Abstract: A method for automated detection of landmarks from 3D medical image data using deep learning according to the present inventive concept, the method includes receiving a 3D volume medical image, generating a 2D intensity value projection image based on the 3D volume medical image, automatically detecting an initial anatomical landmark using a first convolutional neural network based on the 2D intensity value projection image, generating a 3D volume area of interest based on the initial anatomical landmark and automatically detecting a detailed anatomical landmark using a second convolutional neural network different from the first convolutional neural network based on the 3D volume area of interest.

Abstract: A device for recognizing an application in a mobile terminal includes: a coprocessor, configured to identify an image obtained by a signal processing device to obtain a first image category, and determine a first application corresponding to the first image category; and a main processor, configured to receive an identifier of the first application from the coprocessor, and activate the first application or prompt a user to download the first application. The device can relatively accurately determine, based on an image from an image sensor, an application that the user may want to use, and activate the application.

Abstract: A detection method and a device based on a laser radar, and a computer readable storage medium are disclosed. The detection method includes: obtaining scanning data of the laser radar; performing an algorithm splitting on a feature algorithm for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm; and performing a heterogeneous acceleration for the at least one sub-algorithm to process the scanning data and to obtain a processing result; and obtaining a detected position of an obstacle and a detected drivable area based on the processing result.

Abstract: A method for automated determination of a growth rate of an object in 3D data sets is described wherein the method may comprise: a first trained 3D detection deep neural network (DNN) determining one or more first VOIs in a current 3D data set and second VOIs in prior 3D data set, a VOI being associated with an abnormality; a registration algorithm, preferably a registration algorithm based on a trained 3D registration DNN, determining a mapping between the one or more first and second VOIs, the mapping providing for a first VOI in the current 3D data set a corresponding second VOI in the prior 3D data set; a second trained 3D segmentation DNN segmenting voxels of a first VOI into first voxels representing the abnormality and voxels of a corresponding second VOI into second voxels representing the abnormality; and, determining a first volume of the abnormality on the basis of the first voxels and a second volume of the abnormality on the basis of the second voxels and using the first and second volume to dete

Abstract: A system and method for head tracking for a head worn display (HWD) includes head-frame and platform-frame IMUs providing high-rate pose data of the wearer's head and a mobile platform. An optical tracker estimates an optical pose of a camera oriented at fiducials based on a comparison of 2D image data captured by the camera (and portraying the fiducials within the image) and the known location of each fiducial in a 3D marker frame. The pIMU also provides low-rate, high-integrity georeferenced pose data of the mobile platform (e.g., in an earth frame). The head tracker provides low-rate, high-integrity absolute head pose solutions (e.g., user head pose in the earth frame) based on the georeferenced pose data and current high-rate head/platform pose data, updating the absolute pose solutions with high-rate updates based on combined extended Kalman filter propagation of the high-rate head and platform pose data with refined optical pose data.

Abstract: Methods and systems are described that mark and/or correct margin lines and/or other features of dental sites. In one example a three-dimensional model of a dental site is generated from intraoral scan data of the dental site, the three-dimensional model comprising a representation of a preparation tooth. An image of the preparation tooth is received or generated. Data from the image is processed using a trained machine learning model that has been trained to identify margin lines of preparation teeth, wherein the trained machine learning model outputs a probability map comprising, for each pixel in the image, a probability that the pixel depicts a margin line. The three-dimensional model of the dental site is then updated by marking the margin line on the representation of the preparation tooth based on the probability map.

Abstract: An image captured by an imaging device in low light conditions may cause the captured image to have poor contrast, blurring, and otherwise not display one or more objects in the image clearly. According to various aspects, systems and methods are provided for enhancing images that are captured in low light conditions.

Abstract: A method of processing optical coherence tomography (OCT) scans through a subject's skin, the method comprising: receiving at least a plurality of OCT scans through the subject's skin (7), each scan representing an OCT signal in a slice through the subject's skin (7), with the plurality of OCT scans being spaced apart such that the OCT scans represent the OCT signal through a volume through the subject's skin; processing each OCT scan using a processor so as to so as to detect at least one three dimensional structure in the skin; classifying, using a processor, the structure as belonging to a class of a plurality of classes of structures; and outputting at least one of the structure and the class.

