Multi-modality medical image analysis method and apparatus for brain disease diagnosis

A multi-modality medical image analysis method for brain disease diagnosis, which is executed by a medical image analysis apparatus having at least one processor, the method comprising:

• acquiring medical images with different modalities for a patient;
• selecting, from a plurality of pre-trained analysis models, the analysis models having modalities corresponding to the modalities of the acquired medical images;
• inputting the medical images correspondingly to the analysis models selected with respect to the modalities of the medical images to produce output values related to a plurality of factors related to at least one brain disease;
• converting the output values to produce a plurality of feature vectors corresponding to the output values;
• and
• inputting the plurality of feature vectors to at least one or more diagnosis models pre-trained to predict a degree of progression of the at least one brain disease,
• wherein when there is a defect analysis model of the analysis models that does not have the medical images with the modalities corresponding thereto, the method further comprises detecting medical images of another patient having the same brain disease, and
• wherein the producing the output values related to the plurality of factors comprises performing imputation of the defect analysis model with the detected medical images of the another patient to output the output values.