Skip to content
Preprint

MtReNIRS: A Self-Supervised Foundation Model for Transferable Neurovascular Representations

Oct 2026 · 0 citations · 40 references
Engineering Biology

TL;DR

This work shows that a montage-aware masked autoencoder, pretrained on 12 public fNIRS datasets spanning 268 subjects and 567 recording hours, learns generalizable neurovascular representations that transfer across three external datasets, providing a scalable foundation for label-efficient brain-computer interfaces.

Abstract

Functional near-infrared spectroscopy (fNIRS) offers a promising approach for monitoring the hemodynamic correlates of neural activity in naturalistic settings. However, existing modeling methods for fNIRS are largely confined to small-scale, single-session data with low sampling rates, driven by high heterogeneity in acquisition devices and optode montages. In addition, the complex spatiotemporal structure of fNIRS signals further restricts model scalability and transferability across settings. As a result, it remains unclear whether a single fNIRS encoder can learn representations that are usable across subjects, tasks, devices, and montages. Here, we show that a montage-aware masked autoencoder, pretrained on 12 public fNIRS datasets spanning 268 subjects and 567 recording hours, learns generalizable neurovascular representations that transfer across three external datasets. Our fNIRS foundation model, MtReNIRS, jointly encodes the slow hemodynamic nature of fNIRS, chromophore identity and cortical location by mapping heterogeneous montages onto a shared anatomical reference. MtReNIRS achieves cross-subject performance on par or better than fully supervised baselines and excels in few-shot within-subject calibration, where it outperforms prior state-of-the-art fNIRS models, while remaining robust to sparse montages. Our results indicate that montage-aware self-supervision yields physiologically grounded, transferable fNIRS representations, providing a scalable foundation for label-efficient brain-computer interfaces.

View source

Similar papers

Open access Sep 2026

Cross-Modal Learning from fMRI to DOT: Leveraging Large-Scale Neuroimaging for fNIRS Decoding.

Functional near-infrared spectroscopy (fNIRS) and its tomographic extension, diffuse optical tomography (DOT), offer portable and non-invasive measures of functional brain hemodynamics with great potential for everyday neuroimaging. However, the present scarcity of publicly available DOT data hinders deep learning mode...

Shakiba Moradi, Theekshana Dissanayake, Nils Harmening et al. · 2 citations
Preprint Oct 2026

CoFlu-Brain: An fMRI Connectivity Foundation Model with Cofluctuation-Based Pretraining

Functional magnetic resonance imaging (fMRI) commonly characterizes large-scale brain organization through functional connectivity (FC). However, an observed FC matrix is not a fixed brain graph but a statistical estimate determined by the finite BOLD frames used to compute it. Conventional FC-based approaches typicall...

Zi-Teng Sui, Jun-Feng Xia, Wen-Hao Ye et al. · 0 citations
#machine learning Preprint Oct 2026

GeoNIRS: Geometry-Aware Pretraining for Heterogeneous fNIRS Representation Learning

Functional near-infrared spectroscopy (fNIRS) provides portable and spatially localized measurements of cortical hemodynamics, but differences in channel count, probe layout, signal distribution, and task type hinder joint pretraining across datasets. We introduce GeoNIRS, a geometry-aware framework that uses source-de...

Zong-Sheng Li, Sheng-Yuan Zhu, Zi-Mo Zhang et al. · 1 citation · ⚡1
#human-computer interacti... Preprint Oct 2026

Towards fNIRS foundation models: A curated corpus of 67 public datasets with channel positions in MNI space

Brain foundation models have transformed medical artificial intelligence and computational neuroscience by reducing the reliance on task-specific data through large-scale pretraining. While they have been successful with non-invasive neuroimaging data, their adoption in functional near-infrared spectroscopy (fNIRS) has...

Santiago Posso-Murillo · 0 citations

Flattening the Connectome Spectrum: A Spectral Filter for FC Induces a Pretraining Target for fMRI Encoders

It is shown that KRR is weighted by the eigenvalues of the FC which are miscalibrated for phenotype prediction, and an efficient spectral filter is applied to recalibrate the eigenvalues of each subject's FC matrix, enabling the model to exploit more inter-individual variance.

G. Marraffini, Victoria Shevchenko, Carlo Alberto Barbano et al. · 0 citations
Book Open access Nov 2026

UniNeuro: A Unified EEG Foundation Model for Heterogeneous EEG Task Learning

Electroencephalography (EEG), as a non-invasive physiological signal, plays an increasingly important role in human-centered multimedia systems, supporting applications ranging from emotion recognition and cognitive load assessment to motor-imagery-based control. However, due to the substantial differences in temporal...

Hai-Yang Lu, Liang-Hua He, Hong-Zhou Chen et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.