论文标题
神经植入物的边缘深度学习
Edge Deep Learning for Neural Implants
论文作者
论文摘要
提供实时神经活动分类和控制的植入设备越来越多地用于治疗神经系统疾病，例如癫痫和帕金森氏病。分类性能对于确定适合治疗作用的大脑状态至关重要。但是，在离线研究（特别是深度学习方法（DL）方法）中表现出希望的高级算法尚未在资源繁殖的神经植入物上部署。在这里，我们设计并优化了三种常用结构的嵌入式DL模型，并在癫痫发作检测案例研究中评估了其推论性能。深层神经网络（DNN），卷积神经网络（CNN）和长期记忆（LSTM）网络的设计旨在对CHB-MIT头皮EEG数据库中的Ictal，perictal和untictal阶段进行分类。在迭代模型压缩和量化后，将算法部署在通用的，现成的微控制器上。推理灵敏度，假阳性率，执行时间，内存大小和功耗。对于癫痫发作的检测，对DNN，CNN和LSTM模型的灵敏度和FPR（H-1）分别为87.36％/0.169、96.70％/0.102和97.61％/0.071。预测早期警告的癫痫发作也是可行的。实施的压缩和量化可显着节省功率和内存，精度降解小于0.5％。 Edge DL模型的性能与许多没有时间或计算资源限制的先前实现相当。通用微控制器可以提供所需的内存和计算资源，而模型设计可以迁移到ASIC以进行进一步优化。结果表明，边缘DL推断是将来神经植入物改善分类性能和治疗结果的可行选择。
Implanted devices providing real-time neural activity classification and control are increasingly used to treat neurological disorders, such as epilepsy and Parkinson's disease. Classification performance is critical to identifying brain states appropriate for the therapeutic action. However, advanced algorithms that have shown promise in offline studies, in particular deep learning (DL) methods, have not been deployed on resource-restrained neural implants. Here, we designed and optimized three embedded DL models of commonly adopted architectures and evaluated their inference performance in a case study of seizure detection. A deep neural network (DNN), a convolutional neural network (CNN), and a long short-term memory (LSTM) network were designed to classify ictal, preictal, and interictal phases from the CHB-MIT scalp EEG database. After iterative model compression and quantization, the algorithms were deployed on a general-purpose, off-the-shelf microcontroller. Inference sensitivity, false positive rate, execution time, memory size, and power consumption were quantified. For seizure event detection, the sensitivity and FPR (h-1) for the DNN, CNN, and LSTM models were 87.36%/0.169, 96.70%/0.102, and 97.61%/0.071, respectively. Predicting seizures for early warnings was also feasible. The implemented compression and quantization achieved a significant saving of power and memory with an accuracy degradation of less than 0.5%. Edge DL models achieved performance comparable to many prior implementations that had no time or computational resource limitations. Generic microcontrollers can provide the required memory and computational resources, while model designs can be migrated to ASICs for further optimization. The results suggest that edge DL inference is a feasible option for future neural implants to improve classification performance and therapeutic outcomes.
