Abstract: An anomaly detection method by which a computer performs anomaly detection includes: obtaining first feature data outputted through N (N is an integer not less than 1) convolutional layers of a convolutional neural network configured as an encoder when an image is inputted to the convolutional neural network; obtaining second feature data outputted through M (M is an integer not less than 1, and M?N) convolutional layers of the convolutional neural network and different in size from the first feature data; and performing anomaly detection on the image by using features indicated by the first feature data and the second feature data that are different in size.

Abstract: A training method performed through batch learning by a computer includes: obtaining training data including first time-series data and second time-series data different from the first time-series data; performing first training processing of training a neural process (NP) model, which outputs, using a stochastic process, a prediction result that takes uncertainty into account, to predict first and second time-series data distributions, based on the first time-series data and second time-series data; and performing, using a contrastive learning algorithm, second training processing of (i) training the NP model to bring close to each other first sampling data items generated by sampling from the first time-series data distribution, (ii) training the NP model to bring close to each other second sampling data items generated by sampling from the second time-series data distribution, and (iii) training the NP model to push away the first and second sampling data items far from each other.

Abstract: A training method includes: generating a first image by adding noise to a first area; generating a second image by adding noise to a second area; generating a combined image by weighted addition of the first image and the second image; generating a first training label for the first image; generating a second training label for the second image; generating a combined training label by weighted addition of the first training label and the second training label; and generating a learning model by machine learning using the combined image and the combined training label.

Abstract: A risk prediction method is executed by a computer of a risk predictor using a convolutional neural network. The method includes making the convolutional neural network acquire an input image taken by an in-vehicle camera installed on a vehicle. The method also includes making the convolutional neural network estimate a risk area and a feature of the risk area, the risk area being in the acquired input image. The risk area has a possibility that a moving object may appear from the risk area into a travelling path of the vehicle, and the moving object may collide with the vehicle in a case where the vehicle simply continues running. The method further includes making the convolutional neural network output the estimated risk area and the estimated feature of the risk area as a risk predicted for the input image.

Abstract: A method for controlling an information terminal causes a computer of the information terminal to receive, from a case retrieval system, a plurality of similar medical images having a feature quantity of a region of interest and a certain degree of similarity in accordance with the region of interest included in a target medical image, displays a display screen displaying the plurality of received similar medical images on a touch panel display, the display screen including a display region in which at least some of the plurality of received similar medical images are displayed, displays, if selection of a first similar medical image from among the at least some of the plurality of received similar medical images displayed in the display region is detected, the first similar medical image across the display region, and displays, if a swipe operation performed on the first similar medical image is detected, a second similar medical image, which has second highest similarity next to the first similar medical im

Abstract: A display screen including a first display area displaying plural similar medical images received from a case search system and a second display area displaying disease names is displayed on a display of an information terminal. Each similar medical image is a medical image contained in an electronic medical book, and has additional information which includes the number of a page containing the similar medical image. If the user's selection of a similar medical image is detected for the first time after the display of the display screen, a page of an electronic medical book containing the selected similar medical image is displayed on the display screen. If the user's selection of a disease name is detected for the first time after the display of the display screen, a page of an electronic medical book containing differential items corresponding to the selected disease name is displayed on the display screen.

Abstract: First optimization processing for optimizing parameters of a DNN and second optimization processing for optimizing hyperpararneters for each sample used in data augmentation processing are alternately performed. The first optimization processing includes causing the DNN to predict a first augmentation label from a first augmented sample, calculating a first error function between the first augmentation label and a first correct label for a first sample, and updating the parameters in accordance with the first error function. The second optimization processing includes acquiring a second sample, causing the DNN after the updating of the parameters to predict a second label from the second sample, calculating a second error function between the second label and a second correct label for the second sample, and updating the hyperparameter in accordance with a gradient obtained by differentiation of the second error function with respect to the hyperparameter.

Abstract: Provided is a behavior recognition device that selects a candidate behavior as a recognition candidate from among target behaviors based on position information on an image sensor and room layout information, acquires image data detected by the image sensor, determines one or more recognizers corresponding to the candidate behavior, calculates a feature value of the image data using the one or more recognizers, and recognizes the candidate behavior based on the feature value.

Abstract: In a camera device, light is emitted from a turn lamp, a caution lamp, and a welcome lamp. Here, the turn lamp, the caution lamp, and the welcome lamp are provided in a lamp unit. Thus, it is possible to reduce a space between the turn lamp, the caution lamp, and the welcome lamp and reduce the arrangement space of the turn lamp, the caution lamp, and the welcome lamp.

Abstract: A user input obtainer receives an image movement instruction including identification information specifying a position shift or an image capture time shift to be performed and also including a displacement amount. When the identification information specifies the position shift, a slice position selector determines a tomographic image at a destination of the position shift based on the displacement amount from a set of tomographic images captured at the same time. On the other hand, when the identification information specifies the image capture time shift, the image capture time selector determines a tomographic image at a destination of the shift based on the displacement amount from sets of tomographic images that are identical to each other in terms of a patient, an examination portion, and a modality. A displaying image obtainer reads out the determined tomographic image from an image storage device and gives it to a display information generator.

