6 results about "Digital medicine" patented technology

A method for secure aggregation, by a server, of client-provided inputs includes receiving, from each of a plurality of clients, a respective client input, for which a commitment is published. The commitments were computed using randomness and are aggregated by at least two super-clients and a sum of the aggregated commitments is published by each super-client. A sum of the received client inputs is published such that validity of the sum is checkable, by the clients, by comparing the sum of the received client inputs to a verification algorithm result that uses a sum of additive shares computed by the clients using the randomness, and by verifying that the published sum of the aggregated commitments is the same for each super-client. The method can be applied to use cases, for example, in digital medicine using medical data or smartcity applications to support decision-making.

The application discloses a medical intention recognition method and device, a storage medium and computer equipment, relates to the technical field of digital medicine, and mainly aims to solve the problems of poor recognition accuracy and effectiveness of the existing medical intention. The method comprises the following steps: obtaining dialogue text to be recognized; performing recognition on the dialogue text based on an intention recognition model that has completed model training, to obtain a first medical intention, wherein the intention recognition model is trained based on a loss function constructed based on a positive sample pair and a negative sample pair, and text training samples; the positive sample pair and the negative sample pair are determined based on opposite statements in the text training samples; determining a second medical intention matched with the first medical intention based on a generated medical knowledge graph, and generating a medical intention recognition result based on the first medical intention and the second medical intention.

The application relates to the fields of artificial intelligence and digital medicine, and provides a mechanical arm control method, device, equipment and storage medium. The method detects left and right images, obtains first and second elliptical boundaries, selects first and second clustering ellipses and performs matching, obtains target left and right ellipses, identifies an initial distance between a desktop cup and an electronic device according to a left center point of the target left ellipse and a right center point of the target right ellipse, identifies a cup diameter of the desktop cup according to first and second left long axis vertices of the target left ellipse and first and second right long axis vertices of the target right ellipse, generates a distance relationship between a mechanical arm and the desktop cup based on a positional relationship between the electronic device and the mechanical arm and the initial distance, and controls the mechanical arm to operate, and the application is suitable for a dynamic scene. In addition, the application also relates to a blockchain technology, and the distance relationship and the cup diameter can be stored in the blockchain.

The application relates to the field of digital medicine, in particular to an emotion recognition method and device based on electroencephalogram (EEG) data, an equipment and a medium. The method is based on a source encoder and a classifier trained based on first type EEG data and corresponding emotion classification results, parameters of the source encoder are used to initialize a target encoder to obtain an initial target encoder, the source encoder and the initial target encoder encode the first type EEG data and second type EEG data respectively, a discriminator is used to discriminate the encoder to which the encoding result belongs, parameters of the initial target encoder are optimized according to the discrimination result, until the loss determined by the discrimination result is minimum, the optimized target encoder is used to encode the second type EEG data, the encoding result is input into the classifier, and a classification result is output as an emotion recognition result corresponding to the second type EEG data. Based on EEG data with a known emotion classification result, emotion recognition of another EEG data can be realized without other marking operations, and the recognition efficiency is improved.

The application relates to the technical field of artificial intelligence and the field of digital medicine, and discloses a medical data processing method, device and equipment based on artificial intelligence and a medium, which comprises the following steps: preprocessing collected initial medical data to obtain a medical sample set; training a multi-modal basic model by using the medical sample set based on a self-supervised learning method to obtain an initial medical analysis model; fine-tuning the initial medical analysis model by using a multi-task learning strategy to obtain a first medical analysis model; optimizing the first medical analysis model by using a multi-objective optimization strategy to obtain a second medical analysis model; improving the second medical analysis model by using a feedback learning method to obtain a target medical analysis model; and evaluating target medical data based on the target medical analysis model to obtain a medical evaluation result. The medical analysis model can be stored in a blockchain. The application improves the evaluation efficiency of medical data and ensures the accuracy of the medical evaluation result.

The application relates to the technical field of computer software, can be applied to the field of science and technology finance / digital medicine, and discloses a dynamic linkage code review method of a security rule base. The method comprises the following steps: receiving a code submission request, obtaining source code data and running environment identification data associated with the code submission request; determining rule screening conditions according to the running environment identification data, generating rule query parameters, and sending a rule acquisition request to the security rule base based on the rule query parameters; receiving a target rule data set returned by the security rule base and corresponding to the running environment identification data; performing matching processing on the source code data according to the target rule data set, and generating risk level data corresponding to the source code data according to the matching processing result; generating a code submission control instruction according to the risk level data, and changing the submission process state of the source code data according to the code submission control instruction. The method improves the pertinence and process controllability of code security review.