Intelligent diagnosis system for Parkinson's disease based on magnetic resonance image

The invention discloses an intelligent diagnosis system for Parkinson's disease based on a magnetic resonance image, and belongs to the technical field of medical image processing. According to the diagnosis system, a magnetic resonance image is input into a sub-region division unit to obtain a corresponding sub-region set; the detection unit obtains corresponding effective sub-region data based on the effective sub-regions set by the setting unit and distributes the effective sub-regions to the corresponding neural network models to obtain corresponding diagnosis results, and the calculation unit sets the weights of the effective sub-regions based on the setting unit. If the diagnosis result is larger than the threshold value, the diagnosis result indicates that the patient is sick, if the diagnosis result is smaller than the threshold value, the patient is healthy, and the diagnosis result is output and displayed through an output unit. Based on the characteristics that the magnetic resonance image is high