Abstract: Infrared (IR) vibrational scattering scanning near-field optical microscopy (s-SNOM) has advanced to become a powerful nanoimaging and spectroscopy technique with applications ranging from biological to quantum materials. However, full spatiospectral s-SNOM continues to be challenged by long measurement times and drift during the acquisition of large associated datasets. Various embodiments provide for a novel approach of computational spatiospectral s-SNOM by transforming the basis from the stationary frame into the rotating frame of the IR carrier frequency. Some embodiments see acceleration of IR s-SNOM data collection by a factor of 10 or more in combination with prior knowledge of the electronic or vibrational resonances to be probed, the IR source excitation spectrum, and other general sample characteristics.

Abstract: The present invention relates to an ultrasound imaging system comprising: an image processor configured to receive at least one set of volume data resulting from a three-dimensional ultrasound scan of a body and to provide corresponding display data, an anatomy detector configured to detect a position and orientation of an anatomical object of interest within the at least one set of volume data, a slice generator for generating a plurality of two-dimensional slices from the at least one set of volume data, wherein said slice generator is configured to define respective slice locations based on the results of the anatomy detector for the anatomical object of interest so as to obtain a set of two-dimensional standard views of the anatomical object of interest, wherein the slice generator is further configured to define for each two-dimensional standard view which anatomical features of the anatomical object of interest are expected to be contained, and an evaluation unit for evaluating a quality factor for e

Abstract: Systems and methods for determining position and orientation of a bone of an anatomical feature are described. These include the use of a wearable holder configured to be mounted about an outer-skin surface of the anatomical feature, such that the anatomical feature and the bone are positioned in fixed relation with respect to the wearable holder when the wearable holder is mounted about the anatomical feature. Reference marker arrays are fixedly mounted to the wearable holder, each being positioned on the wearable holder to identify a landmark of the bone within the wearable holder when the wearable holder is mounted to the anatomical feature. The position and orientation of the reference markers are trackable to determine position and orientation of the wearable holder in a reference coordinate system, thereby enabling position and orientation of the landmarks on the bone to be determined.

Abstract: Various embodiments of access control systems and fingerprint sensing systems are disclosed. One or more fingerprints of an authorized person are recorded in a fingerprint database together with a sequence of angular positions. The authorized person may subsequently gain access to a secured item by scanning the authorized person's finger or fingers in accordance with the sequence of angular positions. Various embodiments of fingerprint sensors for determining the angular position of a finger on the sensor are also disclosed.

Abstract: A system and associated method are provided for fulfillment of medication requisitions corresponding to contained medication units. Requisition fulfillment logic may be included to provide decision data to a patient care provider for use in selecting one of a plurality of different fulfillment sites to fill a given medication requisition. A requisition router may route the medication requisition to a selected one of the plurality of fulfillment sites. The fulfillment sites may provide medication requisition metadata (e.g., data relating to the preparation and handling of medication units) to a medication requisition database in corresponding relation to the corresponding medication requisitions fulfilled by the fulfillment sites. The medication requisition metadata may be stored in the medication requisition database and accessed to facilitate enhanced management functionalities in relation to medication units dispensed by patient care providers.

Abstract: Disclosed is a system and associated methods that account for the change in coloring or tint that some wavelengths of light have on materials of an object, and that generate an object model with the accounted for change in coloring or tint removed from the pixels or constructs of that model. The system receives spectral data in different electromagnetic spectrum bands for a particular surface of the object. The system measures a first quality of the light that illuminates the object, and determines a reactivity of the particular surface to the first quality of the light based on the spectral data matching a spectral signature of a material having that reactivity. The system removes the light effects on the particular surface by adjusting the spectral data according to the reactivity to the first quality of the light and measuring the first quality in the light illuminating the object.

