Comparing Post-Hoc Explainable AI Methods for Interpreting Black-Box EEG Models in Depression Detection

• URL: http://arxiv.org/abs/2605.28977v1
• Date: Wed, 27 May 2026 18:32:57 GMT
• Title: Comparing Post-Hoc Explainable AI Methods for Interpreting Black-Box EEG Models in Depression Detection
• Abstract summary: This study investigates multiple post-hoc explainability methods applied to an InceptionTime architecture trained for EEG-based depression detection.<n>The evaluated methods revealed partially convergent attribution patterns, with recurring emphasis on frontal, temporal, and posterior EEG regions.<n>Overall, the results suggest that different post-hoc explainability approaches capture partially overlapping relevance structures in EEG-based deep learning models for depression detection.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in deep learning have enabled increasingly accurate electroencephalography (EEG)-based classification of Major Depressive Disorder (MDD), but the decision-making processes of high-capacity models remain difficult to interpret. This study investigates multiple post-hoc explainability methods applied to an InceptionTime architecture trained for EEG-based MDD detection. The analysis includes Shapley-based, gradient-based, and perturbation-based attribution approaches: DeepSHAP, Integrated Gradients, GradCAM, Occlusion, and Permutation Feature Importance. Explainability analysis was performed within a subject-level stratified 5-fold cross-validation framework using global attribution aggregation across EEG segments and subjects. The evaluated methods revealed partially convergent attribution patterns, with recurring emphasis on frontal, temporal, and posterior EEG regions, particularly in the right hemisphere. Quantitative comparison demonstrated substantial agreement between gradient- and perturbation-based approaches, while DeepSHAP produced comparatively distinct attribution distributions. At the same time, variability between explainability methods highlighted the influence of methodological assumptions on the resulting explanations. Overall, the results suggest that different post-hoc explainability approaches capture partially overlapping relevance structures in EEG-based deep learning models for depression detection. Although the observed attribution patterns are broadly consistent with several previous EEG studies of MDD, the analysis should be interpreted as exploratory rather than evidence of definitive neurophysiological biomarkers or clinical applicability. The study highlights both the usefulness and limitations of post-hoc explainability for interpreting black-box EEG classifiers in psychiatric applications.

