Schedule of Events | Symposia

Sketchpad Series

Topological Data Analysis of Pre- and Post-ECT Participant Data

Poster Session D - Monday, March 31, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

Connor M. Harris^1,2, Shane W. Walsh^1,2, Samadrita Chowdhury^1,2, Joan A. Camprodon^1,2; ¹Massachusetts General Hospital, ²Harvard Medical School

Topological Data Analysis (TDA) offers a novel way to analyze high-dimensional datasets by capturing topological features. In this study, TDA is applied to electroconvulsive therapy (ECT) datasets to investigate brain connectivity changes before and after treatment. ECT, a therapy for major depressive disorder (MDD), likely alleviates symptoms by altering brain connectivity, though its mechanisms remain unclear. While graph theory focuses on pairwise relationships in brain networks, it may miss higher-order interactions and topological structure, which TDA could better capture. Analyses were conducted on data from 32 participants (16 male, 16 female; age 33 ± 12) with treatment-resistant MDD who underwent ECT and pre- and post-treatment MRI scans. Resting-state functional MRI data were preprocessed using fmriprep (v.24.1.0) and correlation-based adjacency matrices were created. These matrices were harmonized using NeuroComBat (0.2.12) for consistency between studies. The Python-based MULTINET TDA library generated simplicial complexes via a Vietoris-Rips filtration of each participant's pre- and post-ECT connectivity matrices, capturing topological features like persistence barcodes and Betti numbers. Statistical tests quantified differences between pre- and post-treatment groups. To the best of our knowledge, this study represents the first application of TDA to ECT data. Although no conclusive results are available at this stage, exploratory statistical analyses are being conducted to identify potential topological differences. It is important to note that these analyses are preliminary, and further work will be required to validate any findings and determine whether the observed differences can serve as consistent and replicable biomarkers of ECT-related changes in neural connectivity and functional reorganization.

Topic Area: METHODS: Neuroimaging