Multimodal MRI Brain Tumor Image Segmentation Using Sparse Subspace Clustering Algorithm

Table 1

Evaluation indicator introduction.

Number	Index	Explanation
1		The Dice coefficient is a set similarity measurement method. In the image, it mainly refers to the degree to which the actual segmentation result and the golden segmentation result overlap each other, and the value is . Among them, 0 represents that there is no overlap between the actual segmentation result and the golden segmentation result, which represents the worst segmentation accuracy at this time, and 1 represents that the actual segmentation result and the golden segmentation result completely overlap, which represents the optimal segmentation accuracy at this time.
2		The Jaccard coefficient is a method similar to the Dice coefficient that relies on similarity as a measure. It describes the degree of overlap between the actual segmentation result and the golden segmentation result from another perspective.
3		The false positive rate (Precision) reflects the accuracy of the actual segmentation result. The ratio of the overlap between the actual segmentation result and the golden segmentation result is used for description. The higher the ratio, the higher the proportion of the golden result included in the actual segmentation result.
4		The true positive rate (Recall) reflects the accuracy of the actual results in the actual segmentation results. It refers to the ratio of the overlap between the actual and golden section results. The higher the ratio, the higher the proportion of the true segmentation result in the golden section.