Brain network multi-frequency fusion map kernel Alzheimer's disease auxiliary diagnosis device and method

Abstract

The invention provides an Alzheimer's disease auxiliary diagnosis device and method for brain network multi-frequency fusion map kernel, and relates to the technical field of computer-aided diagnosis. The device includes an image preprocessing module, an image frequency division module, an image kernel generation module, an image kernel fusion module, and an auxiliary diagnosis module; the image frequency division module matches the functional MRI image with the AAL template, and performs frequency division processing; The generation module constructs a multi-frequency brain network for the frequency-divided image and forms a matrix; the image kernel fusion module fuses all the image kernels into one image kernel; the auxiliary diagnosis module combines the fused image kernel with the kernel extreme learning machine to realize the Alzheimer's disease diagnosis. The invention also provides a diagnosis method using the device. The Alzheimer's disease auxiliary diagnosis device and method provided by the present invention can fully represent the difference of brain activity information in multiple frequency bands, and make full use of the signal information of the functional magnetic resonance image.

Description

technical field [0001] The present invention relates to the technical field of computer-aided diagnosis, in particular to a device and method for assisting the diagnosis of Alzheimer's disease with brain network multi-frequency fusion map kernel. Background technique [0002] In recent years, neuroimaging technology, especially functional imaging, has made rapid progress, which provides corresponding techniques for studying the functions of various brain regions in patients with Alzheimer's disease. As one of the mature functional imaging detection technologies, functional magnetic resonance imaging has the advantage of non-invasive monitoring of brain function and activity with high spatial and temporal resolution. Measuring correlations between various brain regions using a blood oxygen level-dependent approach has proven to be a powerful tool for studying the functional organization of the brain. In the resting state functional Magnetic Resonance Imaging (fMRI), it can b...

AI Technical Summary

Problems solved by technology

[0003] In fMRI data, signals in different frequency bands contain brain activity information in different frequency bands, while the brain networks in existing studies are constructed based on time-domain signals, ignoring the difference information in each frequency band in the frequency domain. The obtained graph kernel also cannot more accurately describe the frequency domain information of the brain

Images