from a depth sensor (12), depth information is matched (70) with 3D information. Depending on the shape captured in depth image information, different objects may benefit from more or less pose density from different perspectives. The database (48) is created by bootstrap aggregation (64). Possible additional poses are tested (70) for nearest neighbors already in the database (48). Where the nearest neighbor is far, then the additional pose is added (72). Where the nearest neighbor is not far, then the additional pose is not added. The resulting database (48) includes entries for poses to distinguish the pose without overpopulation. The database (48) is indexed and used to efficiently determine pose from a depth camera (12) of a given captured image.

Abstract: The present embodiments relate to generating synthetic depth data. By way of introduction, the present embodiments described below include apparatuses and methods for modeling the characteristics of a real-world light sensor and generating realistic synthetic depth data accurately representing depth data as if captured by the real-world light sensor. To generate accurate depth data, a sequence of procedures are applied to depth images rendered from a three-dimensional model. The sequence of procedures simulate the underlying mechanism of the real-world sensor. By simulating the real-world sensor, parameters relating to the projection and capture of the sensor, environmental illuminations, image processing and motion are accurately modeled for generating depth data.

Abstract: The present embodiments relate to automatically estimating a three]dimensional pose of an object from an image captured using a camera with a structured light sensor. By way of introduction, the present embodiments described below include apparatuses and methods for training a system for and estimating a pose of an object from a test image. Training and test images are sampled to generate local image patches. Features are extracted from the local image patches to generate feature databased used to estimate nearest neighbor poses for each local image patch. The closest nearest neighbor pose to the test image is selected as the estimated three]dimensional pose.

Abstract: A method and system for classification of endoscopic images is disclosed. An initial trained deep network classifier is used to classify endoscopic images and determine confidence scores for the endoscopic images. The confidence score for each endoscopic image classified by the initial trained deep network classifier is compared to a learned confidence threshold. For endoscopic images with confidence scores higher than the learned threshold value, the classification result from the initial trained deep network classifier is output. Endoscopic images with confidence scores lower than the learned confidence threshold are classified using a first specialized network classifier built on a feature space of the initial trained deep network classifier.

Abstract: A method for predicting short-term cloud coverage includes a computer calculating an estimated cloud velocity field at a current time value based on sky images. The computer determines a segmented cloud model based on the sky images, a future sun location corresponding to a future time value, and sun pixel locations at the future time value based on the future sun location. Next, the computer applies a back-propagation algorithm to the sun pixel locations using the estimated cloud velocity field to yield propagated sun pixel locations corresponding to a previous time value. Then, the computer predicts cloud coverage for the future sun location based on the propagated sun pixel locations and the segmented cloud model.

Abstract: A method of characterizing a specimen for HILN (H, I, and/or L, or N). The method includes capturing images of the specimen at multiple different viewpoints, processing the images to provide segmentation information for each viewpoint, generating a semantic map from the segmentation information, selecting a synthetic viewpoint, identifying front view semantic data and back view semantic data for the synthetic viewpoint, and determining HILN of the serum or plasma portion based on the front view semantic data with an HILN classifier, while taking into account back view semantic data. Testing apparatus and quality check modules adapted to carry out the method are described, as are other aspects.

Abstract: The present disclosure is directed to a method and device for managing medical data. The method may include receiving medical image data of a plurality of patient cases acquired by at least one image acquisition device. The method may further include determining diagnosis results, by a processor, of the medical image data using an artificial intelligence method. The method may also include determining, by the processor, priority scores for the medical image data based on the respective diagnosis results, and sorting, by the processor, the medical image data based on the priority score. The method may yet further include presenting a queue of the medical image data on a display according to the sorted order.

Abstract: Methods and systems for artificial intelligence based medical image segmentation are disclosed. In a method for autonomous artificial intelligence based medical image segmentation, a medical image of a patient is received. A current segmentation context is automatically determined based on the medical image and at least one segmentation algorithm is automatically selected from a plurality of segmentation algorithms based on the current segmentation context. A target anatomical structure is segmented in the medical image using the selected at least one segmentation algorithm.

Abstract: Embodiments of the disclosure provide systems and methods for generating a report based on medical images of a patient. An exemplary system includes a communication interface configured to receive the medical images acquired by an image acquisition device. The system may further include at least one processor. The at least one processor is configured to receive a user selection of at least one medical image in at least one view. The at least one processor is further configured to automatically generate keywords describing the selected medical image based on a learning network including a convolutional neural network and a recursive neural network connected in series. The at least one processor is also configured to receive a keyword selection among the generated keywords and generate the report based on the keyword selection. The exemplary system additionally includes a display configured to display the selected medical image and the report.

Abstract: A probe pose is detected in fluoroscopy medical imaging. The pose of the probe through a sequence of fluoroscopic images is detected. The detection relies on an inference framework for visual tracking overtime. By applying visual tracking, the pose through the sequence is consistent or the pose at one time guides the detection of the probe at another time. Single frame drop-out of detection may be avoided. Verification using detection of the tip of the probe and/or weighting of possible detections by separate detection of markers on the probe may further improve the accuracy.

