NormMoX: A Normative Mixture of Experts for Explainable Autism Classification from Structural and Functional MRI
Organizations: School of Computing Science, University of Glasgow, UK · School of Computing Science, James Watt School of Engineering, University of Glasgow, UK
Abstract
While structural and functional magnetic resonance imaging (sMRI, fMRI) offer profound potential for researching autism spectrum disorder (ASD) and understanding individual neurodivergence, current classifiers lack the explanatory power needed for scientific study, clinical use and personal-level insights. Most models learn from imaging features that also vary with age, sex, scanner site and head motion. They fuse the two modalities without a common reference and explain decisions against arbitrary reference inputs rather than a typical brain. We present NormMoX, a normative mixture of experts that makes the typical brain the shared reference for prediction, fusion and explanation. Normative models trained on large healthy cohorts give the value each brain measure is expected to take for a person's age and sex. NormMoX learns from each person's deviations from these expected values. For each modality, a neural network expert turns the deviations into a score of evidence for ASD. A fusion head, MoX, compares each score with those of typical controls before combining them. It therefore needs no imputation when a modality is missing and shows in which modality each person is atypical. On 1,035 participants of ABIDE I, NormMoX reached an AUROC of in repeated cross-validation, above linear, single-modality and deep fusion baselines. NormMoX thus shows not only whether a person is likely to have ASD, but where, and in which modality, their brain departs from typical. The implementation is available at https://github.com/Novelex/NormMox.
Figures & tables
| Model | AUROC | AUPRC | BAcc | Sens | Spec | F1 |
|---|---|---|---|---|---|---|
| Input ablation: logistic regression | ||||||
| Raw features | 0.675 | 0.632 | 0.607 | 0.657 | 0.616 | |
| ABIDE-normed | 0.678 | 0.634 | 0.614 | 0.654 | 0.620 | |
| Normative deviations | 0.688 | 0.637 | 0.619 | 0.654 | 0.624 | |
| Modality ablation: single expert | ||||||
| sMRI ( , ) | 0.651 | 0.602 | 0.517 | 0.686 | 0.559 | |
| fMRI (136) | sMRI (370) | All (506) | |
| Faithful: AUROC (full model 0.728) | |||
| All inputs reset to typical | |||
| Extra drop, top-30 vs random-30 | |||
| Stable across folds | |||
| Spearman of maps | 0.71 | 0.76 | 0.75 |
| Top-10 overlap / same sign | 0.79 / 1.00 | 0.69 / 1.00 | 0.71 / 1.00 |
Autism Spectrum Disorder
Autism Spectrum Disorder research studies its developmental, behavioral, and neural features and ways to improve assessment and support.
See all 41 papers