Abstract: The embodiments of the present disclosure disclose method for calculating image highlight reflection separation. A specific implementation of the method includes: controlling a polarized color sensor to shoot a polarized color image, and calculating a polarized chromaticity image; clustering the pixels of diffuse reflection components whose chromaticity difference is within a preset range, to obtain a pixel sequence set; using a robust principal component analysis algorithm for each pixel, to obtain an initial separation result; further processing the polarized color image for highlight reflection separation, to obtain a highlight reflection separation result, and generating a diffuse reflection image. This implementation can achieve highlight reflection separation for images without being affected by the lighting information, and obtain a highlight reflection separation result with higher accuracy through the color and polarization information.

Abstract: Systems and methods described herein may provide a system that enables execution of workloads within a distributed computing environment that includes one or more edge computing devices and/or one or more user computing devices. A computing device may determine a workload for execution within a distributed computing environment that includes a plurality of computing devices. The computing device may determine valid execution slots within the distributed computing environment and may select a first execution slot of the valid execution slots. A first edge computing device may be assigned to execute the workload during the first execution slot. The first execution slot may be selected based on scores computed for the valid execution slots. The scores may be determined based on at least one criteria selected from the group consisting of (i) workload durations, (ii) carbon emissions, (iii) utility costs, and (iv) a disruption penalty.

Abstract: Disclosed are an information processing method and a terminal device. The method comprises: acquiring first information, wherein the first information is information to be processed by a terminal device; calling an operation instruction in a calculation apparatus to calculate the first information so as to obtain second information; and outputting the second information. By means of the examples in the present disclosure, a calculation apparatus of a terminal device can be used to call an operation instruction to process first information, so as to output second information of a target desired by a user, thereby improving the information processing efficiency. The present technical solution has advantages of a fast computation speed and high efficiency.

Abstract: Identifying insect species integrates image processing, feature selection, unsupervised clustering, and a support vector machine (SVM) learning algorithm for classification. Results with a total of 101 mosquito specimens spread across nine different vector carrying species demonstrate high accuracy in species identification. When implemented as a smart-phone application, the latency and energy consumption were minimal. The currently manual process of species identification and recording can be sped up, while also minimizing the ensuing cognitive workload of personnel. Citizens at large can use the system in their own homes for self-awareness and share insect identification data with public health agencies.

Abstract: Disclosed herein are systems and methods for of assessing stain titer levels. An exemplary method includes generating a set of field of views for the image or the region of the image, selecting field of views from the set of field of views that meet predefined criteria, creating a series of patches within each of the selected field of views, retaining patches from the series of patches that meet predefined criteria indicative of a presence of the stain for which the titer is to be estimated, deriving stain color features and stain intensity features pertaining to the stain from the retained patches, estimating a titer score for each of the retained patches based on the stain color features and the stain intensity features, and calculating a weighted average score for the titer of the stain based on the estimated titer score for each of the retained patches.

Abstract: The present disclosure provides a computer-implemented method for correlating at least one infrared 2D-image to a 3D-representation of at least a part of a tooth displayed in a graphical user-interface, of a hand-held scanning device, on a screen, comprising the steps of: obtaining a first set of 2D-images of the at least part of the tooth; forming a 3D-representation of the at least a part of the tooth from the first set of 2D-images; displaying, in the graphical user-interface, the 3D-representation; obtaining a second set of 2D-images, wherein the second set of 2D images are infrared 2D-images acquired within the at least part of the tooth; displaying, in the user-interface, at least one of the 2D-images from the second set of 2D-images; displaying, in the user-interface, a manipulator configured to change between 2D-images in the second set of 2D-images.

Abstract: A computer-implemented method includes generating, via a processor, synthetic vessels. The method also includes performing, via the processor, three-dimensional (3D) computational fluid dynamics (CFD) on the synthetic vessels for different flow rates to generate 3D CFD data. The method further includes extracting, via the processor, 3D image patches from the synthetic vessels. The method even further includes obtaining, via the processor, pressure drops across the 3D image patches from the 3D CFD data. The method yet further includes training, via the processor, a deep neural network utilizing the 3D image patches, the pressure drops, and associated flow rates to generate a trained deep neural network.

