Abstract: A medical diagnostic method according to an embodiment includes receiving a plurality of slide images of a three-dimensional (3D) medical image obtained by capturing an object and clinical data for the object, extracting one or more slide images including a lesion among the plurality of slide images, extracting a region of interest (ROI) including the lesion from each of the extracted one or more slide images, generating latent space data including a feature of an image of the extracted ROI using a pre-trained deep learning-based ROI abstraction model, and generating a diagnostic result for the object from the latent space data and the clinical data using a pre-trained deep learning-based medical diagnostic model.

Abstract: An apparatus for learning of generative model according to an embodiment includes an encoder configured to extract a feature from input data and output a feature vector, a decoder configured to restore the input data on the basis of the feature vector, and a domain module configured to generate domain information to be learned through a generative model into domain vector blocks, each with a size corresponding to the feature vector, concatenate the feature vector and the domain vector block, and input a concatenated vector to the decoder.

Abstract: A method for image analysis according to an embodiment may include generating a prediction result for an original image using a pre-trained image classification model, learning a plurality of masks using the original image, the prediction result, and the image classification model, and generating a map visualizing a importance of each area of the original image for the prediction result based on at least one of the plurality of masks.

Abstract: An image processing apparatus includes a superpixel extractor configured to extract a plurality of superpixels from an input original image, a backbone network including N feature extracting layers (here, N is a natural number of two or more) which divide the input original image into grids including a plurality of regions and generate an output value including a feature value for each of the divided regions, and a superpixel pooling layer configured to generate a superpixel feature value corresponding to each of the plurality of superpixels using a first output value to an Nth output value output from each of the N feature extracting layers.

Abstract: An apparatus for transmitting content includes a content request transmitter to transmit a request message for content requested by a client device to content servers, a response delay time calculator to calculate a response delay time of each content server on the basis of a response message of each content server for the request message, a byte range determiner to determine one or more content servers to which a byte range request for the content is to be transmitted, from among the content servers, on the basis of the response message and/or the response delay time and determine a size of a byte range request and the number of byte range request for each content server, and a content acquirer to acquire the content from the content servers through the byte range request based on the size of a byte range request and the number of byte range requests.

Abstract: According to aspects of the present disclosure, a change detection method for ROBOT process automation (RPA) performed by a computer device includes obtaining object information of a target object from first working environment information at a first time, and determining a comparison object detected in second working environment information at a second time as an object corresponding to the target object based on the obtained object information of the target information. Thus, even if an error occurs due to a change in a working environment, an RPA apparatus can operate normally by correcting the error by itself without the need for a user to recreate a new bot or modify the existing bot. Also, even if the position or shape of an object referred to by RPA in a screen is changed, the RPA apparatus can accurately find an object to manipulate based on surrounding information and perform RPA normally.

Abstract: Provided is a method of managing a target process. The method performed by a process management apparatus includes: generating a reference pattern indicating a normal state based on reference observed data on a process factor measured while the target process is maintained in the normal state; obtaining observed data on the process factor measured for a specified observation period; calculating a dissimilarity between the reference pattern and the observed data; and constructing a regression tree for the target process by using the observed data and the dissimilarity, wherein the process factor is set as an independent variable of the regression tree, and the dissimilarity is set as a dependent variable of the regression tree.

Abstract: A method according to an aspect of the disclosure is performed by a computing device and includes rendering a video frame, displaying an object generated by user manipulation on the rendered video frame, calculating a relative position of the object with respect to a reference point of the video frame, and transmitting object information generated based on the relative position of the object. According to the method, even if the streaming screen continuously changes when streaming video between remote terminals, the shared 3D object can be placed at an accurate position, and objective position information of the shared 3D object can be provided regardless of the surrounding environment or situation.

Abstract: A method for performing an operation according to one embodiment includes performing a homomorphic operation using one or more ciphertexts that are homomorphically encrypted based on an encryption key, determining a count value for a ciphertext generated through the homomorphic operation based on count values for each of the one or more ciphertexts, requesting a key management apparatus, which holds the encryption key and a decryption key corresponding to the encryption key, to re-encrypt the generated ciphertext based on the determined count value, acquiring, from the key management apparatus, a ciphertext generated by re-encrypting the generated ciphertext through decryption based on the decryption key and encryption based on the encryption key; and determining a count value for the acquired ciphertext to be a preset initial value.

Abstract: A ciphertext comparison method uses homomorphic encryption. An apparatus performs the ciphertext comparison method using the homomorphic encryption. The ciphertext comparison method using homomorphic encryption includes acquiring a first ciphertext obtained by encrypting first real-number data using a homomorphic evaluation algorithm and a second ciphertext obtained by encrypting second real-number data using the homomorphic evaluation algorithm, and generating a ciphertext for a result for comparison in size between the first real-number data and the second real-number data using homomorphic evaluation for the first ciphertext, the second ciphertext, and an approximation function for approximating a signum function.

