Research map / Label-efficient and robust learning / Self-supervised and contrastive
PPi: Pretraining Brain Signal Model for Patient-independent Seizure Detection
Zhizhang Yuan, Daoze Zhang, Yang Yang, Junru Chen, Yafeng Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add channel discrimination task (ablation) | PPi · Clinical SEEG | F2: 24.63 -> 35.51 | Improved |
| add context swapping task (ablation) | PPi · Clinical SEEG | F2: 23.73 -> 35.51 | Improved |
| add self-supervised pretraining tasks (ablation) | PPi · MAYO | F2: 23.77 -> 61.07 | Improved |
| add self-supervised pretraining tasks (ablation) | PPi · Clinical SEEG | F2: 16.08 -> 35.51 | Improved |
| add reconstruction loss (ablation) | PPi · FNUSA | F2: 60.02 -> 70.55 | Improved |
| add spectral power (PSD) features (ablation) | PPi · MAYO | F2: 28.31 -> 61.07 | Improved |
| replace mean pooling with self-attention aggregation (ablation) | PPi · Clinical SEEG | F2: 32.59 -> 35.51 | Improved |
| add channel background subtraction (ablation) | PPi · Clinical SEEG | F2: 26.74 -> 35.51 | Improved |
| add channel background subtraction (ablation) | PPi · MAYO | F2: 30.05 -> 61.07 | Improved |
| add brain region enhancement (multi-class region labels) (ablation) | PPi · Clinical SEEG | F2: 28.33 -> 35.51 | Improved |
| increase context half length h (ablation) | PPi · Clinical SEEG | score | Improved |
| PPi vs best baseline | PPi · Clinical SEEG | F2: 54.93% | Improved |
Closest papers by meaning
- DMNet: Self-comparison Driven Model for Subject-independent Seizure Detection (NeurIPS 2024 · similarity 0.88)
- Long-Term EEG Partitioning for Seizure Onset Detection (AAAI 2025 · similarity 0.85)
- REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates (ICML 2024 · similarity 0.84)
- Seizure-Semiology-Suite($S^3$): A Clinically Multimodal Dataset, Benchmark, and Models for Seizure Semiology Understanding (ICML 2026 · similarity 0.83)
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.83)
- EvoBrain: Dynamic Multi-Channel EEG Graph Modeling for Time-Evolving Brain Networks (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