Abstract: A computational systems pathology spatial analysis platform includes: (i) a spatial heterogeneity quantification component configured for generating a global quantification of spatial heterogeneity among cells of varying phenotypes in multi-parameter cellular and subcellular imaging data; (ii) a microdomain identification component configured for identifying a plurality of microdomains for tissue samples based on the global quantification, each microdomain being associated with a tissue sample; and (iii) a weighted graph component configured for constructing a weighted graph for the multi-parameter cellular and subcellular imaging data, the weighted graph having a plurality of nodes and a plurality of edges each being located between a pair of the nodes, wherein in the weighted graph each node is a particular one of the microdomains and the edge between each pair of microdomains in the weighted graph is indicative of a degree of similarity between the pair of the microdomains.

Abstract: Platforms for enzymatic assays for biomarkers, including systems, methods, and measuring devices by which a biomarker, such as creatinine, is measured using a small amount of biological fluid, such as blood, plasma, or serum. The measuring device or biosensor can be a test strip including a layered active component assembly positioned between two outer layers which enables multi-step enzymatic reactions operating in kinetic and/or endpoint (in which the reaction is allowed to near completion), and generally includes multiple layers with primary enzyme(s), coupling enzyme(s), and reagents to produce an optical signal correlated to the concentration of a biomarker in the sample. The test strip can be read using a portable optical reader coupled to a smart phone or tablet.

Abstract: Provided are a medical document creation support apparatus, a method, and a non-transitory computer readable recording medium storing a program capable of efficiently correcting medical documents such as interpretation reports. A first display control unit displays a medical document including at least one of a plurality of findings indicating features related to abnormal shadows included in a medical image in a first display region on a display screen. A second display control unit displays a list of the plurality of findings in a second display region on the display screen so that each finding is correctable. A correction unit corrects the medical document according to a correction instruction for a designated finding in the list of the plurality of findings.

Abstract: Illustrative embodiments are directed to a method and apparatus for evaluating boundary detection in an image. A processed image is received, wherein a detected boundary of an image of an object is identified in the processed image. A Radon transform is applied to the processed image for a plurality of angles to form a processed image histogram corresponding to the detected boundary for each of the plurality of angles. The processed image histogram for each of the plurality of angles and a corresponding ground truth histogram for each of the plurality of angles is normalized to provide a normalized processed image histogram and a normalized ground truth histogram for each of the plurality of angles, wherein the ground truth histogram corresponds to a ground truth boundary of the object for a corresponding angle. An indication of the edges of the normalized processed image histogram for each of the plurality of angles is plotted to form a boundary detection evaluation visualization.

Abstract: Methods and systems are provided for generating mattes for input images. A neural network system is trained to generate a matte for an input image utilizing contextual information within the image. Patches from the image and a corresponding trimap are extracted, and alpha values for each individual image patch are predicted based on correlations of features in different regions within the image patch. Predicting alpha values for an image patch may also be based on contextual information from other patches extracted from the same image. This contextual information may be determined by determining correlations between features in the query patch and context patches. The predicted alpha values for an image patch form a matte patch, and all matte patches generated for the patches are stitched together to form an overall matte for the input image.

Abstract: A system to compare erythemal zones includes a multispectral camera to acquire a multispectral image of each compared zone and a computer processing device to retrieve the images acquired by the multispectral camera. The computer processing device includes a memory, a display, an input device and a microprocessor. The system further includes an image processing program configured to perform spectral de-mixing of the acquired images and isolating the erythemal component of the images, and a program to compare the images based on the intensity of the erythemal components of each image.

Abstract: Oral care based imaging computer-implemented systems and methods for determining perceived attractiveness of a facial image portion of at least one person depicted in a digital image. The method has the following steps: a) obtaining a digital image comprising at least one oral feature of at least one person, wherein the digital image includes a facial image portion of the at least one person, the facial image portion having both positive and negative attributes as defined by pixel data of the digital image; b) analyzing the facial image portion; c) generating an Attractiveness Score indicative of a perceived attractiveness of the facial image portion based on the analyzed facial image portion in the obtained digital image; d) further generating an image description that identifies at least one area in said facial image portion based on the Attractiveness Score; and e) presenting the image description to a use.

