Uncovering the Topology of Time-Varying fMRI Data using Cubical Persistence
Authors: Bastian Rieck, Tristan Yates, Christian Bock, Karsten Borgwardt, Guy Wolf, Nicholas Turk-Browne, Smita Krishnaswamy
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We apply both clustering and trajectory analysis techniques to a group of participants watching the movie Partly Cloudy. We observe significant differences in both brain state trajectories and overall topological activity between adults and children watching the same movie. |
| Researcher Affiliation | Academia | Bastian Rieck Dept. Biosystems (D-BSSE) ETH Zurich & Swiss Institute of Bioinformatics, Switzerland Tristan Yates Dept. of Psychology Yale University New Haven, CT, USA Christian Bock Dept. Biosystems (D-BSSE) ETH Zurich & Swiss Institute of Bioinformatics, Switzerland Karsten Borgwardt Dept. Biosystems (D-BSSE) ETH Zurich & Swiss Institute of Bioinformatics, Switzerland Guy Wolf Dept. of Math. and Stat. Univ. de Montréal; Mila Montreal, QC, Canada Nicholas Turk-Browne Dept. of Psychology Yale University New Haven, CT, USA Smita Krishnaswamy Depts. of Gene. & Comp. Sci. Yale University New Haven, CT, USA |
| Pseudocode | No | The paper does not contain explicitly labeled pseudocode or algorithm blocks. The method is described textually and through diagrams like Figure 1. |
| Open Source Code | Yes | We make our code publicly available4 to ensure reproducibility. 4https://github.com/BorgwardtLab/fMRI_Cubical_Persistence |
| Open Datasets | Yes | We evaluate our topological pipeline using open-source f MRI data [46], available on the Open Neuro database (accession number ds000228). [46] H. Richardson, G. Lisandrelli, A. Riobueno-Naylor, and R. Saxe. Development of the social brain from age three to twelve years. Nature Communications, 9(1):1 12, 2018. |
| Dataset Splits | Yes | Using a ridge regression and leave-one-out cross-validation (see Section A.4 for detailed descriptions of all comparison partners and Section A.5 for additional experimental details), we train models on either the curves of summary statistics (not the embeddings) the baseline matrices, and additional topological baselines, reporting the correlation coefficient in the table in Figure 3. |
| Hardware Specification | No | The paper does not explicitly describe the hardware used to run its experiments. |
| Software Dependencies | No | The paper mentions using "DIPHA [7]" and "Scikit-TDA [53]" but does not provide specific version numbers for these software components. It also refers to scikit-learn but again without a version. |
| Experiment Setup | Yes | Subsequently, we use r = 20 and a Gaussian kernel with σ = 1.0; Ψ is known to be impervious to such choices [1]. Using a ridge regression and leave-one-out cross-validation (see Section A.4 for detailed descriptions of all comparison partners and Section A.5 for additional experimental details)... |