Abstract: Provided are search methods and apparatuses capable of defining a search result screen in a simple and easy way. According to an aspect of the present disclosure, there is provided a search method performed by a computing device. The search method comprises receiving a search term and automatically displaying a first graphic user interface (GUI) object configured to receive a search result screen configuration scheme in response to a first user input for the received search term without an additional user input after the first user input. The first unit input is a user input other than a user input editing the search term.

Abstract: Methods for recommending next user input using pattern analysis of user input is provided. According to an aspect of the present disclosure, a method comprising obtaining information on a series of user inputs entered through a graphic user interface (GUI), analyzing the information on the series of user inputs to identify a pattern formed by the series of user inputs, and when the pattern is identified, automatically displaying next input recommendation information determined depending on the identified pattern and a last user input of the series of user inputs without additional user input after the series of user inputs, is provided.

Abstract: A decentralized key management system according to an embodiment of the present disclosure includes a bootstrap for generating a key and obtaining a certificate corresponding to the generated key, a memory for receiving the key and the certificate from the bootstrap and storing the key and the certificate, a container, in response to a mount command of the bootstrap, for reading the key and the certificate from the memory and being mounted with the key and the certificate, and a controller for generating the bootstrap, and deleting the bootstrap after the container mounts the key and the certificate.

Abstract: Disclosed are an acoustic echo cancelling apparatus and an acoustic echo cancelling method. The acoustic echo cancelling apparatus comprises a first capturer configured to capture a remote audio signal received from a remote machine; a second capturer configured to capture a local audio signal that is input to a local microphone; an acoustic echo canceller (AEC) configured to cancel an acoustic echo of the local audio signal using the remote audio signal and output a local audio signal from which the acoustic echo is cancelled; and a transmitter configured to transmit an output signal of the AEC to the remote machine.

Abstract: A method of sharing information on the basis of anchoring and an anchoring device supporting the same and more particularly are provided. One of the methods includes, acquiring anchoring information including first field information permitted for sharing from a target transaction record recorded in a first blockchain and recording the acquired anchoring information in a second blockchain.

Abstract: Provided is a method of managing a target process. The method performed by a process management apparatus includes: generating a reference pattern indicating a normal state based on reference observed data on a process factor measured while the target process is maintained in the normal state; obtaining observed data on the process factor measured for a specified observation period; calculating a dissimilarity between the reference pattern and the observed data; and constructing a regression tree for the target process by using the observed data and the dissimilarity, wherein the process factor is set as an independent variable of the regression tree, and the dissimilarity is set as a dependent variable of the regression tree.

Abstract: Provided is a door lock mortise. The door lock mortise includes a structure in which a latch fixing member moves between a latch operating member and a latch cover member when a dead bolt is locked or unlocked. According to the embodiments of the present invention, the door lock mortise including the latch fixing member to have a structural characteristic capable of addressing a problem in that a latch bolt may be moved when the dead bolt is unlocked to release a door lock so that an impact is transmitted only to the dead bolt in a state in which the dead bolt is not unlocked and durability is reduced is disclosed.

Abstract: An apparatus for extracting object information according to one embodiment includes: a padding image generator for generating a padding image including an original image; a partial image acquirer for acquiring a plurality of partial images of the padding image; an object classification result acquirer for acquiring an object classification result for each of the plurality of partial images using an object classification model; a confidence map generator for generating a confidence map having a size the same as that of the padding image and including a confidence value on the basis of the object classification result; and an object information acquirer for acquiring information on an object in the padding image on the basis of the confidence map.

Abstract: A method for serving cloud of quantum computing according to an embodiment may include receiving a request to perform quantum computing from a client and providing a standard resource description to the client, receiving, from the client, resource-related information corresponding to the standard resource description, providing a standard application program interface (API) to the client, and receiving, from the client, standard quantum computing code created through the standard API, selecting quantum computing hardware, which is to execute the standard quantum computing code, among a plurality of quantum computing hardware, based on one or more among the resource-related information and the standard quantum computing code, converting the standard quantum computing code into quantum computing code executable on the selected quantum computing hardware, and executing the converted quantum computing code using the selected quantum computing hardware.

Abstract: A confidence measurement method according to an embodiment includes additionally training a feature extractor of a classification model trained using training data including a plurality of images, using the training data, such that feature vectors for images labeled with the same class among feature vectors for the plurality of images become closer to each other in an embedding space for the feature extractor, and measuring confidence of a classification result for an input image by the trained classification model using the additionally trained feature extractor.