Unsupervised Scoliosis Diagnosis via a Joint Recognition Method with Multifeature Descriptors and Centroids Extraction

Table 2

The performance comparison of our methods and other related works.

Methods
Evaluation criteria	Supervised classification forests [11]	Interpolation theory + shape-constrained [9]	Single-feature + CGAdaBoost-DRLSE	Multifeature + CGAdaBoost-DRLSE
Detection accuracy rate	86%	97%	98%	99%
Centroid location error	4.4 mm	1.1 mm	1.51 mm	0.87 mm