arXiv subjects

Christine Ahrends

Publications and source records attributed to Christine Ahrends.

5 recordsLinked to original sources

arXiv2025

Stacked Regression using Off-the-shelf, Stimulus-tuned and Fine-tuned Neural Networks for Predicting fMRI Brain Responses to Movies (Algonauts 2025 Report)

We present our submission to the Algonauts 2025 Challenge, where the goal is to predict fMRI brain responses to movie stimuli. Our approach integrates multimodal representations from large language models, video encoders, audio models, and vision-language models, combining both off-the-shelf and fine-tuned variants. To improve performance, we enhanced textual inputs with detailed transcripts and summaries, and we explored stimulus-tuning and fine-tuning strategies for language and vision models. Predictions from individual models were combined using stacked regression, yielding solid results. Our submission, under the team name Seinfeld, ranked 10th. We make all code and resources publicly available, contributing to ongoing efforts in developing multimodal encoding models for brain activity.

eess.IV↗

arXiv2025

BrainATCL: Adaptive Temporal Brain Connectivity Learning for Functional Link Prediction and Age Estimation

Functional Magnetic Resonance Imaging (fMRI) is an imaging technique widely used to study human brain activity. fMRI signals in areas across the brain transiently synchronise and desynchronise their activity in a highly structured manner, even when an individual is at rest. These functional connectivity dynamics may be related to behaviour and neuropsychiatric disease. To model these dynamics, temporal brain connectivity representations are essential, as they reflect evolving interactions between brain regions and provide insight into transient neural states and network reconfigurations. However, conventional graph neural networks (GNNs) often struggle to capture long-range temporal dependencies in dynamic fMRI data. To address this challenge, we propose BrainATCL, an unsupervised, nonparametric framework for adaptive temporal brain connectivity learning, enabling functional link prediction and age estimation. Our method dynamically adjusts the lookback window for each snapshot based on the rate of newly added edges. Graph sequences are subsequently encoded using a GINE-Mamba2 backbone to learn spatial-temporal representations of dynamic functional connectivity in resting-state fMRI data of 1,000 participants from the Human Connectome Project. To further improve spatial modeling, we incorporate brain structure and function-informed edge attributes, i.e., the left/right hemispheric identity and subnetwork membership of brain regions, enabling the model to capture biologically meaningful topological patterns. We evaluate our BrainATCL on two tasks: functional link prediction and age estimation. The experimental results demonstrate superior performance and strong generalization, including in cross-session prediction scenarios.

cs.LG↗

arXiv2025

A comprehensive framework for statistical testing of brain dynamics

Neural activity data can be associated with behavioral and physiological variables by analyzing their changes in the temporal domain. However, such relationships are often difficult to quantify and test, requiring advanced computational modeling approaches. Here, we provide a protocol for the statistical analysis of brain dynamics and for testing their associations with behavioral, physiological and other non-imaging variables. The protocol is based on an open-source Python package built on a generalization of the hidden Markov model (HMM) - the Gaussian-linear HMM - and supports multiple experimental modalities, including task-based and resting-state studies, often used to explore a wide range of questions in neuroscience and mental health. Our toolbox is available as both a Python library and a graphical interface, so it can be used by researchers with or without programming experience. Statistical inference is performed by using permutation-based methods and structured Monte Carlo resampling, and the framework can easily handle confounding variables, multiple testing corrections and hierarchical relationships within the data, among other features. The package includes tools developed to facilitate the intuitive visualization of statistical results, along with comprehensive documentation and step-by-step tutorials for data interpretation. Overall, the protocol covers the full workflow for the statistical analysis of functional neural data and their temporal dynamics.

q-bio.NC↗

arXiv2024

The Gaussian-Linear Hidden Markov model: a Python package

We propose the Gaussian-Linear Hidden Markov model (GLHMM), a generalisation of different types of HMMs commonly used in neuroscience. In short, the GLHMM is a general framework where linear regression is used to flexibly parameterise the Gaussian state distribution, thereby accommodating a wide range of uses -- including unsupervised, encoding and decoding models. GLHMM is implemented as a Python toolbox with an emphasis on statistical testing and out-of-sample prediction -- i.e. aimed at finding and characterising brain-behaviour associations. The toolbox uses a stochastic variational inference approach, enabling it to handle large data sets at reasonable computational time. The approach can be applied to several data modalities, including animal recordings or non-brain data, and applied over a broad range of experimental paradigms. For demonstration, we show examples with fMRI, electrocorticography, magnetoencephalography and pupillometry.

q-bio.NC↗

arXiv2023

Dynamic Functional Connectivity

Most generally, dynamic functional connectivity (FC) refers to the non-instantaneous couplings across timeseries from a set of brain areas, here as measured by fMRI. This is in contrast to static FC, which is defined as purely instantaneous relations. In this chapter, we provide a hands-on description of a non-exhaustive selection of different methods used to estimate dynamic FC (such as sliding windows, clustering approaches, Hidden Markov Models, and multivariate autoregressive models), and we explain, using practical examples, how data should be prepared for dynamic FC analyses and how models of dynamic FC can be evaluated. We also discuss current developments in the dynamic FC research field, including challenges of reliability and reproducibility, and perspectives of using dynamic FC for prediction.

q-bio.NC↗