This website hosts the Precision in Neuroimaging (PIN) study dataset. The dataset includes MRI scans of the brain from individuals scanned at multiple time points and on different scanners.

Please see our two current papers published on this database:

NeuroImage: Reliability of structural MRI measurements: The effects of scan session, head tilt, inter-scan interval, acquisition sequence, FreeSurfer version and processing stream by Hedges et al.

Psychiatry Research: Neuroimaging: Impact of physiological factors on longitudinal structural MRI measures of the brain by Zahid et al.

The dataset includes raw MRI from 24 subjects who gave their consent for their defaced data to be made publicly available. There is also demographic and physiological data available for each subject and visit including: blood pressure, hydration, caffeine intake, height, weight and hours of sleep.

All images were acquired with an ADNI3 accelerated T1 image on the same GE MR750 Discovery 3 Tesla Scanner unless otherwise stated. Session A was at baseline, session B was at 3 weeks and C and D were at 1 year later.

A1= Baseline image

A2= Repetition (rescan)

A3 = Reposition (participant leaves scanner room and re-enters)

A4= Head tilt forward 10 degrees

B1= Scan after 3 weeks

C1= Scan at 1 year

C3= Scan at 1 year with ADNI2 sequence

D1= Scan at 1 year on a different GE MR750 Discovery 3 Tesla Scanner

Download Data

Download PIN MRI dataset

Download PIN demographic and physiological data

Please also see the GitHub page by M Dimitrov which has a copy of the R scripts used in the reliability analysis: https://github.com/misho-dimitrov/pin_study

FreeSurfer results using the PIN datatset (Values are intraclass correlation coefficients, ICCs,

see our paper above for further information)

Mean ICC values over regions using FreeSurfer 7.1.0 longitudinal stream

Mean ICC values for 3 week comparison for different FreeSurfer versions and stream

Milly Hedges, Uzma Zahid, Mihail Dimitrov and Matthew Kempton