FreeSurfer subcortical normative data.

This article contains a spreadsheet computing estimates of the expected subcortical regional volumes of an individual based on its characteristics and the scanner characteristics, in addition to supplementary results related to the article "Normative data for subcortical regional volumes over the lifetime of the adult human brain" (O. Potvin, A. Mouiha, L. Dieumegarde, S. Duchesne, 2016) [1] on normative data for subcortical volumes. Data used to produce normative values was obtained by anatomical magnetic resonance imaging from 2790 healthy individuals aged 18-94 years using 23 samples provided by 21 independent research groups. The segmentation was conducted using FreeSurfer. The spreadsheet includes formulas in order to compute for a new individual, significance test for volume abnormality, effect size and estimated percentage of the normative population with a smaller volume while taking into account age, sex, estimated intracranial volume (eTIV), and scanner characteristics. Detailed R-squares of each predictor for all formula are also reported as well as the difference of subcortical volumes segmented by FreeSurfer on two different computer hardware setups.

Specifications Table

Value of the data

The data provides the first subcortical regional normative values in a very large sample of healthy individuals with a wide age range and diversity of scanner manufacturer and magnetic field strength.

The calculator can be used to assess deviation from normality for any given individual patient or healthy control.

These values can be useful for multicenter studies using various scanner manufacturers and magnetic field strengths.

Data

A Microsoft Excel spreadsheet computing expected subcortical regional volumes for an individual according to his age, sex, intracranial volume and the scanner characteristics is provided (see Subcortical_Norms_Calculator.xlsm file online). Table 1 reports detailed R-squares of each predictor for all models predicting subcortical volumes. Table 2 shows the difference of subcortical volumes segmented by FreeSurfer on two different computer hardware setups.

Table 1

Percentage of the variance explained (R2) by each predictor in models predicting subcortical regional volumes.

Regions	Age	Age2	Age3	Sex	eTIV	eTIV2	eTIV3	MFS	GE / Siemens	Philips / Siemens	GE X MFS	Philips X MFS	eTIV X MFS	Age X Sex	eTIV X GE	eTIV X Philips	Total R2	Validation R2
Accumbens L	25.6	1.1	–	1.6	0.2	0.0	–	1.0	0.4	1.4	1.8	5.1	–	0.4	–	–	38.5	34.2
Accumbens R	28.7	0.3	0.1	2.6	0.1	–	–	1.0	0.1	1.4	0.5	3.0	–	0.2	–	–	37.8	28.6
Amygdala L	14.0	1.1	0.1	13.1	4.2	0.1	–	8.1	0.0	0.0	0.5	0.0	–	0.1	0.1	0.1	41.4	39.0
Amygdala R	9.6	0.1	0.2	12.7	3.5	–	–	4.4	0.1	0.0	0.0	0.4	–	–	–	–	31.1	33.9
Brainstem	3.1	0.9	0.3	21.5	26.7	0.2	–	0.0	0.3	0.0	0.9	0.1	–	0.2	–	–	54.1	61.1
Caudate L	12.8	3.7	0.1	7.1	15.1	0.2	0.0	–	0.0	2.0	–	–	–	0.2	0.0	0.1	41.2	37.0
Caudate R	9.0	7.2	–	6.8	11.7	0.0	–	0.0	0.4	5.5	0.0	0.5	0.0	0.2	0.0	0.2	41.7	31.4
Hippocampus L	21.4	5.8	0.0	6.9	10.6	0.2	–	3.3	0.7	1.6	0.2	0.0	0.0	0.2	–	–	50.9	48.2
Hippocampus R	18.0	6.7	0.1	7.2	11.1	–	–	5.3	0.4	0.6	0.2	0.1	–	0.2	–	–	49.7	51.6
Pallidum L	14.5	3.0	0.1	8.8	8.6	0.2	–	1.4	0.6	0.6	0.0	1.8	–	0.6	–	–	40.0	37.8
Pallidum R	19.5	1.5	0.5	8.5	6.9	0.1	–	1.1	0.1	3.6	0.3	1.1	–	0.3	–	–	43.4	42.4
Putamen L	34.6	1.9	–	6.2	3.3	0.0	0.0	0.1	0.1	3.8	0.1	1.5	0.2	0.3	–	–	52.0	41.9
Putamen R	34.7	2.9	0.0	7.3	3.2	0.0	–	0.2	0.1	3.4	0.0	2.1	–	0.5	–	–	54.2	47.2
Thalamus L	27.3	1.8	0.4	10.8	17.1	0.5	–	1.7	0.0	0.8	0.1	0.5	0.2	0.3	–	–	61.5	57.3
Thalamus R	34.8	0.4	0.3	12.1	17.0	0.3	–	0.6	0.0	0.5	0.0	0.1	–	0.3	0.0	0.2	66.6	66.3
Ventral DC L	17.9	0.6	0.6	17.9	20.5	0.4	–	1.3	0.0	1.3	0.1	0.0	–	0.1	–	–	60.8	66.9
Ventral DC R	26.2	0.2	–	16.7	17.9	0.3	–	0.6	0.0	0.6	0.1	0.0	–	0.2	0.1	0.0	62.8	64.1
Ventricles1	40.2	3.3	–	4.9	7.6	–	–	0.0	0.1	0.4	–	–	0.0	0.3	0.0	0.1	56.9	66.9
Lateral L1	39.3	2.3	–	3.6	7.4	–	–	–	0.1	0.5	–	–	–	0.3	0.0	0.0	53.4	61.7
Lateral R1	38.6	2.8	–	4.0	6.9	–	–	–	0.1	0.3	–	–	–	0.2	0.0	0.1	53.0	65.2
Inferior lateral L1	21.7	9.0	0.0	4.4	1.0	0.2	0.0	1.5	0.1	1.1	0.7	0.0	–	0.3	0.1	0.2	40.4	43.4
Inferior lateral R1	16.0	8.9	0.4	3.3	0.2	–	–	1.9	0.0	1.4	–	–	0.2	0.5	0.0	0.2	33.0	32.6
3rd1	42.6	3.5	0.0	7.5	5.3	0.1	–	0.1	0.2	0.1	0.2	0.1	0.1	0.2	–	–	59.9	64.1
4th	0.2	0.7	–	6.4	5.1	0.0	0.1	0.8	0.5	0.0	–	–	–	–	0.1	0.1	13.9	11.4
Corpus callosum	17.7	5.0	0.2	2.0	6.5	0.0	0.1	2.4	0.2	0.0	–	–	0.1	–	0.4	0.1	34.8	32.7
Subcortical GM	41.0	0.1	0.0	15.5	16.8	0.2	–	0.8	0.0	0.5	0.1	0.3	–	0.4	–	–	75.6	72.0