Abstract: Embodiments of the disclosure provide systems and methods for segmenting a medical image. An exemplary system includes a communication interface configured to receive the medical image acquired by an image acquisition device. The system further includes a memory configured to store a multi-level learning network including at least a first convolution block and a second convolution block. The second convolution block has at least one convolution layer. The system also includes a processor. The processor is configured to determine a first feature map by applying the first convolution block to the medical image, and determine a second feature map by applying the second convolution block to the first feature map. The processor is further configured to determine a first level feature map by concatenating the first feature map and the second feature map. The processor is also configured to obtain a first level segmented image based on the first level feature map.

Abstract: Systems, methods, and computer-readable media are disclosed for determining feature representations of 2.5D image data using deep learning techniques. The 2.5D image data may be synthetic image data generated from 3D simulated model data such as 3D CAD data. The 2.5D image data may be indicative of any number of pose estimations/camera poses representing virtual or actual viewing perspectives of an object modeled by the 3D CAD data. A neural network such as a convolution neural network (CNN) may be trained using the 2.5D image data as training data to obtain corresponding feature representations. The pose estimations/camera poses may be stored in a data repository in association with the corresponding feature representations. The learnt CNN may then be used to determine an input feature representation from an input 2.5D image and index the input feature representation against the data repository to determine matching pose estimation(s).

Abstract: Embodiments of the disclosure provide systems and methods for detecting cancer metastasis in a whole-slide image. The system may include a communication interface configured to receive the whole-slide image and a learning model. The whole-slide image is acquired by an image acquisition device. The system may also include a memory configured to store a plurality of tiles derived from the whole-slide image in a queue. The system may further include at least one processor, configured to apply the learning model to at least two tiles stored in the queue in parallel to obtain detection maps each corresponding to a tile, and detect the cancer metastasis based on the detection maps.

Abstract: Embodiments of the disclosure provide systems and methods for generating a diagnosis report based on a medical image of a patient. The system includes a communication interface configured to receive the medical image acquired by an image acquisition device. The system further includes at least one processor. The at least one processor is configured to detect a medical condition of the patient and parameters associated with the medical condition based on the medical image. The at least one processor is further configured to construct the diagnosis report based on the medical image, wherein the diagnosis report includes at least one view of the medical image and a description of the medical condition using the parameters. The system also includes a display configured to display the diagnosis report.

Abstract: Embodiments of the disclosure provide systems and methods for segmenting a medical image. The system includes a communication interface configured to receive the medical image acquired by an image acquisition device. The system also includes a memory configured to store a plurality of learning networks jointly trained using first training images of a first imaging modality and second training images of a second imaging modality. The system further includes a processor, configured to segment the medical image using a segmentation network selected from the plurality of learning networks.

Abstract: Systems, methods, and computer-readable media are disclosed for automated identification of parts of a parts assembly using image data of the parts assembly and 3D simulated model data of the parts assembly. The 3D simulated model data may be 3D CAD data of the parts assembly. An image of the parts assembly is captured by a mobile device and sent to a back-end server for processing. The back-end server determines a feature representation corresponding to the image and searches a repository to locate a matching feature representation stored in association with a corresponding pose estimation. The matching pose estimation is rendered as an overlay on the image of the parts assembly, thereby enabling automated identification of parts within the image or some user-selected portion of the image.

Abstract: A method for analyzing digital holographic microscopy (DHM) data for hematology applications includes receiving a plurality of DHM images acquired using a digital holographic microscopy system. One or more connected components are identified in each of the plurality of DHM images and one or more training white blood cell images are generated from the one or more connected components. A classifier is trained to identify a plurality of white blood cell types using the one or more training white blood cell images. The classifier may be applied to a new white blood cell image to determine a plurality of probability values, each respective probability value corresponding to one of the plurality of white blood cell types. The new white blood cell image and the plurality of probability values may then be presented in a graphical user interface.

Abstract: A method for identifying a feature in a first image comprises establishing an initial database of image triplets, and in a pose estimation processor, training a deep learning neural network using the initial database of image triplets, calculating a pose for the first image using the deep learning neural network, comparing the calculated pose to a validation database populated with images data to identify an error case in the deep learning neural network, creating a new set of training data including a plurality of error cases identified in a plurality of input images and retraining the deep learning neural network using the new set of training data. The deep learning neural network may be iteratively retrained with a series of new training data sets. Statistical analysis is performed on a plurality of error cases to select a subset of the error cases included in the new set of training data.

Abstract: A computer-implemented method for automatically detecting a target object from a 3D image is disclosed. The method may include receiving the 3D image acquired by an imaging device. The method may further include detecting, by a processor, a plurality of bounding boxes as containing the target object using a 3D learning network. The learning network may be trained to generate a plurality of feature maps of varying scales based on the 3D image. The method may also include determining, by the processor, a set of parameters identifying each detected bounding box using the 3D learning network, and locating, by the processor, the target object based on the set of parameters.

Abstract: The present disclosure is directed to a method and system for automatically detecting a physiological condition from a medical image of a patient. The method may include receiving the medical image acquired by an imaging device. The method may further include detecting, by a processor, target objects and obtaining the corresponding target object patches from the received medical image. And the method may further include determining, by the processor, a first parameter using a first learning network for each target object patch. The first parameter represents the physiological condition level of the corresponding target object, and the first learning network is trained by adding one or more auxiliary classification layers. This method can quickly, accurately, and automatically predict target object level and/or image (patient) level physiological condition from a medical image of a patient by means of a learning network, such as 3D learning network.