Real-Time DCT Learning-based Reconstruction of Neural Signals

Abstract

Wearable and implantable body sensor network systems are one of the key technologies for continuous monitoring of patient's vital health status such as temperature and blood pressure, and brain activity. Such devices are critical for early detection of emergency conditions of people at risk and offer a wide range of medical facilities and services. Despite continuous advances in the field of wearable and implantable medical devices, it still faces major challenges such as energy-efficient and low-latency reconstruction of signals. This work presents a power-efficient real-time system for recovering neural signals. Such systems are of high interest for implantable medical devices, where reconstruction of neural signals needs to be done in realtime with low energy consumption. We combine a deep network and DCT-learning based compressive sensing framework to propose a novel and efficient compression-decompression system for neural signals. We compare our approach with state-of-the-art compressive sensing methods and show that it achieves superior reconstruction performance with significantly less computing time.

Sparse Signal Representation, Sampling, and Recovery in Compressive Sensing Frameworks

“…The results obtained were favourable and a positive sign for researchers working on learning-based techniques. Similarly, the limitations of random subsampling can be mitigated by using structured subsampling, as demonstrated in [39] and [40]. In order to improve image recovery in the CS domain, an image dataset is used to train a sampling matrix and for real time neural signals reconstruction [40].…”

“…Similarly, the limitations of random subsampling can be mitigated by using structured subsampling, as demonstrated in [39] and [40]. In order to improve image recovery in the CS domain, an image dataset is used to train a sampling matrix and for real time neural signals reconstruction [40]. Using deep Convolutional Neural Networks (CNN) and the linear reconstruction method, this work adopted the training mechanism proposed by [39] for real-time applications.…”

Learning-Based Optimization of the Under-Sampling Pattern in MRI

“…However, these machine learning-based methods are typically optimized for a specific under-sampling pattern provided by the user. Furthermore, there are also techniques that are optimizing the subsampling patterns for given reconstruction methods [27,28,29,30]. The reconstruction model's performance will depend significantly on the sub-sampling pattern.…”