Key result
Multimodal automatic detection systems for tonic seizures achieved sensitivities up to 1.0 and accuracies up to 0.97, significantly outperforming single-modality approaches.
Why the study?
This study aimed to evaluate the effectiveness of automatic detection systems for tonic seizures, focusing on different noninvasive modalities, algorithms, and performance metrics.
Do multimodal automatic detection systems improve the detection accuracy of tonic seizures compared to single-modality approaches?
Systematic Review (n=19)
Do multimodal automatic detection systems improve the detection accuracy of tonic seizures compared to single-modality approaches?
Multimodal automatic detection systems integrating physiologic and neural signals with machine learning offer high sensitivity and accuracy for detecting tonic seizures, outperforming single-modality approaches.
May advance noninvasive tonic seizure monitoring in epilepsy care; leaves open optimal modality and algorithm selection pending prospective validation.
PURPOSE: This systematic review and meta-analysis aimed to evaluate the effectiveness of automatic detection systems for tonic seizures, focusing on different noninvasive modalities, algorithms, and performance metrics. METHODS: Following Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines, a comprehensive search was conducted across PubMed, Scopus, and Wiley databases for studies published between 2014 and 2024. Inclusion criteria targeted studies assessing automatic detection systems for tonic seizures using various modalities. Performance metrics such as sensitivity, specificity, accuracy, and false alarm rates per hour (False-Positive Alarm Rate per hour) were analyzed and recalculated where necessary to ensure comparability. RESULTS: A total of 19 studies met the inclusion criteria. Multimodal systems integrating signals from accelerometry, gyroscopes, and other sensors demonstrated the highest sensitivities (up to 1.0) and accuracies (up to 0.97), significantly outperforming single-modality approaches. False alarm rates were lowest in controlled settings, particularly for ECG-based systems, but real-world applications highlighted variability and challenges with noise and signal acquisition. The findings underscore the potential of combining physiologic and neural signals with machine learning techniques to improve detection accuracy. CONCLUSIONS: Although recent advances in neurotechnology have enabled substantial progress in tonic seizure detection, significant challenges remain, including variability in performance metrics, generalizability to diverse populations, and scalability for real-world applications. Future research should focus on standardizing evaluation frameworks, diversifying training data sets, and validating systems in clinical and outpatient settings.
No takes yet. Share an insight, caveat, or question.
Guerrero‐Aranda et al. (2025) conducted a systematic review in Tonic seizures (n=19). Multimodal automatic detection systems vs. Single-modality approaches was evaluated on Performance metrics including sensitivity, specificity, accuracy, and false alarm rates per hour. Multimodal automatic detection systems for tonic seizures achieved sensitivities up to 1.0 and accuracies up to 0.97, significantly outperforming single-modality approaches.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: