abstract |
A method for identifying a neurological disease by using a magnetic resonance imaging image for identifying a magnetic resonance image of a brain of a subject, and comprising: generating a plurality of brain image partitions according to the magnetic resonance imaging image, and each The brain image partition corresponds to at least one diffusion index and a plurality of diffusion index values of each diffusion index; for each diffusion index, the diffusion index value of the diffusion index of the brain image partition is calculated by using a statistical transformation related to the normal distribution Performing conversion; for each diffusion index corresponding to each brain image partition, generating characteristic parameters according to the diffusion index values of the diffusion index; and classifying the whole of the characteristic parameters by using a pair of neurological disease classifiers, And sorting into one of a plurality of predetermined categories related to the neurological disease. |