Abstract: A computer-implemented method for task-guided graph augmentation and editing includes receiving an input graph in an observed financial transaction network. A data augmentation function is learned, where the data augmentation function maintains a true data distribution of the input graph. An augmented financial transaction network is generated that enhances performance of a downstream task and preserves topological and temporal properties of the input graph.

Abstract: An optical marker for positioning a medical device outside a body and the medical device are disclosed. The optical marker comprises: a base having a concave or a convex, the concave or the convex being provided with a non-coplanar optical mark(s), and the optical mark(s) being visible on the entire surface of the concave or the convex; and a connecting portion connected to the base and used for connecting with the medical device. In the optical marker, the base provided with the optical mark(s) has a non-planar structure, being able to increase the recognizable angle of the medical device and improve the recognition accuracy and stability.

Abstract: The present solution provides a surgical navigation system and method, and an electronic device and a readable storage medium. The surgical navigation system comprises: an image obtaining module, for obtaining a surgical scene image; an image recognition module, for executing image recognition on the surgical scene image to obtain a first recognition result, wherein the first recognition result is used to represent an identifier contained in the surgical scene image; an instruction obtaining module, for, based on the first identification result, obtaining a corresponding interaction instruction; an instruction execution module, for, based on the interaction instruction, controlling the surgical navigation system to execute a corresponding surgical navigation step. The implementation of the technical solutions of the present application can reduce the chance of misjudgment when the surgical navigation system is controlled.

Abstract: A dynamic position recognition and prompt method and system. The system comprises: a recognition module (1), a judgment module (2) and a prompt module (3); wherein the recognition module (1) is used to obtain a current scene image via an image obtaining device and recognize a current position of a specific object in the current scene image; the judgment module (2) is used to judge whether the current position of the specific object is within a preset recognition range; the prompt module (3) is used to performing navigation for the specific object if the current position of the specific object is not within the preset recognition range of the image obtaining device, to help a user in finding the specific object to be in the predetermined recognition range, so that the image and the navigation are more accurate and precise.

Abstract: An optical marker for positioning a medical device outside a body and the medical device are disclosed. The optical marker comprises: a base having a concave or a convex, the concave or the convex being provided with a non-coplanar optical mark(s), and the optical mark(s) being visible on the entire surface of the concave or the convex; and a connecting portion connected to the base and used for connecting with the medical device. In the optical marker, the base provided with the optical mark(s) has a non-planar structure, being able to increase the recognizable angle of the medical device and improve the recognition accuracy and stability.

Abstract: A set of sentences within a natural language text document are parsed, generating a word-level graph corresponding to a sentence in the set of sentences. Within the word-level graph using a trained entity identification model, a set of entity candidates are identified. From a set of graphs modelling relationships between portions of the set of sentences, a set of embeddings is generated. From a set of pairs of embeddings in the set of embeddings using a set of deconvolution layers, a set of links between entity candidates within the set of entity candidates is extracted. From the set of links and the set of entity candidates, an output graph modelling linkages between portions of the set of sentences within the natural language text document is generated.

Abstract: A set of sentences within a natural language text document are parsed, generating a word-level graph corresponding to a sentence in the set of sentences. Within the word-level graph using a trained entity identification model, a set of entity candidates are identified. From a set of graphs modelling relationships between portions of the set of sentences, a set of embeddings is generated. From a set of pairs of embeddings in the set of embeddings using a set of deconvolution layers, a set of links between entity candidates within the set of entity candidates is extracted. From the set of links and the set of entity candidates, an output graph modelling linkages between portions of the set of sentences within the natural language text document is generated.

Abstract: A microdroplet treatment device, comprising: a sample introduction system, a microfluidic chip system (46), a temperature control system, a droplet recognition system, a droplet detection system and a control system (45); the sample introduction system is used for introducing, into the microdroplet treatment device, samples of aqueous phase and oil phase, and the sample introduction system comprises at least: a sample introduction system I (41) and a sample introduction system II (42), the sample introduction system I (41) being used for introducing, into the microdroplet treatment device, samples of oil phase, and the sample introduction system II (42) being used for introducing into the microdroplet treatment device, samples of aqueous phase; the microfluidic chip system (46) comprises a substrate (4), pipes (1a, 1b, 1c, 1d, 2, 3) formed in the substrate, and a first detection window (9) and a second detection window (10); the temperature control system comprises: a temperature sensor and a temperature cont

Abstract: This application provides a method for obtaining a virtual reality (VR) resource by a terminal and a terminal, and relates to the field of communications technologies, so that the terminal can quickly and conveniently find a VR resource required by a user, to improve user experience. The method specifically includes, when a terminal detects an operation of choosing, by a user, to search for a VR resource related to a specific picture, automatically obtaining, by the terminal, feature information of the specific picture, and generating a request based on the feature information, sending, by the terminal, the request to a VR server, and then receiving, by the terminal, a list of related VR resources found by the VR server through matching based on the request.