# Ensemble learning with 3D convolutional neural networks for connectome-based prediction

arXiv: https://arxiv.org/abs/1809.06219 · HTML: https://mlsanity.com/paper/1809.06219/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Sep 2018.

## Built on by (most cited first, 10 of 10)

- [Challenges for machine learning in clinical translation of big data imaging studies](https://mlsanity.com/paper/2107.05630/) (2021)
- [Spatially Covariant Lesion Segmentation](https://mlsanity.com/paper/2301.07895/) (2023)
- [Automatic autism spectrum disorder detection using artificial intelligence methods with MRI neuroimaging: A review](https://mlsanity.com/paper/2206.11233/) (2022)
- [Predicting Brain Responses To Natural Movies With Multimodal LLMs](https://mlsanity.com/paper/2507.19956/) (2025)
- [Modern Views of Machine Learning for Precision Psychiatry](https://mlsanity.com/paper/2204.01607/) (2022)
- [MADE-for-ASD: A Multi-Atlas Deep Ensemble Network for Diagnosing Autism Spectrum Disorder](https://mlsanity.com/paper/2407.07076/) (2024)
- [Multimodal Brain Disease Classification with Functional Interaction Learning from Single fMRI Volume](https://mlsanity.com/paper/2208.03028/) (2022)
- [BrainMT: A Hybrid Mamba-Transformer Architecture for Modeling Long-Range Dependencies in Functional MRI Data](https://mlsanity.com/paper/2506.22591/) (2025)
- [Beyond Anatomy: Explainable ASD Classification from rs-fMRI via Functional Parcellation and Graph Attention Networks](https://mlsanity.com/paper/2603.02518/) (2026)
- [Multimodal Connectome Fusion via Cross-Attention for Autism Spectrum Disorder Classification Using Graph Learning](https://mlsanity.com/paper/2603.15168/) (2026)

## Abstract (arXiv)

The specificty and sensitivity of resting state functional MRI (rs-fMRI) measurements depend on pre-processing choices, such as the parcellation scheme used to define regions of interest (ROIs). In this study, we critically evaluate the effect of brain parcellations on machine learning models applied to rs-fMRI data. Our experiments reveal a remarkable trend: On average, models with stochastic parcellations consistently perform as well as models with widely used atlases at the same spatial scale. We thus propose an ensemble learning strategy to combine the predictions from models trained on connectivity data extracted using different (e.g., stochastic) parcellations. We further present an implementation of our ensemble learning strategy with a novel 3D Convolutional Neural Network (CNN) approach. The proposed CNN approach takes advantage of the full-resolution 3D spatial structure of rs-fMRI data and fits non-linear predictive models. Our ensemble CNN framework overcomes the limitations of traditional machine learning models for connectomes that often rely on region-based summary statistics and/or linear models. We showcase our approach on a classification (autism patients versus healthy controls) and a regression problem (prediction of subject's age), and report promising results.