A Machine Learning Model Based on Unsupervised Clustering Multihabitat to Predict the Pathological Grading of Meningiomas

Table 2

Distinguishing ability of clustering parameters.

	AUC (95% CI)	Specificity	Sensitivity	Youden index
Inertia	0.597 (0.504-0.685)	46.67%	73.33%	0.20
Calinski-Harabasz Index	0.596 (0.502-0.684)	41.67%	76.67%	0.183
Silhouette coefficient	0.754 (0.668-0.828)	83.33%	56.67%	0.40
Separation	0.571 (0.477-0.661)	46.67%	71.67%	0.184
Davies-Bouldin Index	0.674 (0.583-0.757)	48.33%	80.0%	0.283