Brain age prediction method and device based on artificial intelligence, equipment and storage medium

Abstract

The embodiment of the invention discloses a brain age prediction method and device based on artificial intelligence, equipment and a storage medium, and the method comprises the steps: inputting a target brain grey matter three-dimensional image and a target brain white matter three-dimensional image of a target object into a brain age prediction model corresponding to the target actual age of the target object for brain age prediction, and obtaining an initial brain age, wherein the brain age prediction model is a model obtained by training a deep learning residual model based on cavity convolution; and performing deviation correction on the initial brain age by adopting the target actual age and a target deviation correction function corresponding to the target actual age to obtain a brain age prediction result corresponding to the target object. Brain age prediction is carried out by adopting a model for distinguishing age groups, the operation speed is higher during prediction, the prediction accuracy is higher, and the accuracy of brain age prediction is further improved through deviation correction; through a deep learning residual error model and cavity convolution, the error of predicting the brain age by the model is reduced.

Description

technical field [0001] The present invention relates to the technical field of image processing, and in particular, to a brain age prediction method, device, device and storage medium based on artificial intelligence. Background technique [0002] Medical imaging refers to the non-invasive imaging of internal tissues of the human body or a part of the human body for medical treatment or medical research. The traditional brain age prediction method first extracts handcrafted features (including underlying features and intermediate features) from the magnetic resonance images of the brain, and then classifies or regresses to predict brain age according to the handcrafted features. The whole process relies on tissue segmentation and feature selection in the image. The tedious workflow, the prediction results are affected by manual features, and the error of the prediction results is large. At present, one model is used to predict the brain age of multiple age groups. During mo...

AI Technical Summary

Problems solved by technology

[0006] Based on this, it is necessary to use one model to predict the brain age of multiple age groups. During model training, the proportion of training samples for each age group is unbalanced. If the actual age of the object to be predicted corresponds to the smallest If the ratio of training samples is higher, there will be a technical problem that the accuracy of model prediction will become lower. An artificial intelligence-based brain age prediction method, device, equipment, and storage medium are proposed.

Images