Explainable AI techniques for dynamic functional brain imaging: validation and analysis of E/I imbalance in autism
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Deep neural networks are increasingly crucial for analysing dynamic functional brain imaging data, offering unprecedented accuracy in distinguishing brain activity patterns across health and disease. However, they often function as black boxes obscuring the neurobiological features driving classifications between groups. This study systematically investigates explainable AI (xAI) methods to address this challenge, employing two complementary simulation approaches: recurrent neural networks for controlled parameter exploration, and The Virtual Brain for biophysically realistic whole-brain modelling. These simulations generate fMRI datasets with known regional alterations in excitation/inhibition (E/I) balance, mimicking mechanisms implicated in psychiatric and neurological disorders. Our comprehensive validation demonstrates that Integrated Gradients and DeepLift successfully identify ground-truth affected regions across challenging conditions, including high noise (−10dB SNR), low prevalence (1% of regions), and subtle E/I alterations. This performance remains robust across three different attribution methods and baseline choices, establishing the reliability of xAI for functional neuroimaging analysis. Critically, successful cross-species validation using both human (68-region) and mouse (426-region) connectomes demonstrates the approach’s ability to detect mechanistic alterations across different scales of brain organization. Application to the multisite ABIDE resting-state fMRI dataset (N=834) reveals that regions within the default mode network, particularly the posterior cingulate cortex and precuneus, most clearly differentiated children with autism from neurotypical controls. The convergence between these empirical findings and our biophysical simulations of E/I imbalance provides computational support for mechanistic theories of E/I imbalance in autism while demonstrating how xAI can bridge cellular-level mechanisms with clinical biomarkers. This work establishes a framework for reliable interpretation of deep neural network models in functional neuroimaging, with implications for understanding brain disorders and developing targeted brain stimulation strategies.
Author Summary
Our research tackles a key challenge in brain imaging analysis: making complex artificial intelligence methods more transparent and biologically meaningful. When scientists and clinicians use AI to analyse functional brain scans, these powerful tools can accurately discriminate conditions like autism, but they typically don’t explain which brain regions are involved or why. This creates a “black box” problem that limits our understanding of brain disorders. We developed and tested methods that reveal which brain areas contribute most to AI decisions when analysing brain activity. Using both computer simulations and real patient data, we demonstrate that these methods can reliably identify regions where the balance between excitation and inhibition is disrupted - a mechanism thought to underlie autism and other conditions. Our findings specifically highlight the posterior cingulate cortex and precuneus, key regions of the “default mode network” involved in self-directed and social cognition. This work contributes to bridging the gap between cellular-level mechanisms and whole-brain patterns observed in clinical scans, potentially helping scientists develop more targeted treatments and better understand the biological basis of brain disorders.
