Abstract: Implementations generally relate to detecting dominant tools in surgical videos. In some implementations, a method includes receiving at least one image frame. The method further includes detecting one or more objects in the at least one image frame. The method further includes classifying the one or more objects into one or more tool classifications, where the one or more objects are tools. The method further includes determining a handedness of the one or more tools. The method further includes determining a dominant tool from the one or more tools based at least in part on the one or more classifications of the one or more tools and based at least in part on the handedness of the one or more tools.

Abstract: A method for 2D feature tracking by cascaded machine learning and visual tracking comprises: applying a machine learning technique (MLT) that accepts as a first MLT input first and second 2D images, the MLT operating on the images to provide initial estimates of a start point for a feature in the first image and a displacement of the feature in the second image relative to the first image; applying a visual tracking technique (VT) that accepts as a first VT input the initial estimates of the start point and the displacement, and that accepts as a second VT input the two 2D images, processing the first and second inputs to provide refined estimates of the start point and the displacement; and displaying the refined estimates in an output image.

Abstract: An assistive apparatus for organ localization, includes storing a 3D representation and CT images of an anatomical portion of the body of a subject. A localization circuitry determines a rib region and a spine region in the CT images and calculates first and second number of voxels within a first and second region of the 3D representation, respectively. The localization circuitry determines the right side of the body in the CT images, based on a comparison result for the first and second number of voxels. The localization circuitry detects a first bottom portion of right lung based on a distribution of intensity values of pixels in a region of right lung. The localization circuitry detects a second bottom portion of the rib region and localizes the liver organ in the CT images, from a reference of the detected first bottom portion and the detected second bottom portion.

Abstract: An electronic device that handles recognition of mental behavioral, affect, emotional, mental states, mental health, or mood-based attributes based on deep neural networks (DNNs), stores a set of EEG signals and a set of bio-signals associated with a subject. The electronic device trains a plurality of first recognition models on a training set of EEG signals and a training set of bio-signals associated with different training subjects. The electronic device trains a second recognition model on a feature vector from output layers of the plurality of first recognition models. The electronic device estimates a plurality of dependency or relationship data by application of the trained plurality of first recognition models on the set of EEG signals and bio-signals. The electronic device identifies a mental behavioral attribute of the subject by application of the trained second recognition model on the plurality of signals and their relationship data.

Abstract: A method for 2D feature tracking by cascaded machine learning and visual tracking comprises: applying a machine learning technique (MLT) that accepts as a first MLT input first and second 2D images, the MLT operating on the images to provide initial estimates of a start point for a feature in the first image and a displacement of the feature in the second image relative to the first image; applying a visual tracking technique (VT) that accepts as a first VT input the initial estimates of the start point and the displacement, and that accepts as a second VT input the two 2D images, processing the first and second inputs to provide refined estimates of the start point and the displacement; and displaying the refined estimates in an output image.

Abstract: An assistive apparatus for organ localization, includes storing a 3D representation and CT images of an anatomical portion of the body of a subject. A localization circuitry determines a rib region and a spine region in the CT images and calculates first and second number of voxels within a first and second region of the 3D representation, respectively. The localization circuitry determines the right side of the body in the CT images, based on a comparison result for the first and second number of voxels. The localization circuitry detects a first bottom portion of right lung based on a distribution of intensity values of pixels in a region of right lung. The localization circuitry detects a second bottom portion of the rib region and localizes the liver organ in the CT images, from a reference of the detected first bottom portion and the detected second bottom portion.

Abstract: A method for automatic organ segmentation in CT images comprises in a first step, rough region segmentation; in a second step, coarse organ segmentation; and in a third step, refinement of organ segmentation. The organ may be a liver. Rough region segmentation may comprise applying standard anatomical knowledge to the CT images. Coarse segmentation may comprise identifying organ voxels using a probabilistic model. Refinement of organ segmentation may comprise refinement based on intensity, followed by refinement based on shape. Apparatuses configured to carry out the method are also disclosed.

Abstract: Implementations generally relate to determining tool handedness for surgical videos. In some implementations, a method includes receiving at least one image frame of a plurality of image frames. The method further includes detecting one or more objects in the at least one image frame. The method further includes classifying the one or more objects into one or more tool classifications, where the one or more objects are tools. The method further includes determining, for each tool, if the tool is assistive or non-assistive. The method further includes determining, for each tool, a handedness of the tool.

Abstract: Implementations generally relate to detecting dominant tools in surgical videos. In some implementations, a method includes receiving at least one image frame. The method further includes detecting one or more objects in the at least one image frame. The method further includes classifying the one or more objects into one or more tool classifications, where the one or more objects are tools. The method further includes determining a handedness of the one or more tools. The method further includes determining a dominant tool from the one or more tools based at least in part on the one or more classifications of the one or more tools and based at least in part on the handedness of the one or more tools.

Abstract: Various aspects of a system and a method to provide assistance in a surgery in presence of tissue deformation are disclosed herein. In accordance with an embodiment, the system includes an electronic device that receives one or more tissue material properties of a plurality of surface structures of an anatomical portion. One or more boundary conditions associated with the anatomical portion may also be received. Surface displacement of the anatomical portion may be determined by matching a first surface of the anatomical portion before deformation with a corresponding second surface of the anatomical portion after the deformation. The volume displacement field of the anatomical portion may be computed based on the determined surface displacement, the received one or more tissue material properties, and the received one or more boundary conditions.

