Awake Infant fMRI (Ellis 2020)
Re-imagining fMRI for awake behaving infants (Princeton and Yale cohorts)
Task fMRI and T1-weighted PETRA anatomical scans of 26 awake infants aged 4 to 33 months across 45 sessions, from two cohorts scanned at Princeton and Yale by the Turk-Browne lab, released on Dryad under CC0.
Overview
This dataset holds the awake infant fMRI data that the Turk-Browne lab collected while developing a protocol for scanning infants who are awake and watching cognitive tasks, rather than sedated or asleep. Ellis and colleagues describe the apparatus, the session procedure, an experiment menu system and an analysis pipeline in Nature Communications (2020), and deposited the anatomical and functional MRI of both study cohorts on Dryad under CC0, first in July 2020 and last updated in August 2021.
Composition
The paper reports 26 infants who took part in 45 sessions and completed 57 experiments. Cohort I, scanned at Princeton University, has 11 infants (5 female) aged 6 to 33 months over 23 sessions, with 1 to 8 sessions per infant. Cohort II, scanned at Yale University, has 15 infants (8 female) aged 4 to 10 months over 22 sessions, with 1 or 2 sessions per infant. Sessions in which the infant would not lie down are not counted. All tasks used silent visual stimuli.
Acquisition
Cohort I was acquired on a 3 T Siemens Skyra and Cohort II on a 3 T Siemens Prisma, using only the bottom half of a 20-channel head and neck coil so that the infant could see the ceiling-projected screen and be filmed for gaze coding. Anatomy comes from a T1-weighted PETRA sequence at 0.94 mm isotropic. A T2-weighted SPACE scan was tried in two infants. Functional runs are gradient-echo EPI with TR 2 s, TE 28 ms, 36 slices and 3 mm isotropic voxels, without multiband acceleration.
Annotations
There are no image labels. Gaze was coded manually from video of the infant's face, and the pipeline marks time-points and task epochs for exclusion based on head motion and whether the eyes were on screen. A file with the burn-in volumes of each raw run was added to the record in 2021.
Known limitations
- The cohorts are small, and the two sites differ in scanner model and age range.
- In Cohort II, about 4.9 minutes of usable awake functional data were obtained per session on average. Motion sometimes moved the cerebellum and brain stem out of the field of view.
- The order and mix of tasks varies between sessions, because the protocol adapts to each infant's state.
- The paper does not report health or developmental screening criteria.
Cohort
Aggregate numbers from the sources below. Bars are relative to the 26 subjects.
Sex
- Female 13 100%
Covers 13 of 26 subjects.
Age
Contrast / sequence
subjects, values can overlap
- BOLD fMRI 26 100%
Scanner vendor
subjects
- Siemens Healthineers 26 100%
Field strength
subjects
- 3 T 26 100%
Country
subjects
- United States 26 100%
License and access
Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer
Download without an account
Creative Commons Zero 1.0 Universal
Public domain dedication. Do anything with the data, including commercial use, without asking and without having to give credit.
The Dryad record (version 8) states the license https://spdx.org/licenses/CC0-1.0.html.
What you can do
- Yes
- Yes
- Yes
- Yes
- Yes
What you can share
- Yes
- Yes
- Yes
What you must do
- No
- Share alike No
- No
- No
- Manuscript review No
- Release code No
- Return results No
- Delete after use No
Limits
- No
- Location limits No
Citation
Ellis CT, Skalaban LJ, Yates TS, Bejjanki VR, Córdova NI, Turk-Browne NB. Re-imagining fMRI for awake behaving infants. Nature Communications 11:4523 (2020). doi:10.1038/s41467-020-18286-y
All numbers
Every number on this page, as stored in stats.csv, with its source.
| Measure | Breakdown | Value | Source |
|---|---|---|---|
| Subjects | total 11 infants in Cohort I (Princeton) and 15 in Cohort II (Yale), Methods: Participants | 26 | ellis2020 Discussion |
| Studies | total 23 sessions in Cohort I and 22 in Cohort II, Methods: Participants; sessions in which the infant would not lie down are not counted | 45 | ellis2020 Discussion |
| Subjects | sex=female sum of 5 females in Cohort I and 8 in Cohort II (disjoint cohorts) | 13 | ellis2020 Methods: Participants |
| Subjects | age=0-2 Cohort I aged 6 to 33 months and Cohort II aged 4 to 10 months at scan | 26 | ellis2020 Methods: Participants |
| Subjects | contrast=bold all counted sessions were fMRI sessions | 26 | ellis2020 Methods: Participants |
| Subjects | country=US Princeton University and Yale University | 26 | ellis2020 Methods: Participants |
| Subjects | vendor=siemens 3T Siemens Skyra (Cohort I) and 3T Siemens Prisma (Cohort II) | 26 | ellis2020 Results: Data quantity and quality |
| Subjects | field_strength=3 | 26 | ellis2020 Results: Data quantity and quality |
Sources
The keys used in the table above.
- ellis2020 Ellis et al. 2020, Re-imagining fMRI for awake behaving infants, Nature Communications paper
- dryad-8gtht76k3 Dryad record doi:10.5061/dryad.8gtht76k3 (version 8, 2021-08-31) website