1 INFORMATIVE METHOD FOR PARKINSON'S DISEASE DIAGNOSIS BASED ON AIArtificial Intelligence USING NIGROSOME-1 AREA CLASSIFIED BY MACHINE LEARNING AND MULTIPLE PREDICTION RESULT BY MULTIPLE LEARNING MODELS

摘要

According to an aspect of the present invention, a method of providing information for diagnosis of Parkinson's disease includes: a first step of obtaining a first image related to a multi-echo size and phase from an MRI of a patient's brain by an image acquisition unit; A second step of post-processing the acquired first image by an image processing unit to enable observation of the nigrosome 1 region and black matter used as image biomarkers of Parkinson's disease; A third step of classifying a second image including the Nygrosome 1 region by analyzing the post-processed first image by an image analysis unit; A fourth step of detecting the nigrosome 1 region from the classified second image by the image analysis unit; And a fifth step of diagnosing the presence or absence of Parkinson's disease of the patient by analyzing whether the detected Nygrosome 1 region is normal or not; including, in the second step, the image processing unit, a quantitative susceptibility mapping algorithm Based on, the post-processing is performed by applying a quantitative susceptibility map mask to the first image to generate a susceptibility map weighted imaging image, and between the second and third steps, the image processing unit It may further include a step 2.5 of further performing at least one of an angle adjustment, an image magnification, and a reslice operation on the generated susceptibility map-weighted imaging image.