Clinical decision support systems for brain tumor characterization using advanced magnetic resonance imaging techniques

Table 1 Validation results of the clinical decision support systems based on chemical shift imaging data

Ref.	Voxel assignment	Accuracy
De Edelenyi et al[27]	Low-grade gliomas	92.9%
	High-grade gliomas	79.16%
	Metastasis	60%
	Meningiomas	100%
	Necrosis	100%
	Healthy tissue	100%
	Cerebrospinal fluid	100%
Simonetti et al[29]	Healthy tissue	100%
	Cerospinal fluid	97%
	Glioma grade II	83%
	Glioma grade III	88%
	Glioma grade IV	100%
Luts et al[32]	Glioma II	66.6%
	Glioma II/III	100%
	Glioma IV	100%
	Meningioma	100%
McKnight et al[28]	Low grade gliomas vs grade III	89%
Li et al[34]	Glioblastoma multiforme	100%
	Glioma II	100%

Table 2 Validation results of the clinical decision support systems based on single voxel data

Ref.	CDSS	Differentiation problem	Accuracy						Supportive raw files
			Short TE			Long TE	Short + Long TE
Pérez-Ruiz et al[38]	INTERRET	Low grade meningiomas vs low grade glial tumors	94a	89b	89c	83b	84c	89c
		Pseudotumoural diseasedvs tumorsevs normal brain	86c			81c	92c
García-Gómez et al[41]	eTUMOUR	Low grade glioma vs high grade tumor	92			84	92		1.5 Tesla MRS data of Philips (sdat/spar) GE up to 9X (SAGE Pxxxx with an shf or sdf/shf) siemens scanners (numaris 4) jMRUI[58] text file
		Meningioma vs glioma/Met	92			78	94
		Low men vs glioma/Met vs low grade glioma	87			75	90
Sáez et al[44]	HealthAgents	Aggressive tumor vs meningioma vs low grade glial	94			-
		Meningioma vs metastasis	91			-
		High grade tumor vs low grade tumor	87			68 (ch)
		Affected tissue vs non affected tissue	99			-
		Tumor vs non tumor	97			-
		Aggressive tumor vs non aggressive tumor	81			72 (ch)
		Glioma vs embryonal tumor	-			72 (ch)
		Glioblastoma vs low grade glioma	84			-
		Glioblastoma vs meningioma	91			-
		Meningioma vs low grade glioma	92			-
		Metastasis vs low grade glioma	85			-
Vicente et al[46]	CURIAM BT	Aggressive tumor vs non aggressive tumor	85			87 (ch)			1.5 or 3 Tesla MRS data of different manufactures (Siemens, GE, Philips) by means of jMRUI[58] and jDMS[36]
		Pilocytic astrocytoma/ependymoma grade II vs medulloblastoma	88 (ch)			85 (ch)	89 (ch)
		Pilocytic astrocytoma vs medulloblastoma	92 (ch)			94 (ch)	95 (ch)
		Pilocytic astrocytoma vs ependymoma grade II vs medulloblastoma	76 (ch)			69 (ch)	92 (ch)

It is indicated where the classification accuracy corresponds to classifier trained on pediatric tumor data (ch).

^aInternational Network for Pattern Recognition of Tumors Using Magnetic Resonance (INTERPRET) version 1.1;

^bINTERPRET version 2.0;

^cINTERPRET version 3.0;

^dPseudotumoural disease: Acute infarct, multiple sclerosis, acute disseminated encephalomyelitis, and no specific pseudotumoral disease; eTumors: Astrocytoma World Health Organization (WHO) grade II, oligodendroglioma WHO grade II, oligoastrocytoma WHO grade II, astrocytoma WHO grade III, oligoastrocytoma WHO grade III. CDSS: Clinical decision support systems; MRS: Magnetic resonance spectroscopy; MRUI: Magnetic Resonance User Interface.