Scientific Reports

Table 2 Variables selected by the GA models for the levels of symptoms studied.

From: Spectrochemical approach combined with symptoms data to diagnose fibromyalgia through paper spray ionization mass spectrometry (PSI-MS) and multivariate classification

Symptom	Model	Selected variables (m/z)
VAS (mild)	GA-LDA	22	37	143	180	563	575	645	827	844	850
		365	448	659	681	707	726	778	841
		113	160	161	189	336	413	794
VAS (moderate)	GA-SVM	133	349	418	459	551	559	679	686
		92	144	182	326	601	646	881
		17	18	153	391	486	797	840	846	893
FIQ (moderate)	GA-SVM	26	160	196	201	343	627	895
		91	365	518	622	643	776
		131	132	141	225	259	389	492	766	888
FIQ (severe)	GA-LDA	261	407	449	598	722	732	784	841	849
		5	8	259	267	389	397	434	779
		200	204	334	375	444	599	838	888
HAM (moderate)	GA-SVM	67	122	468	699	700	805
		159	169	195	231	255	302	717	879	887
		75	147	414	445	477	891
HAM (severe)	GA-LDA	66	174	386	538	546	846	871
		39	273	492	629	702	719	832
		249	277	330	391	421	857
QL (regular)	GA-SVM	31	72	155	361	614	708	797	886
		57	263	404	530	570	741	823
		131	225	366	385	485	533	636	729	780	827
TAMPA (moderate)	GA-SVM	380	381	383	543	562	649	684	792	877
		72	119	183	394	478	584	596	773	803
		76	252	320	387	537	586	624	707	869
TAMPA (severe)	GA-LDA	102	255	402	544	602	646	785
		186	336	352	501	535	560	639	645
		100	375	432	530	651	712	787	848	880
CAT (mild)	GA-LDA	128	168	393	491	582	683	687	751
		5	101	252	350	522	532	618	821
		187	218	413	423	749	814	860
CAT (moderate)	GA-SVM	55	199	366	478	597	679	786	866	889
		31	69	80	133	428	449	455	553	872
		95	121	637	673	695	824	848

Symptoms: VAS Visual Analog Scale of Pain, FIQ Fibromyalgia Impact Questionnaire, HAM Hamilton Anxiety Rating Scale, QL Quality of Life index, TAMPA The Tampa Kinesiophobia Scale, CAT Pain Catastrophizing Scale. Models: GA-LDA genetic algorithm with linear discriminant analysis, GA-SVM genetic algorithm with support vector machines.
The values in bold represent the variables that occurred in more than one model, totaling 31 different variables.

Back to article page

Search

Advanced search

Quick links