Exploring the feasibility of olfactory brain–computer interfaces

Abstract In this study, we explore the feasibility of single-trial predictions of odor registration in the brain using olfactory bio-signals. We focus on two main aspects: input data modality and the processing model. For the first time, we assess the predictability of odor registration from novel e...

Full description

Saved in:
Bibliographic Details
Main Authors: Nona Rajabi, Irene Zanettin, Antônio H. Ribeiro, Miguel Vasco, Mårten Björkman, Johan N. Lundström, Danica Kragic
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-01488-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849705310212587520
author Nona Rajabi
Irene Zanettin
Antônio H. Ribeiro
Miguel Vasco
Mårten Björkman
Johan N. Lundström
Danica Kragic
author_facet Nona Rajabi
Irene Zanettin
Antônio H. Ribeiro
Miguel Vasco
Mårten Björkman
Johan N. Lundström
Danica Kragic
author_sort Nona Rajabi
collection DOAJ
description Abstract In this study, we explore the feasibility of single-trial predictions of odor registration in the brain using olfactory bio-signals. We focus on two main aspects: input data modality and the processing model. For the first time, we assess the predictability of odor registration from novel electrobulbogram (EBG) recordings, both in sensor and source space, and compare these with commonly used electroencephalogram (EEG) signals. Despite having fewer data channels, EBG shows comparable performance to EEG. We also examine whether breathing patterns contain relevant information for this task. By comparing a logistic regression classifier, which requires hand-crafted features, with an end-to-end convolutional deep neural network, we find that end-to-end approaches can be as effective as classic methods. However, due to the high dimensionality of the data, the current dataset is insufficient for either classifier to robustly differentiate odor and non-odor trials. Finally, we identify key challenges in olfactory BCIs and suggest future directions for improving odor detection systems.
format Article
id doaj-art-3744a1992ee744efabf063266c2156ff
institution DOAJ
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-3744a1992ee744efabf063266c2156ff2025-08-20T03:16:31ZengNature PortfolioScientific Reports2045-23222025-05-0115111310.1038/s41598-025-01488-zExploring the feasibility of olfactory brain–computer interfacesNona Rajabi0Irene Zanettin1Antônio H. Ribeiro2Miguel Vasco3Mårten Björkman4Johan N. Lundström5Danica Kragic6Department of Intelligent Systems, KTH Royal Institute of TechnologyDepartment of Clinical Neuroscience, Karolinska InstituteDepartment of Information Technology, Uppsala UniversityDepartment of Intelligent Systems, KTH Royal Institute of TechnologyDepartment of Intelligent Systems, KTH Royal Institute of TechnologyDepartment of Clinical Neuroscience, Karolinska InstituteDepartment of Intelligent Systems, KTH Royal Institute of TechnologyAbstract In this study, we explore the feasibility of single-trial predictions of odor registration in the brain using olfactory bio-signals. We focus on two main aspects: input data modality and the processing model. For the first time, we assess the predictability of odor registration from novel electrobulbogram (EBG) recordings, both in sensor and source space, and compare these with commonly used electroencephalogram (EEG) signals. Despite having fewer data channels, EBG shows comparable performance to EEG. We also examine whether breathing patterns contain relevant information for this task. By comparing a logistic regression classifier, which requires hand-crafted features, with an end-to-end convolutional deep neural network, we find that end-to-end approaches can be as effective as classic methods. However, due to the high dimensionality of the data, the current dataset is insufficient for either classifier to robustly differentiate odor and non-odor trials. Finally, we identify key challenges in olfactory BCIs and suggest future directions for improving odor detection systems.https://doi.org/10.1038/s41598-025-01488-z
spellingShingle Nona Rajabi
Irene Zanettin
Antônio H. Ribeiro
Miguel Vasco
Mårten Björkman
Johan N. Lundström
Danica Kragic
Exploring the feasibility of olfactory brain–computer interfaces
Scientific Reports
title Exploring the feasibility of olfactory brain–computer interfaces
title_full Exploring the feasibility of olfactory brain–computer interfaces
title_fullStr Exploring the feasibility of olfactory brain–computer interfaces
title_full_unstemmed Exploring the feasibility of olfactory brain–computer interfaces
title_short Exploring the feasibility of olfactory brain–computer interfaces
title_sort exploring the feasibility of olfactory brain computer interfaces
url https://doi.org/10.1038/s41598-025-01488-z
work_keys_str_mv AT nonarajabi exploringthefeasibilityofolfactorybraincomputerinterfaces
AT irenezanettin exploringthefeasibilityofolfactorybraincomputerinterfaces
AT antoniohribeiro exploringthefeasibilityofolfactorybraincomputerinterfaces
AT miguelvasco exploringthefeasibilityofolfactorybraincomputerinterfaces
AT martenbjorkman exploringthefeasibilityofolfactorybraincomputerinterfaces
AT johannlundstrom exploringthefeasibilityofolfactorybraincomputerinterfaces
AT danicakragic exploringthefeasibilityofolfactorybraincomputerinterfaces