Abstract: A similar case retrieval apparatus includes: a lesion portion acquirer that acquires partial images including lesion portion images, an image feature extractor that extracts image features of each of the plurality of partial images; a location information acquirer that acquires location information of each of the partial images; a lateral position determiner that determines the right organ or the left organ in which each of the lesion portions exists based on the location information; a unilateral distribution identifier that determines whether or not a distribution of the lesion portions is a unilateral distribution; and a similar case retriever that retrieves case data from a case database including both case data for the unilateral distribution in the right organ and case data for the unilateral distribution in the left organ when the unilateral distribution identifier identifies that the distribution of the lesion portions is the unilateral distribution.

Abstract: A method includes displaying, on a display of an information terminal, thumbnail images of similar cases received from a case search system and displaying, in a disease name list display area, a disease name list formed from a list of disease names of the similar cases and the number of cases. If a plurality of regions of interest are set by a user when a search is conducted, the disease name list enters a mode in which the name of a concomitant disease is displayable therein.

Abstract: Similar medical images having a predetermined similarity to a region of interest in a medical image to be interpreted are received from a case search system. A display screen on which the received similar medical images are displayed is displayed on a display. The display screen includes display frames within which similar medical images are displayed. In response to an instruction for enlarging the similar medical images, corresponding regions of interest in the similar medical images are enlarged and displayed with the sizes of the display frames maintained unchanged on the display screen. In response to an instruction for causing a selected similar medical image among the enlarged and displayed similar medical images to move within a corresponding display frame, the other unselected similar medical image(s) is caused to move within a corresponding display frame synchronously with the movement of the selected similar medical image in the same movement direction.

Abstract: A control method of an information terminal causes a display to display a display screen which includes a first display region and a second display region. The first display region displays an object medical image. The second display region displays M number of similar medical images and includes ND number of individual regions for displaying the M number of the similar medical images. When an instruction from one or more instruction buttons is detected, a display size of each corresponding region of interest that is included in the second number of the similar medical images is changed while maintaining a size of each of the individual regions in the second display region, and each of the similar medical images is magnified at a magnification ratio of a reference thumbnail.

Abstract: A method for controlling an information terminal causes a computer of the information terminal to receive, from a case retrieval system, a plurality of similar medical images having a feature quantity of a region of interest and a certain degree of similarity in accordance with the region of interest included in a target medical image, displays a display screen displaying the plurality of received similar medical images on a touch panel display, the display screen including a display region in which at least some of the plurality of received similar medical images are displayed, displays, if selection of a first similar medical image from among the at least some of the plurality of received similar medical images displayed in the display region is detected, the first similar medical image across the display region, and displays, if a swipe operation performed on the first similar medical image is detected, a second similar medical image, which has second highest similarity next to the first similar medical im

Abstract: A control method of an information terminal includes sensing specification information indicating a region of interest in a medical image, receiving from a case retrieval system one or more similar medical images each having a prescribed degree of similarity with a feature quantity of the region of interest indicated by the specification information in accordance with the region of interest, and displaying the medical image that is managed by a medical information management system in a first display region that is displayed on the display and displaying thumbnail views of a plurality of similar medical images received from the case retrieval system in a second display region that differs from the first display region.

Abstract: A display control unit vertically divides a case display area in accordance with the number of disease names selected by a user to create a number of sub-areas equal to the number of disease names. Each of the sub-areas is vertically elongated so that thumbnail images of similar cases of the corresponding disease name are displayed so as to be aligned in a column. The display control unit displays, in each sub-area, thumbnail images of similar cases of the corresponding disease name so that the thumbnail images are aligned in a column in order of decreasing similarity to a search query image displayed in a layout area.

Abstract: A method for controlling an information terminal causes a computer of the information terminal to receive, from a case retrieval system, a plurality of similar medical images having a feature quantity of a region of interest and a certain degree of similarity in accordance with the region of interest included in a target medical image, displays a display screen displaying the plurality of received similar medical images on a touch panel display, the display screen including a display region in which at least some of the plurality of received similar medical images are displayed, displays, if selection of a first similar medical image from among the at least some of the plurality of received similar medical images displayed in the display region is detected, the first similar medical image across the display region, and displays, if a swipe operation performed on the first similar medical image is detected, a second similar medical image, which has second highest similarity next to the first similar medical im

Abstract: Similar medical images having a predetermined similarity to a region of interest in a medical image to be interpreted are received from a case search system. A display screen on which the received similar medical images are displayed is displayed on a display. The display screen includes display frames within which similar medical images are displayed. In response to an instruction for enlarging the similar medical images, corresponding regions of interest in the similar medical images are enlarged and displayed with the sizes of the display frames maintained unchanged on the display screen. In response to an instruction for causing a selected similar medical image among the enlarged and displayed similar medical images to move within a corresponding display frame, the other unselected similar medical image(s) is caused to move within a corresponding display frame synchronously with the movement of the selected similar medical image in the same movement direction.

Abstract: An apparatus includes a memory, and circuitry which, in operation, performs operations including, storing, in the memory, an object occurrence map defining an occurrence area where there is a possibility that an object appears, detecting the object included in a captured image of a scene seen in a running direction of a vehicle, switching a vehicle drive mode, based on a result of the detection of the object and the object occurrence map, from an automatic drive mode in which the vehicle is automatically driven to a manual drive mode in which the vehicle is driven manually by a driver, and controlling driving of the vehicle in the switched manual drive mode.