Related papers

• EEG-Xplain: Decoding Neural Black-Boxes of EEG Foundation Models [8.430851504111585]
  EEG foundation models such as BIOT, LaBraM, and EEGMamba have achieved remarkable performance in neural signal decoding.<n>We propose a unified attribution framework for interpreting EEG foundation models across heterogeneous architectures.
  arXiv  Detail & Related papers  (2026-09-14T14:56:07Z)
• Dementia Etiology Diagnosis via Collaborative Meta Knowledge Enhancement [81.27196033392559]
  We propose a Collaborative Meta Knowledge Enhancement (COME) framework for dementia etiology diagnosis.<n>COME injects multi-center acquisition semantics, source identifiers, and modality indicators as heterogeneity-aware embeddings into a unified Transformer architecture for scale-up training.<n>Our method achieves state-of-the-art in-domain performance with a mean macro-averaged AUC of 85.62% and a 4.29-point gain over the strongest baseline.
  arXiv  Detail & Related papers  (2026-07-24T04:26:47Z)
• ConceptM$^3$oE: Concept-Guided Multimodal Mixture of Experts for Interpretable Computational Pathology [33.27476666910628]
  ConceptM$3$oE embeds concept formation directly within interaction-aware mixture-of-experts pathways.<n>The architecture decomposes evidence into modality-specific, redundant, and synergistic experts, which are then projected into structured concept bottlenecks.<n>In data-limited regimes, ConceptM$3$oE improves limited-data performance, increasing macro-F1 from 56.41% to 66.70% at small training sizes.
  arXiv  Detail & Related papers  (2026-05-23T04:55:16Z)
• Multimodal Graph-based Classification of Esophageal Motility Disorders [73.90451172929117]
  Diagnosing esophageal motility disorders pose significant challenges due to the complexity of high-resolution impedance manometry data and variability in clinical interpretation.<n>This work explores the feasibility of a multimodal Machine Learning-based classification approach that combines HRIM recordings with patient-specific information and incorporates a graph-based modeling of esophageal physiology.
  arXiv  Detail & Related papers  (2026-05-13T14:52:12Z)
• Clinical Cognition Alignment for Gastrointestinal Diagnosis with Multimodal LLMs [63.535652574541764]
  Multimodal Large Language Models (MLLMs) have demonstrated remarkable potential in medical image analysis.<n>Their application in gastrointestinal endoscopy is currently hindered by two critical limitations.<n>We propose a novel Clinical-Cognitive-Aligned (CogAlign) framework to address these challenges.
  arXiv  Detail & Related papers  (2026-03-21T07:47:37Z)
• PIME: Prototype-based Interpretable MCTS-Enhanced Brain Network Analysis for Disorder Diagnosis [29.231312772459237]
  PIME is an interpretable framework that bridges intrinsic interpretability with minimal-sufficient subgraph optimization.<n> Experiments on three benchmark fMRI datasets demonstrate that PIME achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-02-24T16:04:52Z)
• Investigating the Impact of Histopathological Foundation Models on Regressive Prediction of Homologous Recombination Deficiency [52.50039435394964]
  We systematically evaluate foundation models for regression-based tasks.<n>We extract patch-level features from whole slide images (WSI) using five state-of-the-art foundation models.<n>Models are trained to predict continuous HRD scores based on these extracted features across breast, endometrial, and lung cancer cohorts.
  arXiv  Detail & Related papers  (2026-01-29T14:06:50Z)
• A Knowledge-Driven Cross-view Contrastive Learning for EEG Representation [48.85731427874065]
  This paper proposes a knowledge-driven cross-view contrastive learning framework (KDC2) to extract effective representations from EEG with limited labels. The KDC2 method creates scalp and neural views of EEG signals, simulating the internal and external representation of brain activity. By modeling prior neural knowledge based on neural information consistency theory, the proposed method extracts invariant and complementary neural knowledge to generate combined representations.
  arXiv  Detail & Related papers  (2023-09-21T08:53:51Z)
• Pain level and pain-related behaviour classification using GRU-based sparsely-connected RNNs [61.080598804629375]
  People with chronic pain unconsciously adapt specific body movements to protect themselves from injury or additional pain. Because there is no dedicated benchmark database to analyse this correlation, we considered one of the specific circumstances that potentially influence a person's biometrics during daily activities. We proposed a sparsely-connected recurrent neural networks (s-RNNs) ensemble with the gated recurrent unit (GRU) that incorporates multiple autoencoders. We conducted several experiments which indicate that the proposed method outperforms the state-of-the-art approaches in classifying both pain level and pain-related behaviour.
  arXiv  Detail & Related papers  (2022-12-20T12:56:28Z)
• Tensor Decomposition of Large-scale Clinical EEGs Reveals Interpretable Patterns of Brain Physiology [0.13980986259786218]
  We propose a tensor decomposition approach to discover a parsimonious set of population-level EEG patterns. We validate their clinical value using a cohort of patients including varying stages of cognitive impairment.
  arXiv  Detail & Related papers  (2022-11-24T20:39:22Z)
• fMRI from EEG is only Deep Learning away: the use of interpretable DL to unravel EEG-fMRI relationships [68.8204255655161]
  We present an interpretable domain grounded solution to recover the activity of several subcortical regions from multichannel EEG data. We recover individual spatial and time-frequency patterns of scalp EEG predictive of the hemodynamic signal in the subcortical nuclei.
  arXiv  Detail & Related papers  (2022-10-23T15:11:37Z)
• EEG-based Cross-Subject Driver Drowsiness Recognition with an Interpretable Convolutional Neural Network [0.0]
  We develop a novel convolutional neural network combined with an interpretation technique that allows sample-wise analysis of important features for classification. Results show that the model achieves an average accuracy of 78.35% on 11 subjects for leave-one-out cross-subject recognition.
  arXiv  Detail & Related papers  (2021-05-30T14:47:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.