Abstract: A deep-learning method and systems for analyzing optical coherence tomography (OCT) images based on a convolutional neural network are provided. The method includes extracting a feature from one or more three-dimensional OCT volumetric scan images and classifying the OCT images with respect to diabetic macular edema (DME) based on results the feature extracted. The step of extracting a feature from one or more three-dimensional OCT volumetric scan images is performed by a neural network such as a neural network based on a ResNet-34 architecture. The method can further include extracting a feature from one or more two-dimensional (2D) OCT B-scan images and classifying the OCT images with respect to DME based on results of the 2D feature extracted. The step of extracting a feature from one or more 2D OCT B-scan images is performed by a neural network such as a neural network based on a ResNet-18 architecture.

Abstract: To improve the performance and accuracy of an image segmentation neural network, a cascaded robust learning framework for the segmentation of noisy labeled images includes two stages: a sample selection stage, and a joint optimization stage with label correction. In the first stage, the clean annotated samples are selected for network updating, so that the influence of noisy sample can be interactively eliminated. In the second stage, the label correction module works together with the joint optimization scheme to revise the imperfect labels. Thus, the training of the whole network is supervised by the corrected labels and the original ones.

Abstract: To improve the performance and accuracy of an image segmentation neural network, a cascaded robust learning framework for the segmentation of noisy labeled images includes two stages: a sample selection stage, and a joint optimization stage with label correction. In the first stage, the clean annotated samples are selected for network updating, so that the influence of noisy sample can be interactively eliminated. In the second stage, the label correction module works together with the joint optimization scheme to revise the imperfect labels. Thus, the training of the whole network is supervised by the corrected labels and the original ones.

Abstract: This application discloses a method and apparatus for detecting a molecule binding site, an electronic device, and a storage medium, relating to the field of computer technologies. According to one embodiment, the method includes: obtaining 3D coordinates of at least one site in a target molecule; for each site, obtaining a prediction probability indicating a possibility of the each site being a binding site via a site detection model; and determining a binding site from the at least one site in the target molecule based on the prediction probability of the each of the at least one site.

Abstract: A surgery simulation system and method is disclosed. The surgery simulation system includes a simulation platform provided with a physical model thereon for simulating a human organ for incising and/or piercing; a force feedback device provided with an actuator and a surgical instrument connection component thereon, the surgical instrument connection component being detachably connected to a surgical instrument; a three-dimensional tracker for tracking the surgical instrument; a display device for displaying a virtual organ and/or a medical image; a control module having an electrical signal connection with the force feedback device, the three-dimensional tracker and the display device, controlling the force feedback device according to the signal of the three-dimensional tracker, and displaying a corresponding image on the display device. The surgery simulation system and method vividly simulates the scene of a real operation so as to assist medical staff to quickly master surgical skills.

Abstract: In a computer-based graphics imaging system, binocular tone mapping is employed with an optimization-based framework to generate a binocular single vision image with more visual richness than a single tone-mapped image, the binocular tone mapping being guided by a binocular viewing comfort predictor (BVCP) metric in order to assure stable formation of binocularly tone-mapped image pairs, so that visual information content of the image pair can be maximized without triggering visual discomfort.

Abstract: Disclosed are methods, systems and apparatuses for detection of arterial input function (AIF) in MRI, specially DCE MR images, comprising automatic selection of AIF based on affinity propagation (AP) clustering method.

Abstract: The disclosure relates to apparatus and methods for processing images. The method for generating a manga-style image from an input image comprises: generating a line drawing from the input image; generating a manga-style screening from an input image; and overlaying the line drawing and the manga-style screening to generate the manga-style image. The method can provide a high efficient way for auto-generating manga images.

Abstract: In a computer-based graphics imaging system, binocular tone mapping is employed with an optimization-based framework to generate a binocular single vision image with more visual richness than a single tone-mapped image, the binocular tone mapping being guided by a binocular viewing comfort predictor (BVCP) metric in order to assure stable formation of binocularly tone-mapped image pairs, so that visual information content of the image pair can be maximized without triggering visual discomfort.

