Model training method and device, data processing method, device, client and storage medium

Abstract

The invention relates to model construction in artificial intelligence, and provides a model training method, a data processing method, a data processing device, a client and a storage medium. The method comprises the following steps: obtaining ultrasonic image data containing a target detection object, and pre-training a preset neural network through the ultrasonic image data to obtain a target neural network; extracting a plurality of parts of ultrasonic image data from the ultrasonic image data, and training the target neural network through the plurality of parts of ultrasonic image data to obtain a training result; determining gradient data of the target neural network based on the training result, encrypting the gradient data, and sending the encrypted gradient data to a federated learning server; receiving target gradient data sent by the federated learning server; and updating model parameters of the target neural network according to the target gradient data until the updatedtarget neural network converges to obtain an ultrasonic image data processing model. The detection effect and accuracy of the ultrasonic image data processing model can be improved.

Description

technical field [0001] The present application relates to the technical field of model building in artificial intelligence, and in particular to a model training method, a data processing method, a device, a client, and a storage medium. Background technique [0002] At present, doctors obtain ultrasonic images of target detection objects such as the liver, spleen, and kidneys by performing ultrasound examinations on the human body, and doctors check the characteristics of key parts in the ultrasound images to obtain the ultrasound examination results of target detection objects such as the patient's liver. In the existing deep learning algorithm, the ultrasonic image data processing model can be established through the ultrasonic image data of the target detection object to determine the ultrasonic inspection results of the target detection object. However, due to the privacy of medical data, the existing ultrasonic image data processing model The quality is often limited b...

AI Technical Summary

Problems solved by technology

In the existing deep learning algorithm, the ultrasonic image data processing model can be established through the ultrasonic image data of the target detection object to determine the ultrasonic examination results of the target detection object. However, due to the privacy of medical data, the existing ultrasonic image data processing model The quality is often limited by the scale and quality of the sample data set, which leads to the low detection effect and accuracy of the established ultrasonic image data processing model of the target detection object

Images