Abstract: A framework combines vision and sample-efficient reinforcement-learning based on guided policy search for autonomous driving. A controller extracts environmental information from vision and is trained to drive using reinforcement learning.

Abstract: An image registration system, and a method of operation of an image registration system thereof, including: an imaging unit for obtaining a pre-operation non-invasive imaging volume and for obtaining an intra-operation non-invasive imaging volume; and a processing unit including: a rigid registration module for generating a rigid registered volume based on the pre-operation non-invasive imaging volume, a region of interest module for isolating a region of interest from the intra-operation non-invasive imaging volume, a point generation module, coupled to the region of interest module, for determining feature points of the region of interest, an optimization module, coupled to the point generation module, for matching the feature points with corresponding points of the rigid registered volume for generating a matched point cloud, and an interpolation module, coupled to the optimization module, for generating a non-rigid registered volume based on the matched point cloud for display on a display interface.

Abstract: Various aspects of a system and method for utilization of multiple-camera network to capture static and/or motion scenes are disclosed herein. The system comprises a plurality of unmanned aerial vehicles (UAVs). Each of the plurality of UAVs is associated with an imaging device configured to capture a plurality of images. A first UAV of the plurality of UAVs comprises a first imaging device configured to capture a first set of images of one or more static and/or moving objects. The first UAV is configured to receive focal lens information, current location and current orientation from one or more imaging devices. A target location and a target orientation of each of the one or more imaging devices are determined. Control information is communicated to the one or more other UAVs to modify the current location and current orientation to the determined target location and orientation of each of the one or more imaging devices.

Abstract: The present disclosure describes methods and apparatuses for testing an integrated circuit. In some aspects, a test engine of a system-on-chip (SoC) receives a test trigger from an external test port of a multi-chip module. The multi-chip module includes the SoC and integrated circuitry. Responsive to receiving the test trigger, the test engine initiates execution of a test associated with the integrated circuitry. The test engine also generates test data associated with the test and provides the test data at the external test port of the MCM. In this manner, internal tests can be executed within the multi-chip module. In other aspects, the SoC and the integrated circuitry are not packaged together. The test engine can receive the test trigger from a test port of the SoC and initiate execution of the test. In this way, the SoC may act as a test instrument for the integrated circuitry.

Abstract: Various aspects of a system and method to process an image for generation of a three-dimensional (3D) view of an anatomical portion are disclosed herein. The system includes an image-processing device configured to receive a plurality of two-dimensional (2D) images associated of an anatomical portion, each with a first field-of-view (FOV) value. A first 3D view of the anatomical portion with a second FOV value, is generated from a 3D reconstruction of the anatomical portion. The 3D reconstruction is obtained by use of a set of 2D image frames selected from the received plurality of 2D images. The second FOV value is greater than the first FOV value of each of the plurality of 2D images of the anatomical portion. A second 3D view is generated, based on the alignment of the generated first 3D view with 3D data associated with a pre-operative state of the anatomical portion.

Abstract: An interactive communication system includes at least one server, at least one base station for connecting to the server via a network, and at least one wearable device. The at least one wearable device includes a first sensor, a second sensor, a microcontroller, and a wireless communication module. The first sensor is configured for sensing a moving state of the at least one wearable device, and generating a corresponding first sensing signal. The second sensor is configured for sensing whether an object is contacting to or is close to the wearable device, and generating a corresponding second sensing signal. The microcontroller is configured for generating a first controlling signal according to the first sensing signal and the second sensing signal. The wireless communication module is configured for sending the first controlling signal to the least one server via the least one base station.

Abstract: Certain aspects of an apparatus and method for automatic ER/PR scoring of tissue samples may include for determining a cancer diagnosis score comprising identifying a positive stained nucleus in a slide image of the tissue sample, identifying a negative stained nucleus in the slide image, computing a proportion score based on number of the positive stained nucleus identified and number of the negative stained nucleus identified and determining the cancer diagnosis score based on the proportion.

Abstract: An inter-patient brain registration method for data normalization deformably aligns two brain images obtained from different patients even with tumor presence.

Abstract: Various aspects of a method and system to localize surgical tools during anatomical surgery are disclosed herein. In accordance with an embodiment of the disclosure, the method is implementable in an image-processing engine, which is communicatively coupled to an image-capturing device that captures one or more video frames. The method includes determination of one or more physical characteristics of one or more surgical tools present in the one or more video frames, based on one or more color and geometric constraints. Thereafter, two-dimensional (2D) masks of the one or more surgical tools are detected, based on the one or more physical characteristics of the one or more surgical tools. Further, poses of the one or more surgical tools are estimated, when the 2D masks of the one or more surgical tools are occluded at tips and/or ends of the one or more surgical tools.

Abstract: A 3D multiview reconstruction method takes a sequence of 2D stereo images from a narrow field-of-view imager (e.g., camera) and reconstructs a 3D representation of the wide field-of-view object or scene. The 3D multiview reconstruction method tracks 2D image pixels across neighboring frames and constraints for frame integration via 3D model construction.