Abstract: A system and method for quantitatively assessing a press fit value (and provide a mechanism to evaluate optimal quantitative values) of any implant/bone interface regardless the variables involved including bone site preparation, material properties of bone and implant, implant geometry and coefficient of friction of the implant-bone interface without requiring a visual positional assessment of a depth of insertion. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements.

Abstract: An object is to provide an endoscopic image learning device, an endoscopic image learning method, an endoscopic image learning program, and an endoscopic image recognition device that appropriately learn a learning model for image recognition for recognizing an endoscopic image in which a treatment tool for an endoscope appears. The object is achieved by an endoscopic image learning device including an image generation unit and a machine learning unit. The image generation unit generates a superimposed image where a foreground image in which a treatment tool for an endoscope is extracted is superimposed on a background-endoscopic image serving as a background of the foreground image, and the machine learning unit performs the learning of a learning model for image recognition using the superimposed image.

Abstract: The invention provides for a medical imaging system (100, 400) comprising a memory (110) storing machine executable instructions (120) and a configured artificial neural network (122). The medical imaging system further comprises a processor (104) configured for controlling the medical imaging system. Execution of the machine executable instructions causes the processor to receive (200) magnetic resonance imaging data (124), wherein the magnetic resonance imaging data is BOLD functional magnetic resonance imaging data descriptive of a time dependent BOLD signal (1100) for each of a set of voxels. Execution of the machine executable instructions further causes the processor to construct (202) a set of initial signals (126) by reconstructing the time dependent BOLD signal for each of the set of voxels using the magnetic resonance imaging data.

Abstract: A system and method for quantitatively assessing, during bone preparation, an estimation of future a press fit value (and provide a mechanism to evaluate optimal quantitative values) of any implant/bone interface regardless the variables involved including bone site preparation, material properties of bone and implant, implant geometry and coefficient of friction of the implant-bone interface without requiring a visual positional assessment of a depth of insertion. The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a patent application and its requirements.

Abstract: Aspects of the present disclosure pertain to systems and methods for enhancing brightfield or darkfield images to better enable nucleus detection. In some embodiments, the systems and methods described herein are useful for identifying membrane stain biomarkers as well as nuclear/cytoplasm stain biomarkers in stained images of biological samples. In some embodiments, the presently disclosed systems and methods enable quick and accurate nucleus detection in stained images of biological samples, especially for original stained images of biological samples where the nuclei appear faint.

Abstract: A laser-surgical apparatus includes a laser emitting laser light with a narrowband wavelength distribution, a digital image sensor or a plurality of digital image sensors that separately record(s) different color channels that represent mutually different spectral wavelength distributions with in each case one maximum corresponding to a specific spectral color and produce(s) separate pieces of image information for the individual color channels, and a graphics module for combining the separate pieces of image information into one color image. In addition, it includes a manipulation module for suppressing the laser light in the color image, which makes it possible to electronically manipulate the pieces of image information of the color channel the spectral wavelength distribution of which has the largest overlap with the narrowband wavelength distribution of the laser light before the separate pieces of image information are combined or during the combining of the separate pieces of image information.

Abstract: A method of configuring images for display in a user interface can include accessing an image and information indicating locations of a plurality of features of interest within the image. The method can include determining a first tile arrangement indicating a first quantity of tiles associated with a first zoom level such that each of the first tiles including a portion of the image at a first downsampling such that the first tiles collectively represent the image. The method can include determining a second tile arrangement indicating a second quantity of tiles associated with a second zoom level. The method can include determining a zoom level associated with display of a portion of the image and one or more tiles associated with the portion of the image at the determined zoom level. The method can include generating the determined one or more tiles and displaying them in a browser.

