Skip to content

Latest commit

 

History

History
52 lines (38 loc) · 2.91 KB

README.md

File metadata and controls

52 lines (38 loc) · 2.91 KB

MEGHACK_2023_spring

Location info

https://megcore.nih.gov/index.php/MEG_Hackathon_2023

Github SSH key setup

https://github.com/nih-megcore/MEGHACK_2023_spring/wiki/Connect-To-A-Github-Repository-Using-SSH

Ideas

Eyetracking v ground truth

Co-Leads: Rohini Kumar and Sebastion Montesinos
https://github.com/sebmonte/MEG-Hackathon/tree/main
-- 4 subject datasets are available in the above github repository
-- Actual positions of the stimuli are in the results folder. Eye tracking data is available in the UADC channels
-- Objective - Minimize the average error between the on-screen eye fixation mark and calculated eye location
-- Fork this repository: https://github.com/nih-megcore/nih_to_mne.git
---- Use this as a base file for ideas: nih2mne/eyetracking_preprocessing.py

Epilepsy spike spread vs white matter connectivity

Lead: Price Withers

  • Goal: Replicate figures 3A-B and 4A-B from Integration of white matter architecture to stereo-EEG better describes epileptic spike propagation using MEG data
  • Dataset: 10-15 MEG patients w/ and w/out patient-specific DTI data. Spikes have been marked ahead of time and SC matrices have been created in advance.
  • 2 group members will determine latencies between "Propagation Pairs". The initial spiking region from averaged IEDs will have a latency of 0ms and all other regions which show significant activation will have their own latency at the time of activion. Goal output: 1 vector of size n_parcels per patient with latencies of spiking. Majority of values will be NaN and timings will range from 0ms to ~100ms.
  • 2 group members will write code to convert latency vectors and DWI connectivity matrices into figures 3A-B and 4A-B. Since this will not be available until end of day, they can use random vectors until data are prepared. Additional analyses can be prepared to incorporate pre- and post-operative DWI data and seizure free/persist outcomes.

Harmonization Procedures for consortium data

Allison Nugent
-- Use combat/covbat/etc on initial open MEG data from the Enigma MEG Project

Develop interactive dashboard for exploring MEG datasets

Jeff Stout and Arshitha Basavaraj
-- Build a dashboard in Shiny python https://shiny.rstudio.com/py/
-- https://github.com/jstout211/MEGHack_dashboard

Other projects for next time

Statistical Group analysis

(No Lead Currently)
-- Use the HV data: https://openneuro.org/datasets/ds004215/versions/1.0.2
-- Some of the datasets will be pre-processed to source activations
-- Use the online pages to calculate group statistics on the data: https://mne.tools/stable/statistics.html

Reaction time v activation

(No Lead)
-- HV data can be used and will be prepped
-- Assess reaction time versus stimulus time
-- Regress stimulus time versus activation over group data