An image processing apparatus includes a display and processing circuitry. The processing circuitry generates second image data by subtracting first image data before and after contrast enhancement. The processing circuitry removes information indicating a structure from pieces of the second image data to generate third image data. The processing circuitry correct misalignment of the third image data based on characteristics of the shape of the subject represented in at least any one of the first to third image data. The processing circuitry obtains a variation index between the third image data after alignment correction, and generates fourth image data base on the variation index. The processing circuitry displays a medical image based on the fourth image data on the display.