Human brain distinctiveness based on EEG spectral coherence connectivity (1403.6384v1)

Abstract: The use of EEG biometrics, for the purpose of automatic people recognition, has received increasing attention in the recent years. Most of current analysis rely on the extraction of features characterizing the activity of single brain regions, like power-spectrum estimates, thus neglecting possible temporal dependencies between the generated EEG signals. However, important physiological information can be extracted from the way different brain regions are functionally coupled. In this study, we propose a novel approach that fuses spectral coherencebased connectivity between different brain regions as a possibly viable biometric feature. The proposed approach is tested on a large dataset of subjects (N=108) during eyes-closed (EC) and eyes-open (EO) resting state conditions. The obtained recognition performances show that using brain connectivity leads to higher distinctiveness with respect to power-spectrum measurements, in both the experimental conditions. Notably, a 100% recognition accuracy is obtained in EC and EO when integrating functional connectivity between regions in the frontal lobe, while a lower 97.41% is obtained in EC (96.26% in EO) when fusing power spectrum information from centro-parietal regions. Taken together, these results suggest that functional connectivity patterns represent effective features for improving EEG-based biometric systems.

• The paper proposes using EEG spectral coherence, representing brain region synchrony, as a novel and distinct biometric feature for human identification, moving beyond traditional single-region analysis.
• The study achieved 100% identification accuracy using frontal lobe spectral coherence features, demonstrating the superior discriminatory power of connectivity compared to power spectral density measures.
• Findings suggest EEG spectral coherence is a highly promising and robust biometric modality with potential for secure identification systems and future research focusing on diverse scenarios and hardware improvements.

EEG Spectral Coherence Connectivity for Biometric Identification

The paper by La Rocca et al. investigates the application of electroencephalography (EEG) spectral coherence connectivity as a distinguishing biometric feature for human identification. Traditional EEG-based biometric systems typically emphasize the analysis of single brain region activities, primarily derived from power-spectrum estimates. This approach often neglects the temporal dependencies and connectivity patterns that hold valuable physiological information. This paper introduces a novel methodology incorporating spectral coherence, which defines the level of synchrony between EEG signals from different brain regions as a biometric feature.

Key Methodological Insights

The researchers utilized a dataset from the PhysioNet BCI database, comprising 108 subjects recorded under eyes-closed (EC) and eyes-open (EO) conditions. EEG data were processed using a 56-channel montage, with the signals from each condition split into 10-second epochs. Two primary feature extraction techniques were deployed: power spectral density (PSD) and spectral coherence (COH). The PSD method provides an individual brain region's oscillatory activities, while COH estimates the interdependence and connectivity within brain networks.

To evaluate the distinctiveness of the extracted features, the authors employed a Mahalanobis distance-based classifier. They adapted the data through Fisher’s Z transformation for COH and logarithmic transformation for PSD to normalize feature distributions, thereby improving classification robustness. A cross-validation framework allowed them to partition data into enroLLMent and test phases to assess recognition performance.

Results and Performance

A distinctive outcome of the paper is the report of 100% recognition accuracy in both EC and EO states when focusing on frontal lobe regions utilizing the functional connectivity approach, indicating the paramount significance of connectivity features over traditional PSD measures. In contrast, recognition scores reduced slightly to 97.41% for EC and 96.26% for EO when solely utilizing PSD from centro-parietal regions. Moreover, the paper evidences that aggregating connectivity scores across different zones (frontal, central, parieto-occipital) considerably enhances recognition efficacy across various conditions, with topological arrangements favoring short-range connectivity.

Practical and Theoretical Implications

The paper implies potent implications for EEG-based biometric systems. Practically, the high accuracy reported underscores the potential of EEG spectral coherence as a viable biometric modality, presenting less vulnerability to forgery compared to traditional biometrics like fingerprints or irises. Future developments could move towards more compact and efficient sensor arrays focused on critical frontal zones, increasing the applicability of EEG biometrics in real-world settings.

Theoretically, the findings support the notion of brain connectivity as a more stable and unique identifier than isolated regional activities. The high recognition rates further bolster the hypothesis that synchronous connectivity patterns inherently reflect individual neurophysiological distinctiveness.

Future Directions

Further exploration is warranted to understand the full potential of connectivity-based EEG biometrics. Future investigations should focus on diverse recording scenarios, including tasks involving cognitive load variations, to test the generalization capability of such models. Advancements in EEG hardware, such as the development of non-invasive, dry electrode systems, also hold promise for more practical implementations. Additionally, there is room for improved classifiers that better capture the nuances of complex brain connectivity networks during different states.

In conclusion, La Rocca et al.'s paper offers crucial insights into the potential of EEG spectral coherence as a distinctive and robust feature for biometric applications, presenting a significant step forward in the exploitation of brain connectivity patterns for secure personal identification systems.