人工智能
计算机科学
卷积神经网络
模式识别(心理学)
计算机视觉
人工神经网络
特征(语言学)
眼动
噪音(视频)
可解释性
作者
Wen-yi Tsai,Cheng-Hung Lin
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2026-03-28
卷期号:: 61-64
标识
DOI:10.1109/icct-pacific69083.2026.11518775
摘要
Convolutional neural network (CNN) models utilizing stochastic computing for the classification of 3D optical coherence tomography (OCT) eye images. By representing numbers within [−1, 1] using bit-streams, and performing operations with XNOR gates and multiplexers, the model achieves significant reductions in power consumption and hardware footprint. The proposed model attained an accuracy of 93.02% in classifying OCT eye images, demonstrating its effectiveness in detecting various eye diseases. This approach provides a scalable and energy-efficient solution for deploying eye disease diagnostic systems in rural and remote areas, enhancing healthcare accessibility.
科研通智能强力驱动
Strongly Powered by AbleSci AI