Now showing 1 - 2 of 2
  • Publication
    Low Power Real-Time Seizure Detection for Ambulatory EEG
    Ambulatory Electroencephalograph (AEEG) technology is becoming popular because it facilitates the continuous monitoring of epilepsy patients without interrupting their routine life. As long term monitoring requires low power processing on the device, a low power real time seizure detection algorithm suitable for AEEG devices is proposed herein. The performance of various classifiers was tested and the most effective was found to be the Linear Discriminant Analysis classifier (LDA). The algorithm presented in this paper provides 87.7 (100–70.2)% accuracy with 94.2 (100–78)% sensitivity and 77.9 (100–52.1)% specificity in patient dependent experiments. It provides 76.5 (79.0–73.3)% accuracy with 90.9 (96.2–85.8)% sensitivity and 59.5 (70.9–52.6)% specificity in patient independent experiments. We also suggest how power can be saved at the lost of a small amount of accuracy by applying different techniques. The algorithm was simulated on a DSP processor and on an ASIC and the power estimation results for both implementations are presented. Seizure detection using the presented algorithm is approximately 100% more power efficient than other AEEG processing methods. The implementation using an ASIC can reduce power consumption by 25% relative to the implementation on a DSP processor with reduction of only 1% of accuracy.
  • Publication
    Improved patient specific seizure detection during pre-surgical evaluation
    Objective: There is considerable interest in improved off-line automated seizure detection methods that will decrease the workload of EEG monitoring units. Subject-specific approaches have been demonstrated to perform better than subject-independent ones. However, for pre-surgical diagnostics, the traditional method of obtaining a priori data to train subject-specific classifiers is not practical. We present an alternative method that works by adapting the threshold of a subject-independent to a specific subject based on feedback from the user. Methods: A subject-independent quadratic discriminant classifier incorporating modified features based partially on the Gotman algorithm was first built. It was then used to derive subject-specific classifiers by determining subject-specific posterior probability thresholds via user interaction. The two schemes were tested on 529 h of intracranial EEG containing 63 seizures from 15 subjects undergoing pre-surgical evaluation. To provide comparison, the standard Gotman algorithm was implemented and optimised for this dataset by tuning the detection thresholds. Results: Compared to the tuned Gotman algorithm, the subject-independent scheme reduced the false positive rate by 51% (0.23 to 0.11 h−1) while increasing sensitivity from 53% to 62%. The subject-specific scheme further improved sensitivity to 78%, but with a small increase in false positive rate to 0.18 h−1. Conclusions: The results suggest that a subject-independent classifier scheme with modified features is useful for reducing false positive rate, while subject adaptation further enhances performance by improving sensitivity. The results also suggest that the proposed subject-adapted classifier scheme approximates the performance of the subject-specific Gotman algorithm. Significance: The proposed method could potentially increase the productivity of offline EEG analysis. The approach could also be generalised to enhance the performance of other subject independent algorithms.
      513Scopus© Citations 32