Abstract: A document creation support apparatus comprising at least one processor, wherein the processor is configured to: acquire an image and a character string related to the image; extract at least one feature region included in the image; specify a specific region that is a region corresponding to a phrase included in the character string, in the feature region; and present information for supporting creation of a document including the character string based on a result of the specifying.

Abstract: The invention relates to tracking pieces (110) used in the context of tracking movements of hard tissue of a jaw. According to the invention, a tracking piece (110) to be used in such context is designed to be patient-specific by including in the design of the tracking piece (110) a surface model representing a part of a person's intraoral anatomy, which surface model is generated based on imaging results acquired by imaging the anatomy.

Abstract: A medical image processing device includes a storage and a processor. The storage stores notification setting information for setting a notification content regarding which method a notification is performed in accordance with lesion information indicating content of a lesion, part information representing a part, or procedure information representing a procedure performed at a time of acquisition of a medical image. The processor is configured to acquire the medical image including a subject; perform image analysis processing on the medical image; select use-notification setting information from the notification setting information by using the lesion information, the part information or the procedure information; and perform the notification to notify a user of a result of the image analysis processing with different methods of the notifications for different lesion information, part information or procedure information according to the use-notification setting information.

Abstract: The present specification discloses a method capable of automatically recognizing an accurate wound boundary and a method of generating a 3D wound model based on the recognized wound boundary. The method of automatically recognizing a wound boundary according to the present specification is a method of automatically recognizing a wound boundary based on artificial intelligence, and may photograph several frames of the wound to be recognized with an RGB-D camera, separating measurement information in an image, amplifying image data for learning, and passing the amplified image data through an artificial neural network. The method may include generating a three-dimensional (3D) model by performing boundary recognition post-processing on the data passing through the artificial neural network to match a two-dimensional (2D) image with the 3D model.

Abstract: Embodiments of the present specification disclose a system and method for generating a vehicle damage image on the basis of a GAN model. During operation, the system obtains a real vehicle image, generates an intermediate image based on the real vehicle image by labeling a target box on the real vehicle image and removing a portion of the real vehicle image within the target box, and generates the vehicle damage image based on the intermediate image by inputting the intermediate image into a machine-learning model, which outputs the vehicle damage image by filling a local image indicating vehicle damage into the target box of the intermediate image.

Abstract: Directional propagation editing techniques are described, in one example, a digital image, a depth map, and a direction are obtained by an image editing system. The image editing system then generates features. To do so, the image editing system generates features from the digital image and the depth map for each pixel based on the direction, e.g., until an edge of the digital image is reached. In an implementation, instead of storing a value of the depth directly, a ratio is stored based on a depth in the depth map and a depth of a point along the direction. The image editing system then forms a feature volume using the features, e.g., as three dimensionally stacked features. The feature volume is employed by the image editing system as part of editing the digital image to form an edited digital image.

Abstract: Provided are apparatuses, a non-transitory computer-readable medium or media, and methods for supporting predicting of vascular disease using a fundus image of a subject. In certain aspects, disclosed a method including the steps of: extracting a feature information from a first fundus image of the subject based on a machine learning model; generating a second fundus image having a feature which is corresponding to the feature information by mapping a saliency factor to the first fundus image; and displaying the first fundus image and the second fundus image on a display device.

Abstract: There is provided a radiotherapy system, a data processing method and a storage medium. The radiotherapy system includes a calibration database, a training module and a data processing module. The calibration database can acquire a plurality of the sample data generated in a clinical therapy process. The training module can perform deep learning on the plurality of sample data stored in the calibration database to obtain a therapy algorithm model. The data processing module can process the detection data received in the clinical therapy process according to the therapy algorithm model to generate preliminary therapy data. As the therapy algorithm model is obtained by performing deep learning on a large amount of clinical data, the reliability of the therapy algorithm model is high. Accordingly, the accuracy of the preliminary therapy data generated by the therapy algorithm model is also high.

