Three-dimensional spine MR image automatic segmentation device

The invention discloses an automatic segmentation detection method for a three-dimensional spine MR image, and the method comprises the following steps: obtaining the three-dimensional spine MR image, forming a data set, and dividing the data set into a training set and a test set; preprocessing the acquired three-dimensional spine MR image, converting the three-dimensional image in the training set into two-dimensional slice data, and compressing the three-dimensional image in the test set into an h5 file; constructing a TransUNet network, and training the TransUNet network by using the training set to obtain a trained network model; using the trained network model to perform segmentation test on the data in the test set to obtain a predicted segmentation result; and inputting a to-be-segmented three-dimensional spine MR image into the trained network model to obtain a segmentation result. The automatic segmentation device for the three-dimensional spine MR image is suitable for the spine MR image, and can e