Abstract: Disclosed are methods, systems and apparatuses for detection of arterial input function (AIF) in MRI, specially DCE MR images, comprising automatic selection of AIF based on affinity propagation (AP) clustering method.

Abstract: The disclosure relates to apparatus and methods for processing images. The method for generating a manga-style image from an input image comprises: generating a line drawing from the input image; generating a manga-style screening from an input image; and overlaying the line drawing and the manga-style screening to generate the manga-style image. The method can provide a high efficient way for auto-generating manga images.

Abstract: A method and a system for estimating a longitudinal relaxation time in magnetic resonance imaging. The method includes scanning at least one object to form a sequence of data with a plurality of flip angles; regressing linearly the formed sequence of data based on a first signal intensity model associated with the flip angles to obtain an initial estimation for said longitudinal relaxation time; and regressing nonlinearly the formed sequence of data based on a second signal intensity model associated with the flip angles so as to obtain a final estimation for the longitudinal relaxation time, in which the initial estimation is used as an initial guess for the longitudinal relaxation time based on the second signal intensity model.

Abstract: A computer-based method and system for interactive volume rendering of a large volume data on a conventional personal computer using hardware-accelerated block filtration optimizing uses 3D-textured axis-aligned slices and block filtration. Fragment processing in a rendering pipeline is lessened by passing fragments to various processors selectively in blocks of voxels based on a filtering process operative on slices. The process involves generating a corresponding image texture and performing two-pass rendering, namely a virtual rendering pass and a main rendering pass. Block filtration is divided into static block filtration and dynamic block filtration. The static block filtration locates any view-independent unused signal being passed to a rasterization pipeline. The dynamic block filtration determines any view-dependent unused block generated due to occlusion. Block filtration processes utilize the vertex shader and pixel shader of a GPU in conventional personal computer graphics hardware.

Abstract: A computer-based method and system for interactive volume rendering of a large volume data on a conventional personal computer using hardware-accelerated block filtration optimizing uses 3D-textured axis-aligned slices and block filtration. Fragment processing in a rendering pipeline is lessened by passing fragments to various processors selectively in blocks of voxels based on a filtering process operative on slices. The process involves generating a corresponding image texture and performing two-pass rendering, namely a virtual rendering pass and a main rendering pass. Block filtration is divided into static block filtration and dynamic block filtration. The static block filtration locates any view-independent unused signal being passed to a rasterization pipeline. The dynamic block filtration determines any view-dependent unused block generated due to occlusion. Block filtration processes utilize the vertex shader and pixel shader of a GPU in conventional personal computer graphics hardware.

Abstract: A computer-based method and system for interactive volume rendering of a large volume data on a conventional personal computer using hardware-accelerated block filtration optimizing uses 3D-textured axis-aligned slices and block filtration. Fragment processing in a rendering pipeline is lessened by passing fragments to various processors selectively in blocks of voxels based on a filtering process operative on slices. The process involves generating a corresponding image texture and performing two-pass rendering, namely a virtual rendering pass and a main rendering pass. Block filtration is divided into static block filtration and dynamic block filtration. The static block filtration locates any view-independent unused signal being passed to a rasterization pipeline. The dynamic block filtration determines any view-dependent unused block generated due to occlusion. Block filtration processes utilize the vertex shader and pixel shader of a GPU in conventional personal computer graphics hardware.

Abstract: A computer-based method and system for interactive volume rendering of a large volume data on a conventional personal computer using hardware-accelerated block filtration optimizing uses 3D-textured axis-aligned slices and block filtration. Fragment processing in a rendering pipeline is lessened by passing fragments to various processors selectively in blocks of voxels based on a filtering process operative on slices. The process involves generating a corresponding image texture and performing two-pass rendering, namely a virtual rendering pass and a main rendering pass. Block filtration is divided into static block filtration and dynamic block filtration. The static block filtration locates any view-independent unused signal being passed to a rasterization pipeline. The dynamic block filtration determines any view-dependent unused block generated due to occlusion. Block filtration processes utilize the vertex shader and pixel shader of a GPU in conventional personal computer graphics hardware.