Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks
Activated channels of functional near-infrared spectroscopy are typically identified using the desired hemodynamic response function (dHRF) generated by a trial period. However, this approach is not possible for an unknown trial period. In this paper, an innovative method not using the dHRF is propo...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/10/6/685 |
| _version_ | 1827738512392192000 |
|---|---|
| author | So-Hyeon Yoo Guanghao Huang Keum-Shik Hong |
| author_facet | So-Hyeon Yoo Guanghao Huang Keum-Shik Hong |
| author_sort | So-Hyeon Yoo |
| collection | DOAJ |
| description | Activated channels of functional near-infrared spectroscopy are typically identified using the desired hemodynamic response function (dHRF) generated by a trial period. However, this approach is not possible for an unknown trial period. In this paper, an innovative method not using the dHRF is proposed, which extracts fluctuating signals during the resting state using maximal overlap discrete wavelet transform, identifies low-frequency wavelets corresponding to physiological noise, trains them using long-short term memory networks, and predicts/subtracts them during the task session. The motivation for prediction is to maintain the phase information of physiological noise at the start time of a task, which is possible because the signal is extended from the resting state to the task session. This technique decomposes the resting state data into nine wavelets and uses the fifth to ninth wavelets for learning and prediction. In the eighth wavelet, the prediction error difference between the with and without dHRF from the 15-s prediction window appeared to be the largest. Considering the difficulty in removing physiological noise when the activation period is near the physiological noise, the proposed method can be an alternative solution when the conventional method is not applicable. In passive brain-computer interfaces, estimating the brain signal starting time is necessary. |
| first_indexed | 2024-03-11T02:45:22Z |
| format | Article |
| id | doaj.art-04f0d860f5154e6fab3669ad52a1b15c |
| institution | Directory Open Access Journal |
| issn | 2306-5354 |
| language | English |
| last_indexed | 2024-03-11T02:45:22Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Bioengineering |
| spelling | doaj.art-04f0d860f5154e6fab3669ad52a1b15c2023-11-18T09:21:21ZengMDPI AGBioengineering2306-53542023-06-0110668510.3390/bioengineering10060685Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory NetworksSo-Hyeon Yoo0Guanghao Huang1Keum-Shik Hong2School of Mechanical Engineering, Pusan National University, 2 Busandaehak-ro, Geumjeong-gu, Busan 46241, Republic of KoreaInstitute for Future, School of Automation, Qingdao University, Qingdao 266071, ChinaSchool of Mechanical Engineering, Pusan National University, 2 Busandaehak-ro, Geumjeong-gu, Busan 46241, Republic of KoreaActivated channels of functional near-infrared spectroscopy are typically identified using the desired hemodynamic response function (dHRF) generated by a trial period. However, this approach is not possible for an unknown trial period. In this paper, an innovative method not using the dHRF is proposed, which extracts fluctuating signals during the resting state using maximal overlap discrete wavelet transform, identifies low-frequency wavelets corresponding to physiological noise, trains them using long-short term memory networks, and predicts/subtracts them during the task session. The motivation for prediction is to maintain the phase information of physiological noise at the start time of a task, which is possible because the signal is extended from the resting state to the task session. This technique decomposes the resting state data into nine wavelets and uses the fifth to ninth wavelets for learning and prediction. In the eighth wavelet, the prediction error difference between the with and without dHRF from the 15-s prediction window appeared to be the largest. Considering the difficulty in removing physiological noise when the activation period is near the physiological noise, the proposed method can be an alternative solution when the conventional method is not applicable. In passive brain-computer interfaces, estimating the brain signal starting time is necessary.https://www.mdpi.com/2306-5354/10/6/685functional near-infrared spectroscopyfilteringphysiological noisemaximal overlap discrete wavelet transformlong-short term memory |
| spellingShingle | So-Hyeon Yoo Guanghao Huang Keum-Shik Hong Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks Bioengineering functional near-infrared spectroscopy filtering physiological noise maximal overlap discrete wavelet transform long-short term memory |
| title | Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks |
| title_full | Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks |
| title_fullStr | Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks |
| title_full_unstemmed | Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks |
| title_short | Physiological Noise Filtering in Functional Near-Infrared Spectroscopy Signals Using Wavelet Transform and Long-Short Term Memory Networks |
| title_sort | physiological noise filtering in functional near infrared spectroscopy signals using wavelet transform and long short term memory networks |
| topic | functional near-infrared spectroscopy filtering physiological noise maximal overlap discrete wavelet transform long-short term memory |
| url | https://www.mdpi.com/2306-5354/10/6/685 |
| work_keys_str_mv | AT sohyeonyoo physiologicalnoisefilteringinfunctionalnearinfraredspectroscopysignalsusingwavelettransformandlongshorttermmemorynetworks AT guanghaohuang physiologicalnoisefilteringinfunctionalnearinfraredspectroscopysignalsusingwavelettransformandlongshorttermmemorynetworks AT keumshikhong physiologicalnoisefilteringinfunctionalnearinfraredspectroscopysignalsusingwavelettransformandlongshorttermmemorynetworks |