Provided is an image processing device including a matching unit that performs matching processing between a predetermined pattern on a surface of a 3D model of a biological tissue including an operating site generated on the basis of a preoperative diagnosis image and a predetermined pattern on a surface of the biological tissue included in a captured image during surgery, a shift amount estimation unit that estimates an amount of deformation from a preoperative state of the biological tissue on the basis of a result of the matching processing and information regarding a three-dimensional position of a photographing region which is a region photographed during surgery on the surface of the biological tissue, and a 3D model update unit that updates the 3D model generated before surgery on the basis of the estimated amount of deformation of the biological tissue.

Methods of the disclosure may include obtaining a first set of medical images at a first time point and a second set of medical images at a second time point, each set including at least two medical images. First and second algorithms may be used to calculate, respectively, first and third brain volume (BV) values at the first time point based on two or more images from the first set of medical images and second and fourth BV values at the second time point based on two or more images from the second set of medical images. A mathematical weight may be applied to at least one of the first, second, third, or fourth BV values. The first and third BV values may be averaged, and the second and fourth BV values may be averaged to determine overall BV values at the first and second time points, respectively.

System for performing fully automatic brain tumor and tumor-aware cortex reconstructions upon receiving multi-modal MRI data (T1, T1c, T2, T2-Flair). The system outputs imaging which delineates distinctions between tumors (including tumor edema, and tumor active core), from white matter and gray matter surfaces. In cases where existing MRI model data is insufficient then the model is trained on-the-fly for tumor segmentation and classification. A tumor-aware cortex segmentation that is adaptive to the presence of the tumor is performed using labels, from which the system reconstructs and visualizes both tumor and cortical surfaces for diagnostic and surgical guidance. The technology has been validated using a publicly-available challenge dataset.

Systems and methods for image segmentation are provided. The systems may obtain a target image and a template image relating to the target image. The template image may correspond to an initial mask reflecting initial segmentations of the template image. The systems may determine a first transformation and an intermediate template image by preliminarily registering the template image to the target image and generate an intermediate mask based on the initial mask and the first transformation. The systems may determine, based on the intermediate mask, one or more first regions from the target image and one or more second regions from the intermediate template image. The systems may determine a second transformation by registering each of the one or more second regions to a corresponding first region. The systems may determine a target mask according to which the target image can be segmented based on one or more second transformations.

Methods and apparatus for predicting performance of an individual on a task, the method comprises receiving brain imaging data for the individual, wherein the brain imaging data comprises structural brain data, determining values for at least one characteristic of the structural brain data within regions of interest defined for a population of individuals having different performance levels, and predicting based on the determined values, a performance potential of the individual.

A method and system to treat headaches in a patient by performing surgery via at least one nostril. Data from a computer tomography scan of at least one nasal cavity and one sinus cavity of the patient and a completed headache questionnaire are matched to at least one nasal/sinus surgery plan to operate on at least one of: a nasal septum, at least one sinus cavity and at least one turbinate of the patient. The surgery plan is executed by installing a topical local anesthetic and decongestant onto the at least one turbinate forming an anesthetized decongested nasal cavity; infusing an anesthetic into the anesthetized decongested nasal cavity of the patient; dilating the at least one sinus ostium; incising at least one of: a first mucosal flap or a second mucosal flap of the nasal septum of the anesthetized decongested nasal cavity to expose deviated septal cartilage and bone; removing deviated cartilage and/or bone of the nasal septum; fracturing the at least one turbinate laterally away from the nasal septum; inspecting between the first mucosal flap and the second mucosal flap for a residual broken bone, a residual segment of cartilage or combinations thereof, surgically closing the first mucosal flap and the second mucosal flap of the nasal septum; and suctioning unwanted matter from the anesthetized decongested nasal cavity. An interactive system guides the surgery and provides a record thereof.

The following relates generally to accelerated magnetic resonance imaging (MRI) reconstruction. In some embodiments, a MRI machine learning algorithm is trained based on reference MRI data in non-Cartesian k-space. During the training, at each iteration of a plurality of iterations: (i) a non-Cartesian sampling trajectory ω may be optimized under the physical constraints, and/or (ii) an image reconstructor may be jointly iteratively optimized. Examples of the image reconstructor include a convolutional neural network (CNN) denoiser, a model-based deep learning (MoDL) image reconstructor, iterative image reconstructor, a regularizer, and an invertible neural network.

In order to reduce a radiation dose delivered to an object or an observer, a facility for monitoring handling of the object has an optical unit configured to direct ionizing radiation onto the object and also a filter element in order to attenuate a part of the ionizing radiation. An imaging unit may detect portions of the ionizing radiation passing through the object in order to create an image of the object. A view acquisition system may acquire a viewing movement, and a control unit is configured, during a first operating mode, to control a position of the filter element as a function of the viewing movement. The control unit is configured to identify a predefined sequence of viewing movements and, as a function thereof, to switch into a second operating mode. The position of the filter element is controlled during the second operating mode as a function of an image analysis.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining an artificial neural network architecture corresponding to a sub-graph of a synaptic connectivity graph. In one aspect, there is provided a method comprising: obtaining data defining a graph representing synaptic connectivity between neurons in a brain of a biological organism; determining, for each node in the graph, a respective set of one or more node features characterizing a structure of the graph relative to the node; identifying a sub-graph of the graph, comprising selecting a proper subset of the nodes in the graph for inclusion in the sub-graph based on the node features of the nodes in the graph; and determining an artificial neural network architecture corresponding to the sub-graph of the graph.

A neuroimaging-based approach to predict treatment response in mental disorders by acquiring and analysing brain PET dopamine measures from patients. The method uses a short, simplified protocol for [18F]FDOPA brain PET imaging adapted for clinical practice. Individual [18F]FDOPA brain PET data are then quantified with a fully-automated analysis pipeline to extract information on the dopamine function of the subject. This information coupled with clinical information is run through a prediction algorithm to identify those patients whose illness will not respond to conventional antipsychotics.