# Whole MILC: generalizing learned dynamics across tasks, datasets, and populations

arXiv: https://arxiv.org/abs/2007.16041 · HTML: https://mlsanity.com/paper/2007.16041/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jul 2020.

## Built on by (most cited first, 11 of 11)

- [Transfer Learning in Magnetic Resonance Brain Imaging: a Systematic Review](https://mlsanity.com/paper/2102.01530/) (2021)
- [A deep learning model for data-driven discovery of functional connectivity](https://mlsanity.com/paper/2112.04013/) (2021)
- [Tasting the cake: evaluating self-supervised generalization on out-of-distribution multimodal MRI data](https://mlsanity.com/paper/2103.15914/) (2021)
- [Evaluating deep transfer learning for whole-brain cognitive decoding](https://mlsanity.com/paper/2111.01562/) (2021)
- [Deep Dynamic Effective Connectivity Estimation from Multivariate Time Series](https://mlsanity.com/paper/2202.02393/) (2022)
- [Variational voxelwise rs-fMRI representation learning: Evaluation of sex, age, and neuropsychiatric signatures](https://mlsanity.com/paper/2108.12756/) (2021)
- [Brain dynamics via Cumulative Auto-Regressive Self-Attention](https://mlsanity.com/paper/2111.01271/) (2021)
- [Multi network InfoMax: A pre-training method involving graph convolutional networks](https://mlsanity.com/paper/2111.01276/) (2021)
- [Self-supervised multimodal neuroimaging yields predictive representations for a spectrum of Alzheimer's phenotypes](https://mlsanity.com/paper/2209.02876/) (2022)
- [Self-Supervised Mental Disorder Classifiers via Time Reversal](https://mlsanity.com/paper/2211.16398/) (2022)
- [Rhamba: Region-Aware Hybrid Attention-Mamba Framework for Self-Supervised Learning in Resting-State fMRI](https://mlsanity.com/paper/2605.01240/) (2026)

## Abstract (arXiv)

Behavioral changes are the earliest signs of a mental disorder, but arguably, the dynamics of brain function gets affected even earlier. Subsequently, spatio-temporal structure of disorder-specific dynamics is crucial for early diagnosis and understanding the disorder mechanism. A common way of learning discriminatory features relies on training a classifier and evaluating feature importance. Classical classifiers, based on handcrafted features are quite powerful, but suffer the curse of dimensionality when applied to large input dimensions of spatio-temporal data. Deep learning algorithms could handle the problem and a model introspection could highlight discriminatory spatio-temporal regions but need way more samples to train. In this paper we present a novel self supervised training schema which reinforces whole sequence mutual information local to context (whole MILC). We pre-train the whole MILC model on unlabeled and unrelated healthy control data. We test our model on three different disorders (i) Schizophrenia (ii) Autism and (iii) Alzheimers and four different studies. Our algorithm outperforms existing self-supervised pre-training methods and provides competitive classification results to classical machine learning algorithms. Importantly, whole MILC enables attribution of subject diagnosis to specific spatio-temporal regions in the fMRI signal.