Automated Extraction of Human Functional Brain Network Properties Associated with Working Memory Load through a Machine Learning-Based Feature Selection Algorithm

Table 2

Brain regions that were significantly activated between the control and 2-back or 3-back tasks (FWE-corrected, ).

Task	Region				value	Cluster size
2-back	Precentral L	−34	−2	60	Inf	2584
	Supp Motor Area L
	Parietal Sup L	−26	−70	52	7.65	2865
	Parietal Inf L
	Frontal Sup R	28	0	62	6.66	718
	Parietal Inf R	30	−56	50	6.61	2083
	Precuneus R
	Parietal Sup R
	Insula L	−34	18	8	6.58	331
	Frontal Inf Tri L				5.24
	Caudate L	−16	−6	16	6.07	450
	Thalamus L				5.52
	Precentral R	52	10	36	5.95	290
	Insula R	34	22	6	5.57	113
	Vermis 6	4	−64	−20	5.47	62
	Caudate R	20	−6	24	4.96	4
	Vermis 1 2	2	−34	−16	4.88	8
	Frontal Mid L	−38	50	24	4.76	1
	Varmis 3	−2	−28	−12	4.73	2
	Temporal Inf L	−48	−54	−16	4.72	1
	Frontal Mid R	40	34	36	4.70	2
3-back	Parietal Inf L	−42	−54	52	7.49	2334
	Precentral L	−30	−2	60	7.38	4437
	Supp Motor Area L	−8	10	52	7.13
	Angular R	36	−66	48	7.30	1993
	Parietal Inf R	44	−44	54	6.12
	Frontal Mid R	38	36	32	6.03	299
	Insula R	34	22	8	5.88	204
	Frontal Mid L	−30	54	12	5.58	61
	Putamen L	−20	0	14	5.26	43
	Caudate L	−18	0	22	4.72
	Thalamus L	−16	−22	18	4.93	10
	Pallidum L	−18	4	0	4.77	2