### Abstract

The ratio of spectral power in two different bands and relative band power have been shown to be sometimes more discriminative features than the spectral power in a specific band for binary classification of a time series for seizure prediction. However, why and which ratio of spectral power and relative power features are better discriminators than a band power have not been understood. While general answers to why and which are difficult, this paper partially addresses the answer to these questions. Using auto-regressive modeling, this paper, for the first time, theoretically explains that for high signal-to-noise ratio (SNR) cases, the ratio features may sometime amplify the discriminability of one of the two states in a time series, as compared with a band power. This paper, also for the first time, introduces a novel frequency-domain model ratio (FDMR) that can be used to select the two frequency bands. The FDMR computes the ratio of the frequency responses of the two auto-regressive model filters that correspond to two different states. It is shown that the ratio implicitly cancels the effect of change of variance of the white noise that is input to the auto-regressive model in a non-stationary environment for high SNR conditions. It is also shown that under certain sufficient but not necessary conditions, the ratio of the spectral power and the relative band power, i.e., the band power divided by the total power spectral density, can be better discriminators than band power. Synthesized data and scalp EEG data from the MIT Physionet for patient-specific seizure prediction are used to explain why the ratios of spectral power obtained by a ranking algorithm in the prior literature satisfy the sufficient conditions for amplification of the ratio feature derived in this paper.

Original language | English (US) |
---|---|

Article number | 8716296 |

Pages (from-to) | 645-657 |

Number of pages | 13 |

Journal | IEEE transactions on biomedical circuits and systems |

Volume | 13 |

Issue number | 4 |

DOIs | |

State | Published - Aug 1 2019 |

### Fingerprint

### Keywords

- Time-series
- auto-regressive model
- classification
- discriminability
- frequency-domain model ratio (FDMR)
- prediction error filter
- ratio of band power
- relative band power
- seizure prediction

### PubMed: MeSH publication types

- Journal Article

### Cite this

*IEEE transactions on biomedical circuits and systems*,

*13*(4), 645-657. [8716296]. https://doi.org/10.1109/TBCAS.2019.2917184

**Discriminative Ratio of Spectral Power and Relative Power Features Derived via Frequency-Domain Model Ratio with Application to Seizure Prediction.** / Parhi, Keshab K; Zhang, Zisheng.

Research output: Contribution to journal › Article

*IEEE transactions on biomedical circuits and systems*, vol. 13, no. 4, 8716296, pp. 645-657. https://doi.org/10.1109/TBCAS.2019.2917184

}

TY - JOUR

T1 - Discriminative Ratio of Spectral Power and Relative Power Features Derived via Frequency-Domain Model Ratio with Application to Seizure Prediction

AU - Parhi, Keshab K

AU - Zhang, Zisheng

PY - 2019/8/1

Y1 - 2019/8/1

N2 - The ratio of spectral power in two different bands and relative band power have been shown to be sometimes more discriminative features than the spectral power in a specific band for binary classification of a time series for seizure prediction. However, why and which ratio of spectral power and relative power features are better discriminators than a band power have not been understood. While general answers to why and which are difficult, this paper partially addresses the answer to these questions. Using auto-regressive modeling, this paper, for the first time, theoretically explains that for high signal-to-noise ratio (SNR) cases, the ratio features may sometime amplify the discriminability of one of the two states in a time series, as compared with a band power. This paper, also for the first time, introduces a novel frequency-domain model ratio (FDMR) that can be used to select the two frequency bands. The FDMR computes the ratio of the frequency responses of the two auto-regressive model filters that correspond to two different states. It is shown that the ratio implicitly cancels the effect of change of variance of the white noise that is input to the auto-regressive model in a non-stationary environment for high SNR conditions. It is also shown that under certain sufficient but not necessary conditions, the ratio of the spectral power and the relative band power, i.e., the band power divided by the total power spectral density, can be better discriminators than band power. Synthesized data and scalp EEG data from the MIT Physionet for patient-specific seizure prediction are used to explain why the ratios of spectral power obtained by a ranking algorithm in the prior literature satisfy the sufficient conditions for amplification of the ratio feature derived in this paper.

AB - The ratio of spectral power in two different bands and relative band power have been shown to be sometimes more discriminative features than the spectral power in a specific band for binary classification of a time series for seizure prediction. However, why and which ratio of spectral power and relative power features are better discriminators than a band power have not been understood. While general answers to why and which are difficult, this paper partially addresses the answer to these questions. Using auto-regressive modeling, this paper, for the first time, theoretically explains that for high signal-to-noise ratio (SNR) cases, the ratio features may sometime amplify the discriminability of one of the two states in a time series, as compared with a band power. This paper, also for the first time, introduces a novel frequency-domain model ratio (FDMR) that can be used to select the two frequency bands. The FDMR computes the ratio of the frequency responses of the two auto-regressive model filters that correspond to two different states. It is shown that the ratio implicitly cancels the effect of change of variance of the white noise that is input to the auto-regressive model in a non-stationary environment for high SNR conditions. It is also shown that under certain sufficient but not necessary conditions, the ratio of the spectral power and the relative band power, i.e., the band power divided by the total power spectral density, can be better discriminators than band power. Synthesized data and scalp EEG data from the MIT Physionet for patient-specific seizure prediction are used to explain why the ratios of spectral power obtained by a ranking algorithm in the prior literature satisfy the sufficient conditions for amplification of the ratio feature derived in this paper.

KW - Time-series

KW - auto-regressive model

KW - classification

KW - discriminability

KW - frequency-domain model ratio (FDMR)

KW - prediction error filter

KW - ratio of band power

KW - relative band power

KW - seizure prediction

UR - http://www.scopus.com/inward/record.url?scp=85070945381&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070945381&partnerID=8YFLogxK

U2 - 10.1109/TBCAS.2019.2917184

DO - 10.1109/TBCAS.2019.2917184

M3 - Article

C2 - 31095498

AN - SCOPUS:85070945381

VL - 13

SP - 645

EP - 657

JO - IEEE Transactions on Biomedical Circuits and Systems

JF - IEEE Transactions on Biomedical Circuits and Systems

SN - 1932-4545

IS - 4

M1 - 8716296

ER -