4. How data moves through the lab¶

Every dataset the lab holds went through the same seven stages. Each stage links to the relevant themes in the wiki; the arrows are the hand-offs a coordinator owns.

flowchart LR
  A["1 Acquisition<br>Imaging · Electrophysiology"] --> B["2 Landing<br>cnt-fs / cnt1"]
  S["Site delivery<br>15+ external sites"] --> B
  B --> C["3 Standardization<br>BIDS · Data Structure v1.0"]
  C --> D["4 De-identification<br>EDF · headers · NIfTI"]
  D --> E["5 Derivatives<br>ieeg_recon · FreeSurfer · VoxTool"]
  R["REDCap · RADAR/EPIC"] --> F["6 Clinical metadata joins"]
  D --> F
  E --> G["7 Sharing & archive<br>Pennsieve · ieeg.org · Azure"]
  F --> G

The stages¶

Acquisition — a prospective HUP patient is scanned (3T Research Scans: Scheduling & Billing) or recorded in the EMU (sEEG Phase II Processing: Overview & Timeline); or a collaborating site delivers a batch.
Landing — raw data reaches cnt-fs/cnt1: Flywheel → cnt-fs (3T), Exporting Files from Natus, Moving Data Across cnt-fs, cnt1, BSC, Borel & Leif. Where each kind lives: Data Storage Locations.
Standardization — into the BIDS layout defined in the lab's Data Structure v1.0 specification (being written into Data › Standards); channel maps for EEG: Automated Channel Mapping.
De-identification — before data moves to a destination that requires de-identified data: De-Identifying EDFs and the header and NIfTI scripts beside it.
Derivatives — Electrode Reconstruction: Prep & Software, FreeSurfer, VoxTool outputs into derivatives/.
Clinical metadata joins — Surgical Outcomes REDCap Project, RADAR Pull → REDCap Entry.
Sharing and archive — Processing for ieeg.org (natus2mef → validate → upload), Uploading from cnt1 to Pennsieve, Archiving EEG Data to Azure.

Warning

The line that matters: identified data may be handled on approved PMACS systems, including cnt1, cnt-fs and BSC. BSC permission was confirmed by Nishant Sinha on 2026-09-20. Nothing crosses to Borel, Pennsieve, ieeg.org, GitHub or this wiki until stage 4 is done.