Two of our graduate students, Theo and Erica, recently attended Neurohackademy, a two-week program that provides training in neuroimaging and data science. The program featured a full schedule of lectures and hands-on activities designed to teach researchers the latest techniques and methods for analyzing neuroimaging data.
One of the topics that Erica found particularly interesting was “Measuring and Analyzing Human Functional Brain Networks,” which covered methods for conducting functional connectivity analysis with real data. This is something that Erica plans to do with the MLINDIV data. Another topic that caught her attention was “Machine Learning Methods for Neuroimaging,” which covered how to create and analyze models using real brain data to predict certain outcome measures and then validate the effectiveness of those models.
During the second week of the program, Theo and Erica worked on a team project where they investigated the relationship between metabolic output and brain activity using connectivity measures. Specifically, Erica completed a dynamic community detection analysis and found that the flexibility of the nodes in the default mode network were negatively correlated with FDG-PET values. Overall, Erica and Theo had a great time at Neurohackademy and would recommend it to any student who is interested in learning computational methods for analyzing neuroimaging data. We are proud of Theo and Erica for their participation in Neurohackademy and look forward to seeing how they apply their newly acquired knowledge and skills in their research!
