The invention provides a medical image data rapid auxiliary labeling and storage method, which comprises the steps of data labeling, data storage, data reading, data conversion and the like. The method comprises the following steps: A1, before data labeling, reading an image ID, wherein the image ID comprises the information of patient Id, StudyId, StudyInstanceUid, SertissInstanceUid and SopInstanceUid; A2, labeling the image; After the image labeling, a labeling box appears; A3, customizing a labeling type; B1, storing the annotation information in a JSON format, wherein the annotation information comprises an image ID, an annotation box and a user-defined annotation type; C1, during data reading, comparing the image with the ID of the JSON file, entering the next step if the image is consistent with the ID of the JSON file, and exiting if the image is inconsistent with the ID of the JSON file; C2, reading and displaying the annotation information. The method is compatible with an existing storage scheme. The storage scheme of labeled data in medical images can be effectively solved. The powerful support can be effectively provided for a deep learning algorithm. The invention also comprises an image multi-channel processing process which can be applied to an MRI system.