A Machine Learning Perspective on fNIRS Signal Quality Control Approaches
Despite a rise in the use of functional Near Infra-Red Spectroscopy (fNIRS) to study neural systems, fNIRS signal processing is not standardized and is highly affected by empirical and manual procedures. At the beginning of any signal processing procedure, Signal Quality Control (SQC) is critical to...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9854812/ |
| _version_ | 1797805132647235584 |
|---|---|
| author | Andrea Bizzego Michelle Neoh Giulio Gabrieli Gianluca Esposito |
| author_facet | Andrea Bizzego Michelle Neoh Giulio Gabrieli Gianluca Esposito |
| author_sort | Andrea Bizzego |
| collection | DOAJ |
| description | Despite a rise in the use of functional Near Infra-Red Spectroscopy (fNIRS) to study neural systems, fNIRS signal processing is not standardized and is highly affected by empirical and manual procedures. At the beginning of any signal processing procedure, Signal Quality Control (SQC) is critical to prevent errors and unreliable results. In fNIRS analysis, SQC currently relies on applying empirical thresholds to handcrafted Signal Quality Indicators (SQIs). In this study, we use a dataset of fNIRS signals (N = 1,340) recorded from 67 subjects, and manually label the signal quality of a subset of segments (N = 548) to investigate the pitfalls of current practices while exploring the opportunities provided by Deep Learning approaches. We show that SQIs statistically discriminate signals with bad quality, but the identification by means of empirical thresholds lacks sensitivity. Alternatively to manual thresholding, conventional machine learning models based on the SQIs have been proven more accurate, with end-to-end approaches, based on Convolutional Neural Networks, capable of further improving the performance. The proposed approach, based on machine learning, represents a more objective SQC for fNIRS and moves towards the use of fully automated and standardized procedures. |
| first_indexed | 2024-03-13T05:46:41Z |
| format | Article |
| id | doaj.art-0e3231c11a49437d96920bc5b983cc74 |
| institution | Directory Open Access Journal |
| issn | 1558-0210 |
| language | English |
| last_indexed | 2024-03-13T05:46:41Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Transactions on Neural Systems and Rehabilitation Engineering |
| spelling | doaj.art-0e3231c11a49437d96920bc5b983cc742023-06-13T20:08:03ZengIEEEIEEE Transactions on Neural Systems and Rehabilitation Engineering1558-02102022-01-01302292230010.1109/TNSRE.2022.31981109854812A Machine Learning Perspective on fNIRS Signal Quality Control ApproachesAndrea Bizzego0https://orcid.org/0000-0002-1586-8350Michelle Neoh1https://orcid.org/0000-0001-5960-1634Giulio Gabrieli2https://orcid.org/0000-0002-9846-5767Gianluca Esposito3https://orcid.org/0000-0002-9442-0254Department of Psychology and Cognitive Science, University of Trento, Trento, ItalyPsychology Program, Nanyang Technological University, Jurong West, SingaporePsychology Program, Nanyang Technological University, Jurong West, SingaporeDepartment of Psychology and Cognitive Science, University of Trento, Trento, ItalyDespite a rise in the use of functional Near Infra-Red Spectroscopy (fNIRS) to study neural systems, fNIRS signal processing is not standardized and is highly affected by empirical and manual procedures. At the beginning of any signal processing procedure, Signal Quality Control (SQC) is critical to prevent errors and unreliable results. In fNIRS analysis, SQC currently relies on applying empirical thresholds to handcrafted Signal Quality Indicators (SQIs). In this study, we use a dataset of fNIRS signals (N = 1,340) recorded from 67 subjects, and manually label the signal quality of a subset of segments (N = 548) to investigate the pitfalls of current practices while exploring the opportunities provided by Deep Learning approaches. We show that SQIs statistically discriminate signals with bad quality, but the identification by means of empirical thresholds lacks sensitivity. Alternatively to manual thresholding, conventional machine learning models based on the SQIs have been proven more accurate, with end-to-end approaches, based on Convolutional Neural Networks, capable of further improving the performance. The proposed approach, based on machine learning, represents a more objective SQC for fNIRS and moves towards the use of fully automated and standardized procedures.https://ieeexplore.ieee.org/document/9854812/Deep learningfunctional near infrared spectroscopymachine learningsignal quality control |
| spellingShingle | Andrea Bizzego Michelle Neoh Giulio Gabrieli Gianluca Esposito A Machine Learning Perspective on fNIRS Signal Quality Control Approaches IEEE Transactions on Neural Systems and Rehabilitation Engineering Deep learning functional near infrared spectroscopy machine learning signal quality control |
| title | A Machine Learning Perspective on fNIRS Signal Quality Control Approaches |
| title_full | A Machine Learning Perspective on fNIRS Signal Quality Control Approaches |
| title_fullStr | A Machine Learning Perspective on fNIRS Signal Quality Control Approaches |
| title_full_unstemmed | A Machine Learning Perspective on fNIRS Signal Quality Control Approaches |
| title_short | A Machine Learning Perspective on fNIRS Signal Quality Control Approaches |
| title_sort | machine learning perspective on fnirs signal quality control approaches |
| topic | Deep learning functional near infrared spectroscopy machine learning signal quality control |
| url | https://ieeexplore.ieee.org/document/9854812/ |
| work_keys_str_mv | AT andreabizzego amachinelearningperspectiveonfnirssignalqualitycontrolapproaches AT michelleneoh amachinelearningperspectiveonfnirssignalqualitycontrolapproaches AT giuliogabrieli amachinelearningperspectiveonfnirssignalqualitycontrolapproaches AT gianlucaesposito amachinelearningperspectiveonfnirssignalqualitycontrolapproaches AT andreabizzego machinelearningperspectiveonfnirssignalqualitycontrolapproaches AT michelleneoh machinelearningperspectiveonfnirssignalqualitycontrolapproaches AT giuliogabrieli machinelearningperspectiveonfnirssignalqualitycontrolapproaches AT gianlucaesposito machinelearningperspectiveonfnirssignalqualitycontrolapproaches |