# MT-NAM: An Efficient and Adaptive Model for Epileptic Seizure Detection

arXiv: https://arxiv.org/abs/2503.08251 · HTML: https://mlsanity.com/paper/2503.08251/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 53 references that are in the map.
- First posted on arXiv 11 Mar 2025.

## Builds on (most cited first, 21 of 21)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [Neural Additive Models: Interpretable Machine Learning with Neural Nets](https://mlsanity.com/paper/2004.13912/) (2020)
- [Parameter-free Online Test-time Adaptation](https://mlsanity.com/paper/2201.05718/) (2022)
- [NOTE: Robust Continual Test-time Adaptation Against Temporal Correlation](https://mlsanity.com/paper/2208.05117/) (2022)
- [Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis](https://mlsanity.com/paper/2104.08336/) (2021)
- [Self-Supervised Learning for Anomalous Channel Detection in EEG Graphs: Application to Seizure Analysis](https://mlsanity.com/paper/2208.07448/) (2022)
- [Teacher-Student Architecture for Knowledge Learning: A Survey](https://mlsanity.com/paper/2210.17332/) (2022)
- [Learning Robust Features using Deep Learning for Automatic Seizure Detection](https://mlsanity.com/paper/1608.00220/) (2016)
- [SzCORE: A Seizure Community Open-source Research Evaluation framework for the validation of EEG-based automated seizure detection algorithms](https://mlsanity.com/paper/2402.13005/) (2024)
- [ResOT: Resource-Efficient Oblique Trees for Neural Signal Classification](https://mlsanity.com/paper/2006.07900/) (2020)
- [REST: Efficient and Accelerated EEG Seizure Analysis through Residual State Updates](https://mlsanity.com/paper/2406.16906/) (2024)
- [NeuralTree: A 256-Channel 0.227-μJ/Class Versatile Neural Activity Classification and Closed-Loop Neuromodulation SoC](https://mlsanity.com/paper/2205.06090/) (2022)
- [Closed-Loop Neural Prostheses with On-Chip Intelligence: A Review and A Low-Latency Machine Learning Model for Brain State Detection](https://mlsanity.com/paper/2109.05848/) (2021)
- [Intelligent and Miniaturized Neural Interfaces: An Emerging Era in Neurotechnology](https://mlsanity.com/paper/2405.10780/) (2024)
- [SOUL: An Energy-Efficient Unsupervised Online Learning Seizure Detection Classifier](https://mlsanity.com/paper/2110.02169/) (2021)
- [Knowledge-Distilled Graph Neural Networks for Personalized Epileptic Seizure Detection](https://mlsanity.com/paper/2304.06038/) (2023)

## Abstract (arXiv)

Enhancing the accuracy and efficiency of machine learning algorithms employed in neural interface systems is crucial for advancing next-generation intelligent therapeutic devices. However, current systems often utilize basic machine learning models that do not fully exploit the natural structure of brain signals. Additionally, existing learning models used for neural signal processing often demonstrate low speed and efficiency during inference. To address these challenges, this study introduces Micro Tree-based NAM (MT-NAM), a distilled model based on the recently proposed Neural Additive Models (NAM). The MT-NAM achieves a remarkable 100× improvement in inference speed compared to standard NAM, without compromising accuracy. We evaluate our approach on the CHB-MIT scalp EEG dataset, which includes recordings from 24 patients with varying numbers of sessions and seizures. NAM achieves an 85.3% window-based sensitivity and 95% specificity. Interestingly, our proposed MT-NAM shows only a 2% reduction in sensitivity compared to the original NAM. To regain this sensitivity, we utilize a test-time template adjuster (T3A) as an update mechanism, enabling our model to achieve higher sensitivity during test time by accommodating transient shifts in neural signals. With this online update approach, MT-NAM achieves the same sensitivity as the standard NAM while achieving approximately 50× acceleration in inference speed.