Abstract: Disclosed are a computer-implemented method, a system and a computer program product for feature processing. In the computer-implemented method for feature processing, two input features selected from multiple features of each sample in a sample set are projected to one resulting feature by one or more processing units based on a specified curve. The sample set is updated by replacing the two input features with the one resulting feature for each sample in the sample set by one or more processing units. The projecting and the updating for the sample set are repeated by one or more processing units until the number of features of each sample in the sample set reaches a predetermined criterion.

Abstract: Systems and methods for scan preparation are provided. The systems may obtain a first parameter set of a subject to be scanned by a medical device acquired before a scan is performed on the subject. The first parameter set may relate to a physiological motion of the subject acquired before the scan is performed on the subject. The systems may predict, based on the first parameter set and an estimation model, a second parameter set of the subject. The second parameter set may relate to the physiological motion of the subject. The systems may determine at least one scan parameter for the scan based at least in part on the second parameter set. The systems may cause the medical device to perform the scan on the subject based on the at least one scan parameter.

Abstract: The present invention provides a graph-based prostate diagnosis network (GPD-Net) and a method for using the same to predict a prostate health status of a patient from a 3D magnetic resonance imaging (MRI) scan containing a plurality of 2D MRI slices. The GPD-Net only demands patient-level annotations of MRI scan for training by formulating the diagnosis task of 3D prostate MRI scan in a multi-instance learning (MIL) strategy, and regarding each 2D MRI slice in the 3D prostate MRI scan as an instance. The GPD-Net includes a plurality of importance-guided graph convolutional layers to explore the diagnostic information with the importance-based topology. The present invention provides accurate prediction of prostate diseases and achieve more reliable diagnosis from MRI scans, therefore can effectively alleviate the workload of clinician in viewing the slices of MRI scan.

Abstract: An excrement judgement system according to the embodiment including: an acquisition unit that acquires image information on an image in which excrement excreted in a defecation action per time is captured; a determination unit that determines properties of a plurality of stools included in the image information and stool amounts corresponding to the properties of the plurality of stools; and an outputting unit that associates the properties of the plurality of stools and the stool amounts corresponding to the properties of the plurality of stools that are determined by the determination unit, and outputs the associated properties and stool amounts to a device to be displayed.

Abstract: The present disclosure relates to a system and method (10) for monitoring a set-up for manufacture and/or setting up for manufacture and/or tearing down after manufacture of a biopharmaceutical product. The method comprises: processing (S2, S3, S4) at least one image of a scene comprising the set-up for manufacture of the biopharmaceutical product. The processing of the at least one image comprises performing (S2) a first process on the at least one image for classifying first objects in the image, said first objects being devices such as clamps, pumps, valves and/or sensors and/or any other bio processing equipment. The first process comprising identifying, localizing and classifying the first objects in the image. A second process is performed (S3) on the at least one image for identifying and localizing connections in the images.

Abstract: Embodiments of the present disclosure provide a method and system for co-operative and cascaded inference on the edge device using an integrated Deep Learning (DL) model for object detection and localization, which comprises a strong classifier trained on largely available datasets and a weak localizer trained on scarcely available datasets, and work in coordination to first detect object (fire) in every input frame using the classifier, and then trigger a localizer only for the frames that are classified as fire frames. The classifier and the localizer of the integrated DL model are jointly trained using Multitask Learning approach. Works in literature hardly address the technical challenge of embedding such integrated DL model to be deployed on edge devices. The method provides an optimal hardware software partitioning approach for components or segments of the integrated DL model which achieves a tradeoff between latency and accuracy in object classification and localization.

Abstract: Image co-registration entails: emitting energy, and responsively and dynamically acquiring images, the acquired images having a given dimensionality; selecting, repeatedly, and dynamically with the acquiring, from among the acquired images and, as a result of the selecting, changing, repeatedly, and dynamically, membership in a set of images of the given dimensionality; and, synchronously with the change, dynamically and iteratively registering the set to an image having a dimensionality higher than the given dimensionality. The co-registration is realizable with an imaging probe for the acquiring, and a tracking system for tracking a location, and orientation, of the probe, the registering being specialized for the acquiring with the probe being dynamically, during the acquiring, any one or more of angulated, rotated and translated.