Leednet: a lightweight network for event detection in EEG signals
Mohammad Ali Alqarni, Hira Masood, Hassan Aqeel Khan, Awais Mehmood Kamboh, Saima Shafait, Tooba Jaffarey, Faisal Shafait · 2026
Read Leednet: a lightweight network for event detection in EEG signals by Mohammad Ali Alqarni, Hira Masood, Hassan Aqeel Khan, Awais Mehmood Kamboh, Saima Shafait, Tooba Jaffarey, Faisal Shafait (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13471070.txt.
Description
Mohammad Ali Alqarni and colleagues built LEEDNet in 2026, a lightweight network for detecting diagnostic events in electroencephalogram recordings. Long recordings demand thousands of short-window decisions, straining field triage systems in resource-limited settings. The purpose-built convolutional network classifies two-second windows as normal, slow waves, or spike and sharp waves.
On the expert-labelled NMT-Events set with splits by subject, raw-signal LEEDNet reached 81.2% accuracy and 77.5% macro-F1, beating four state-of-the-art baselines on three of four metrics. It needs only 0.25 million parameters and 37 megaflops per segment, inferring in under five milliseconds on a standard CPU. Fourier and wavelet features added no consistent gains, favouring raw waveforms for deployment.
Catalog record
| CREATOR | Mohammad Ali Alqarni, Hira Masood, Hassan Aqeel Khan, Awais Mehmood Kamboh, Saima Shafait, Tooba Jaffarey, Faisal Shafait |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |