When adding a new project to the listing, please copy and paste the template below.
Project url(s): [link to GitHub repo or other resources]
Contributors: [list of people involved], NAME
Description of project: [a few sentences describing the project]
How to get involved: [optional explanation of how one can get involved in the project]
Project url(s): GitHub repo, Slides
Contributors: Marisa Lytle, Erin Neaton, Veronica Porubsky, Madison Thomasson, Avery Van De Water
Description of project: The goal of the project is to create a predictive model to identify fMRI signals associated with the viewing of social and nonsocial naturalistic images. Stimuli and fMRI data were obtained from the Natural Scenes Dataset (Allen, St-Yves, Wu, Breedlove, Prince, Dowdle, Nau, Caron, Pestilli, Charest, Hutchinson, Naselaris*, & Kay*. A massive 7T fMRI dataset to bridge cognitive neuroscience and artificial intelligence. Nature Neuroscience (2022).
How to get involved: Email [email protected]
Project url(s): GitHub, Presentation Slides, Natural Scenes Dataset
Contributors: Jeanne Barthélemy, Rhideeta Jalal, Jessica Kraft, Yutong Li, Zhen-Qi Liu, Jacqueline Quirke, Kirsten Rhittershofer, Karen Shen, Mike Starrett Ambrose, Haley Wang
Description of project: Our project aims to explore the cognitive and neural mechanisms underlying encoding and recognition (familiarity) memory processes. To this end, we have begun analzying data from the Natural Scenes Dataset (NSD). We aim to analyze behavioral performance using signal detection approaches to characterize memory function and associated effects on reaction times. We are also conducting task-based analyses including functional connectivity (using resting state to test our framework) and multivariate classification using linear support vector machines and convolutional neural networks. ROIs of interest include hippocampus, retrosplenial cortex, and whole brain data.
How to get involved: To-date, we have used a branching workflow. You may access the public GitHub repository (listed above), clone it, contribute and initiate a pull request to merge your changes into the main analysis pipeline.
Project url(s): Github Repo
Contributors: Steven Meisler, Emily Kubota, Chenying Zhao, Alicja Olszewska, Hamsi Radhakrishnan, Brad Caron, Drew Winters, Daniela Cossio, McKenzie Paige Hagen, Tashrif Billah
Description of project: Software that functionally segments white matter connections to generate task-specific subcomponents of fiber bundles.
How to get involved: Post an issue or submit a PR.
Project url(s): Github Repo
Contributors: Patrycja Dzianok
Description of project: A step-by-step pipeline for using NiMARE Python package for neuroimaging meta-analysis. The project shows functionality of NiMARE and sample (basic/undergoing) meta-analysis results of fMRI studies regarding dementia. Additionally, during the Neurohackademy I created a list (SORTED) of interesting neuroscience ideas, tools and links (the idea was born out of Slack discussions with other participants) that will (hopefully) be updated on an ongoing basis and can serve neuroscience community (especially MSc/PhD students and postdocs).
How to get involved: Post an issue or submit a PR.
Project url(s): Github Repo
Contributors: Ana Arsenovic
Description of project: The goal of this project is to investigate the reliability of Freesurfer outputs in individual subjects in the Natural Scenes Dataset (NSD; Allen et al., 2022). The dataset contains in total 8 subjects with multiple 0.8 mm T1 acquisitions each (ranging from 4 to 10). NSD Data Manual is available here.
How to get involved: Post an issue or submit a PR.
Project url(s): Repo with discussion boards; website
Contributors: Team: Sana Ali, Clare McCann, Monica Thieu, Lucy Whitmore
Description of project: A resource hub for ABCD researchers filled with frequently asked questions, code troubleshooting, archived Slack or related message board threads from past workshops, and associated tweets.
How to get involved: Submit a pull request! 🙏
Project url(s): Repo
Contributors: Max van den Boom, Jiyun Shin, Romina Ambrosini, Zach Ladwig, Bram Diamond, Bert Liu, Liberty Hamilson
Description of project: Comparison of functional connectivity, as measured by iEEG and fMRI, in the same subjects watching the same movies. Findings are controversial.
How to get involved: Talk to Max van den Boom or drop by Alder 105.
Project url(s): Github Repo
Contributors: Sajjad Torabian, Setayesh Radkani
Description of project: Using images from the Natural Scenes Dataset, we investigated the neural representations of
implied motion and social relationship. We manually annotated the images and performed analyses of SVM classification & Representational Similarity Analysis (RSA).
How to get involved: Interested in getting involved? Submit a pull request!
Name: Paranormal-events
Project url(s): Github Repo
Contributors: John Andrew Chwe, Ana Fouto, Jiawen Huang, Clara Sava-Segal
Description of project: Previous research have shown a link between individual variations in brain activity to clinical traits 1. For example, individuals that shared clinical traits should process information more similarly. Researchers examined trait level paranoia in this dataset and then had individuals listen to a paranoid-inducing scenario in the scanner, finding that persons with comparable degrees of trait level paranoia processed information in similar ways. This study served as a proof of concept that clinical traits could be linked to neural information processing. There is also evidence that participants chunk continuous stimuli into discrete chunks, and recent algorithms can model the switching between event states. In this project, we aimed to apply such techniques to assess whether clinical traits can be linked to event segmentation, such that individuals more similar in trait paranoia demonstrate more similar event boundaries.
How to get involved: Post an issue or submit a PR.
Name: MLPtryout
Project url: Github Repo,
Presentation Slides,
AOMIC Dataset
Contributors: Raj V Jain, Selin Topel,
Maïwenn Fleig, Woon Ju Park,
Chelsea Xu, Nadine Spychala
Description of project: AOMIC-ID1000 contains about 1000 subjects which have done a video viewing task. They have
fMRI signals, T1w images, physiology data and a battery of tests outside the scanner. We have two aims. One is to predict
the audio features from fMRI and anatomical features. Other is to predict out-of-scanner scores.
How to get involved: Post an issue or submit a PR.