Abstract: Provided is a disease prediction method and apparatus. The disease prediction apparatus may receive a medical image and clinical information, identify, from the medical image, quantitative data comprising a volume of at least one anatomical structure, and predict a disease occurrence based on the clinical information and the quantitative data. An artificial intelligence model may be used for each of identification of an anatomical structure and prediction of the disease occurrence. The disclosure was supported by the “AI Precision Medical Solution (Doctor Answer 2.0) Development” project hosted by Seoul National University Bundang Hospital (Project Serial No.: 1711151151, Project No.: S0252-21-1001).

Abstract: Disclosed is a medical image conversion method and apparatus. The medical image conversion apparatus trains a first artificial intelligence model to output a second contrast-enhanced image, based on first learning data including a pair of a first contrast-enhanced image and a first non-contrast image, and trains a second artificial intelligence model to output a second non-contrast image, based on second learning data including a pair of the first non-contrast image of the first learning data and the second contrast-enhanced image. The disclosure was supported by the “AI Precision Medical Solution (Doctor Answer 2.0) Development” project hosted by Seoul National University Bundang Hospital (Project Serial No.: 1711151151, Project No.: S0252-21-1001).

Abstract: A method and a device for converting a non-contrast image into a contrast image are disclosed. The image conversion device converts the non-contrast image into the contrast image by using a deep learning network trained with learning data including one or more contrast learning images and one or more non-contrast learning images. The disclosure was supported by the “Critical Care Patient Specialized Big Data Construction and AI-based CDSS Development” project hosted by Seoul National University Hospital (Task identification number: HI21C1074, Assignment number: HI21C1074050021).

Abstract: A medical image registration method and apparatus are provided. Upon receiving a medical image captured using a contrast agent, the medical image registration apparatus determines a phase of the medical image by using a classification model, and registers the medical image to a reference image by using a registration model. The disclosure refers to a technology developed through the “Korea Data and Software-Driven Hospitals Consortium 2.0)” project supervised by Seoul National University Bundang Hospital (task number: S0252-21-1001).

Abstract: Provided are a method and an apparatus for analyzing a body composition by using a medical image. The apparatus for analyzing the body composition trains a first artificial intelligence model by using training data labelling a body tissue region, such as muscle or fat, in a medical image for training, and then, segments a body tissue from an examination medical image by using the first artificial intelligence model and outputs body composition information based on a region, a volume, or a weight of the body tissue of the examination medical image. The disclosure is a technique developed through the Seoul Business Agency's 2020 support project for artificial intelligence (AI) technology industrialization (CY20053), “AI Verification and Industrialization of Computed Tomography (CT) Image-based Opportunistic Screening of Metabolic Syndrome, Osteoporosis, and Muscle Reduction.

Abstract: Provided are a method and apparatus for displaying a translucent image. An image display apparatus detects a plurality of voxels overlapping in a depth direction in three-dimensional (3D) image data, determines a first combination ratio for two colors based on transparency values of a first voxel and a second voxel, generates a first mixed color by combining a color of the first voxel with a color of the second voxel at the first combination ratio, and displays a mixed color on a screen.

Abstract: Disclosed are a method and device for analyzing human tissue on the basis of a medical image. A tissue analysis device generates training data including a two-dimensional medical image and volume information of tissue by using a three-dimensional medical image, and trains, by using the training data, an artificial intelligence model that obtains a three-dimensional size, volume, or weight of tissue by dividing at least one or more normal or diseased tissues from a two-dimensional medical image in which a plurality of tissues are displayed overlapping on the same plane. In addition, the tissue analysis device obtains a three-dimensional size, volume, or weight of normal or diseased tissue from an X-ray medical image by using the artificial intelligence model.

Abstract: A method and apparatus for anonymizing a three-dimensional medical image are provided. The apparatus determines a skin region of a three-dimensional medical image, generates a human mask based on a human tissue region of the three-dimensional medical image, the human tissue region including various organs, generates a skin expansion region in which the skin region of the three-dimensional medical image is expanded, generates an anonymization region obtained by removing a region corresponding to the human mask from the skin expansion region, and changes brightness values of voxels corresponding to the anonymization region in the three-dimensional medical image to a predefined value or an arbitrary value.

Abstract: A medical image registration method and apparatus are provided. The medical image registration apparatus generates a two-dimensional rigid image obtained by two-dimensionally projecting rigid tissue segmented from a three-dimensional medical image, generates a two-dimensional soft image obtained by two-dimensionally projecting soft tissue segmented from the three-dimensional medical image, registers a two-dimensional medical image with the two-dimensional rigid image, and registers the two-dimensional medical image with the two-dimensional soft image.

Abstract: Provided is a method and apparatus for reconstructing a medical image. The apparatus for reconstructing a medical image generates at least one base image by reducing a dimensionality of a three-dimensional (3D) medical image, generates at least one segmented image by reducing a dimensionality of a 3D image of a region of a tissue segmented from the 3D medical image or a 3D image of a region excluding the tissue from the 3D medical image, and trains, by using training data including the at least one base image and the at least one segmented image, an artificial intelligence (AI) model that separates at least one tissue from a medical image showing a plurality of tissues overlapping one another on the same plane.

Abstract: Provided is a method and apparatus for segmenting medical images. The apparatus sets a first seed group including voxels belonging to a segmentation target region among voxels of a medical image and a second seed group including voxels belonging to a remaining region thereof, assigns a weight to a link between a start node and an end node and a voxel node, and segments the medical image into two regions by cutting a link having a minimum weight in a shortest path in which a sum of weights of a path connecting the start node and the end node is minimum.

Abstract: Provided is a method and apparatus for segmenting medical images. The apparatus sets a first seed group including voxels belonging to a segmentation target region among voxels of a medical image and a second seed group including voxels belonging to a remaining region thereof, assigns a weight to a link between a start node and an end node and a voxel node, and segments the medical image into two regions by cutting a link having a minimum weight in a shortest path in which a sum of weights of a path connecting the start node and the end node is minimum.

Abstract: Provided is a method and apparatus for controlling a three-dimensional (3D) medical image. The apparatus renders a medical image into a 3D object displayed in a virtual space, and scales up, scales down, rotates, or moves the 3D object or displays a cross section thereof according to a control signal.