Diagnostic support for selected neuromuscular diseases using answer-pattern recognition and data mining techniques: a proof of concept multicenter prospective trial

Table 4 Sensitivity (%) of different classifiers in selected neuromuscular disease groups during 21-fold cross-validation

Diagnostic group Classifier system	MdMy^a (1)	MP (2)	SMA (3)	ALS (4)	PNP (5)	Other^b (6)	NND^c (7)
SVM	92	84	38	78	77	56	89
RF	100	93	69	93	73	63	97
LR	84	86	56	81	73	81	94
NB	94	88	56	52	73	75	92
LD	94	88	69	81	77	81	94
NN	76	81	44	52	64	50	94
Fusion	96	93	69	81	86	81	97

SVM support vector machine, RF random forest, LR logistic regression, NB naive Bayes, LD linear discriminant analysis, NN nearest neighbor
^aincluding patients with Duchenne and Becker muscular dystrophy (MD), oculopharyngeal muscular dystrophy (OPMD), proximal myotonic myopathy (PROMM), facioscapulohumeral MD, progressive MD, limb-girdle-MD, myotonia congenita Thomsen
^bincluding patients with chronic progressive external opthalmoplegia (CPEO) -plus, polymyositis, Lambert-Eaton myasthenic syndrome, Ullrich congenital muscular dystrophy, Miyoshi myopathy, Friedreich ataxia, primary lateral sclerosis (PLS), spinal and bulbar muscular atrophy (SBMA), and chronic inflammatory demyelinating polyneuropathy (CIDP)
^c NND no neuromuscular disease

ISSN: 1472-6947