Unsupervised learning of acquisition variability in structural connectomes via hybrid latent space modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rudravaram, Gaurav, Zuo, Lianrui, Ramadass, Karthik, McMaster, Elyssa, Yoon, Jongyeon, Krishnan, Aravind R., Saunders, Adam M., Gao, Chenyu, Newlin, Nancy R., Kanakaraj, Praitayini, Held, Lori L. Beason, Bilgel, Murat, Barquero, Laura A., DArchangel, Micah, Nguyen, Tin Q., Cutting, Laurie B., Archer, Derek, Hohman, Timothy J., Moyer, Daniel C., Landman, Bennett A.
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911683400695808
author Rudravaram, Gaurav
Zuo, Lianrui
Ramadass, Karthik
McMaster, Elyssa
Yoon, Jongyeon
Krishnan, Aravind R.
Saunders, Adam M.
Gao, Chenyu
Newlin, Nancy R.
Kanakaraj, Praitayini
Held, Lori L. Beason
Bilgel, Murat
Barquero, Laura A.
DArchangel, Micah
Nguyen, Tin Q.
Cutting, Laurie B.
Archer, Derek
Hohman, Timothy J.
Moyer, Daniel C.
Landman, Bennett A.
author_facet Rudravaram, Gaurav
Zuo, Lianrui
Ramadass, Karthik
McMaster, Elyssa
Yoon, Jongyeon
Krishnan, Aravind R.
Saunders, Adam M.
Gao, Chenyu
Newlin, Nancy R.
Kanakaraj, Praitayini
Held, Lori L. Beason
Bilgel, Murat
Barquero, Laura A.
DArchangel, Micah
Nguyen, Tin Q.
Cutting, Laurie B.
Archer, Derek
Hohman, Timothy J.
Moyer, Daniel C.
Landman, Bennett A.
contents Acquisition differences across sites, scanners, and protocols in dMRI introduce variability that complicates structural connectome analysis. This motivates deep learning models that can represent high-dimensional connectomes in a low-dimensional space while explicitly separating acquisition-related effects from biological variation. Conventional dimensionality reduction methods model all variance as continuous, so acquisition effects often get absorbed into a continuous latent space. Recent hybrid latent-space models combine discrete and continuous components to address this, but typically require manual capacity tuning to ensure the discrete component captures the intended variability. We introduce an unsupervised framework that removes this manual tuning by architecturally annealing encoder outputs before decoding, allowing the model to adaptively balance discrete and continuous latent variables during training. To evaluate it, we curated a dataset of N=7,416 structural connectomes derived from dMRI, spanning ages 2 to 102 and 13 studies with 25 unique acquisition-parameter combinations. Of these, 5,900 are cognitively unimpaired, 877 have mild cognitive impairment (MCI), and 639 have Alzheimer's disease (AD). We compare against a standard VAE, PCA with k-means clustering, and hybrid models that anneal only through the loss function. Our architectural annealing produces stronger site learning (ARI=0.53, p<0.05) than these baselines. Results show that a hybrid continuous-discrete latent space, with architectural rather than loss-based annealing, provides a useful unsupervised mechanism for capturing acquisition variability in dMRI: by jointly modeling smooth and categorical structure, the Joint-VAE recovers clusters aligned with scanner and protocol differences.
format Preprint
id arxiv_https___arxiv_org_abs_2605_13933
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Unsupervised learning of acquisition variability in structural connectomes via hybrid latent space modeling
Rudravaram, Gaurav
Zuo, Lianrui
Ramadass, Karthik
McMaster, Elyssa
Yoon, Jongyeon
Krishnan, Aravind R.
Saunders, Adam M.
Gao, Chenyu
Newlin, Nancy R.
Kanakaraj, Praitayini
Held, Lori L. Beason
Bilgel, Murat
Barquero, Laura A.
DArchangel, Micah
Nguyen, Tin Q.
Cutting, Laurie B.
Archer, Derek
Hohman, Timothy J.
Moyer, Daniel C.
Landman, Bennett A.
Machine Learning
Artificial Intelligence
Acquisition differences across sites, scanners, and protocols in dMRI introduce variability that complicates structural connectome analysis. This motivates deep learning models that can represent high-dimensional connectomes in a low-dimensional space while explicitly separating acquisition-related effects from biological variation. Conventional dimensionality reduction methods model all variance as continuous, so acquisition effects often get absorbed into a continuous latent space. Recent hybrid latent-space models combine discrete and continuous components to address this, but typically require manual capacity tuning to ensure the discrete component captures the intended variability. We introduce an unsupervised framework that removes this manual tuning by architecturally annealing encoder outputs before decoding, allowing the model to adaptively balance discrete and continuous latent variables during training. To evaluate it, we curated a dataset of N=7,416 structural connectomes derived from dMRI, spanning ages 2 to 102 and 13 studies with 25 unique acquisition-parameter combinations. Of these, 5,900 are cognitively unimpaired, 877 have mild cognitive impairment (MCI), and 639 have Alzheimer's disease (AD). We compare against a standard VAE, PCA with k-means clustering, and hybrid models that anneal only through the loss function. Our architectural annealing produces stronger site learning (ARI=0.53, p<0.05) than these baselines. Results show that a hybrid continuous-discrete latent space, with architectural rather than loss-based annealing, provides a useful unsupervised mechanism for capturing acquisition variability in dMRI: by jointly modeling smooth and categorical structure, the Joint-VAE recovers clusters aligned with scanner and protocol differences.
title Unsupervised learning of acquisition variability in structural connectomes via hybrid latent space modeling
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.13933