1Log10 transformed. MFS: Magnetic field strength, eTIV: Estimated total intracranial volume. GM: gray matter.

Table 2

Subcortical volumes differences between segmentation on two different computer hardware setups (n=50).

Regions	Mean difference (%)	t	p
Accumbens L	0.05	−0.22	0.825
Accumbens R	1.10	0.97	0.339
Amygdala L	0.95	1.71	0.094
Amygdala R	0.96	1.79	0.080
Brainstem	0.09	0.6	0.552
Caudate L	0.02	−0.05	0.961
Caudate R	0.07	0.18	0.854
Hippocampus L	0.41	1.48	0.144
Hippocampus R	−0.55	−2.01	0.049
Pallidum L	−0.35	−0.46	0.645
Pallidum R	−0.37	−0.73	0.471
Putamen L	0.52	1.3	0.200
Putamen R	0.18	0.65	0.519
Thalamus L	−0.03	−0.17	0.862
Thalamus R	−0.11	−0.44	0.658
Ventral DC L	−0.07	−0.19	0.851
Ventral DC R	0.11	0.36	0.723
Ventricles
All	0.00	−0.18	0.858
Lateral L	−0.01	−0.22	0.830
Lateral R	0.00	0.16	0.874
Inferior lateral L	−0.20	0.65	0.521
Inferior lateral R	0.15	−0.02	0.984
3rd	−0.06	−0.74	0.461
4th	−0.09	−0.09	0.928
Corpus callosum	0.34	0.91	0.366
Subcortical GM	0.10	0.90	0.375

Bonferroni-corrected critical value for significance: .002.

Experimental design, materials and methods

Participants and segmentation

A detailed description of the participants and segmentation procedure can be found in Potvin et al. [1].

Statistical analyses

Regression models predicting subcortical regional volumes were built using age, sex, eTIV, MFS, and scanner manufacturer as predictors. The details about model building can be found in Potvin et al. [1]. Individual predictors׳ weight was measured by squared semi-partial correlations.

The impact of the hardware setup on the volumes generated by FreeSurfer was tested by dependent one-sample t-tests with Bonferroni correction.

Detailed information about the normative statistics included in the Excel spreadsheet can be found in Potvin et al. [1] and in the work of Crawford and colleagues [2], [3].

Subject area	Neuroscience, Neurology, Neurobiology
More specific subject area	Volumetric subcortical normative values
Type of data	Tables, Excel file
How data was acquired	MRI images from open databases, data analyses and normative values generated by statistical models
Data format	Analyzed
Experimental factors	The sociodemographics, the scanner manufacturer and magnetic field strength
Experimental features	Subcortical volumes extracted using FreeSurfer
Data source location	Australia, Austria, Belgium, Canada, Finland, Germany, Ireland, Italy, Netherlands, United Kingdom, and USA
Data accessibility	Data is with this article