Research map / Label-efficient and robust learning / Recognition and long-tail
Omni-iEEG: A Large-Scale, Comprehensive iEEG Dataset and Benchmark for Epilepsy Research
Chenda Duan, Yipeng Zhang, Sotaro Kanai, Yuanyi Ding, Atsuro Daida, Pengyue Yu, Tiancheng Zheng, Naoto Kuroda, Shaun A. Hussain, Eishi Asano, Hiroki Nariai, vwani Roychowdhury
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 |
|---|---|---|---|
| use time-frequency (Morlet) spectrogram input with ResNet18 instead of raw time-domain traces | HFO event classifier · HFO event classification | macro F1: 0.7726 -> 0.8061 (PatchTST vs PyHFO-Omni) | Improved |
| train PyHFO classifier on Omni-iEEG multi-center annotations instead of single-center model | PyHFO · Pathological channel identification | outcome AUC: 0.4972 -> 0.7438 | Improved |
| train on Omni-iEEG multi-center annotations instead of single-center model | PyHFO · Pathological channel identification | channel AUC: 0.6557 -> 0.7351 | Improved |
| use end-to-end TimeConv-CNN on 1-minute segments instead of event-based PyHFO-Omni | TimeConv-CNN · Pathological channel identification | channel AUC: 0.7351 -> 0.8061 | Improved |
| use end-to-end TimeConv-CNN instead of event-based PyHFO-Omni | TimeConv-CNN · Surgical outcome prediction | outcome AUC: 0.7438 vs 0.7380 | No clear effect |
| fine-tune audio-pretrained CLAP instead of SEEG-NET | CLAP · Pathological channel identification | specificity: 0.6049 -> 0.7823 | Improved |
| fine-tune audio-pretrained CLAP instead of SEEG-NET | CLAP · Pathological channel identification | channel AUC: 0.7850 -> 0.7684 | Mixed |
| use CLAP instead of TimeConv-CNN on exploratory tasks | CLAP · Ictal vs interictal | macro F1: 0.8533 -> 0.9245 | Improved |
| use CLAP instead of TimeConv-CNN | CLAP · Anatomical location (12 classes) | macro F1: 0.3087 -> 0.3540 | Improved |
Closest papers by meaning
- DMNet: Self-comparison Driven Model for Subject-independent Seizure Detection (NeurIPS 2024 · similarity 0.86)
- NeuroAtlas: Benchmarking Foundation Models for Clinical EEG and Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.86)
- Seizure-Semiology-Suite($S^3$): A Clinically Multimodal Dataset, Benchmark, and Models for Seizure Semiology Understanding (ICML 2026 · similarity 0.86)
- FAPEX: Fractional Amplitude-Phase Expressor for Robust Cross-Subject Seizure Prediction (NeurIPS 2025 · similarity 0.85)
- Long-Term EEG Partitioning for Seizure Onset Detection (AAAI 2025 · similarity 0.84)
- EvoBrain: Dynamic Multi-Channel EEG Graph Modeling for Time-Evolving Brain Networks (NeurIPS 2025 · similarity 0.84)
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.
