Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Mar 26:16:1514883.
doi: 10.3389/fphys.2025.1514883. eCollection 2025.

Design of a hybrid AI network circuit for epilepsy detection with 97.5% accuracy and low cost-latency

Liufang Sheng  1 , Xuanxu Chen  2 , Yuejun Zhang  2 , Ke Yan  1 , Junping Chen  3 , Zhikang Chen  2 , Hanyu Shi  2 , Yi Gong  2
Affiliations

Design of a hybrid AI network circuit for epilepsy detection with 97.5% accuracy and low cost-latency

Liufang Sheng et al. Front Physiol. .

Abstract

Epilepsy detection using artificial intelligence (AI) networks has gained significant attention. However, existing methods face challenges in accuracy, computational cost, and speed. CNN excel in feature extraction but suffer from high computational latency and power consumption, while SVM rely heavily on feature quality and expensive kernel computations, limiting real-time performance. Additionally, most CNN-SVM hybrid model lack hardware optimization, leading to inefficient implementations with poor accuracy-latency trade-offs. To address these issues, this paper designs a hybrid AI network-based method for epilepsy detection using electroencephalography (EEG) signals. First, a hybrid AI network was constructed using three convolutional layers, three pooling layers, and a Gaussian kernel SVM to achieve EEG epilepsy detection. Then, the design of the multiply-accumulate circuit was completed using a parallel-style row computation method, and a pipelined convolutional computation circuit was used to accelerate the convolutional computation and reduce the computational overhead and delay. Finally, a single-precision floating-point exponential and logarithmic computation circuit was designed to improve the speed and accuracy of data computation. The digital back-end of the hardware circuit was realized under the TSMC 65 nm process. Experimental results show that the circuit occupies an area of 3.20 mm2, consumes 4.28 mW of power, operates at a frequency of 10 MHz, and has an epilepsy detection latency of 0.008 s, which represents a 32% reduction in latency compared to those reported in the relevant literature. The database test results showed an epilepsy detection accuracy of 97.5%, a sensitivity of 97.6%, and a specificity of 97.2%.

Keywords: biomedical diagnostics; convolutional neural network; epilepsy detection; hardware implementation; hybrid AI network model.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

FIGURE 1
FIGURE 1
The epilepsy detection process.
FIGURE 2
FIGURE 2
Hybrid AI network for epilepsy detection.
FIGURE 3
FIGURE 3
Structure of CNN and SVM epilepsy detection model.
FIGURE 4
FIGURE 4
Data flow diagram of hybrid AI network model.
FIGURE 5
FIGURE 5
Hardware structure of the 3 × 3 convolution circuit.
FIGURE 6
FIGURE 6
Hardware structure of the pooling circuit.
FIGURE 7
FIGURE 7
SVM hardware structure.
FIGURE 8
FIGURE 8
Hardware circuit diagram of the CORDIC algorithm.
FIGURE 9
FIGURE 9
DPU hardware circuit diagram.
FIGURE 10
FIGURE 10
EEG waveform of patients with epilepsy at different stages.
FIGURE 11
FIGURE 11
BONN dataset test results.
FIGURE 12
FIGURE 12
CHB-MIT dataset test results.
FIGURE 13
FIGURE 13
Confusion matrix of hardware classification results (A) BONN classification confusion matrix (B) CHB-MIT classification confusion matrix.
FIGURE 14
FIGURE 14
Epilepsy detection circuit (A) circuit layout (B) Key performance parameters (C) Area distribution of modules (D) Power distribution of each module.
FIGURE 15
FIGURE 15
Cumulative delayed test results for epilepsy detection.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Andrzejak R. G., Lehnertz K., Mormann F., Rieke C., David P., Elger C. E. (2001). Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. Phys. Rev. E. 64, 061907. 10.1103/PhysRevE.64.061907 - DOI - PubMed
    1. Balestrini S., Arzimanoglou A., Blümcke I., Scheffer I. E., Wiebe S., Zelano J., et al. (2021). The aetiologies of epilepsy. Epileptic Disord. 23, 1–16. 10.1684/epd.2021.1255 - DOI - PubMed
    1. Cao J. W., Feng Y. M., Zheng R. Z., Cui X. N., Zhao W. J., Jiang T. J., et al. (2022). Two-stream attention 3-D deep network-based childhood epilepsy syndrome classification. IEEE T. Instrum. Meas. 72, 1–12. 10.1109/TIM.2022.3220287 - DOI
    1. Cortes C., Vapnik V. (1995). Support-vector networks. Mach. Learn. 20, 273–297. 10.1007/BF00994018 - DOI
    1. Elhosary H., Zakhari M. H., Elgammal M. A., Abd El Ghany M. A., Salama K. N., Mostafa H. (2019). Low-power hardware implementation of a support vector machine training and classification for neural seizure detection. IEEE Trans. Biomed. Circuits Syst. 13, 1324–1337. 10.1109/TBCAS.2019.2947044 - DOI - PubMed

LinkOut - more resources