Artificial Neural Networks-based Real-time Classification of ENG Signals for Implanted Nerve Interfaces
2403.20234
0
0
🧠
Abstract
Neuropathies are gaining higher relevance in clinical settings, as they risk permanently jeopardizing a person's life. To support the recovery of patients, the use of fully implanted devices is emerging as one of the most promising solutions. However, these devices, even if becoming an integral part of a fully complex neural nanonetwork system, pose numerous challenges. In this article, we address one of them, which consists of the classification of motor/sensory stimuli. The task is performed by exploring four different types of artificial neural networks (ANNs) to extract various sensory stimuli from the electroneurographic (ENG) signal measured in the sciatic nerve of rats. Different sizes of the data sets are considered to analyze the feasibility of the investigated ANNs for real-time classification through a comparison of their performance in terms of accuracy, F1-score, and prediction time. The design of the ANNs takes advantage of the modelling of the ENG signal as a multiple-input multiple-output (MIMO) system to describe the measures taken by state-of-the-art implanted nerve interfaces. These are based on the use of multi-contact cuff electrodes to achieve nanoscale spatial discrimination of the nerve activity. The MIMO ENG signal model is another contribution of this paper. Our results show that some ANNs are more suitable for real-time applications, being capable of achieving accuracies over $90%$ for signal windows of $100$ and $200,$ms with a low enough processing time to be effective for pathology recovery.
Create account to get full access
Overview
- Neuropathies are nerve disorders that can seriously impact a person's life
- Fully implanted devices are emerging as a promising solution to support patient recovery
- These devices, though part of a complex neural system, face many challenges
- This paper focuses on one challenge: classifying motor and sensory stimuli from electronic signals in the sciatic nerve
Plain English Explanation
Nerve disorders, called neuropathies, can be very serious and affect a person's daily life. To help patients recover, doctors are starting to use fully implanted devices. These devices are part of a complex system that interacts with the nerves, but they come with their own challenges.
This paper looks at one of those challenges - how to take the electronic signals coming from the nerves and use them to identify different types of sensory and movement information. The researchers tested four different kinds of artificial neural networks to see how well they could classify these signals. They used data from experiments on rats, where they measured the electronic signals in the sciatic nerve.
The goal was to find neural network models that could accurately classify the signals in real-time, which is important for the devices to work effectively and help patients recover from their nerve disorders. The researchers also looked at how quickly the neural networks could process the information, since real-time performance is crucial.
Technical Explanation
The paper models the electronic signals from the nerve (called the electroneurographic or ENG signal) as a multiple-input, multiple-output (MIMO) system. This reflects how state-of-the-art implanted nerve interfaces use multiple electrodes to pick up spatial information about nerve activity at a very fine scale.
The researchers tested four different types of artificial neural networks (ANNs) to classify the motor and sensory stimuli from the ENG signal. They looked at the performance of the ANNs in terms of accuracy, F1-score (a measure of precision and recall), and prediction time, using different sizes of datasets.
The results show that some of the ANN models are more suitable for real-time applications. Certain models were able to achieve over 90% accuracy in classifying the signals from 100 ms and 200 ms time windows, while also having a fast enough processing time to be effective for assisting in patient recovery.
Critical Analysis
The paper provides a thorough technical evaluation of using ANNs for classifying nerve signals, which is an important step towards developing effective implanted devices for neuropathy patients. However, the research is still at an early stage, and there are some limitations to consider.
The experiments were conducted on rats, so the findings may not directly translate to humans. Additionally, the paper does not address potential challenges with long-term reliability and stability of the implanted devices, which will be crucial for real-world clinical applications.
Further research is needed to validate the findings in human patients and assess the practical implementation challenges. Aspects like power consumption, heat dissipation, and biocompatibility of the devices will also need to be carefully evaluated.
Conclusion
This paper demonstrates the potential of using artificial neural networks to classify motor and sensory signals from the nerves, which is an important step towards developing fully implanted devices to assist neuropathy patients. The researchers found that certain ANN models can achieve high accuracy and fast processing times, making them promising candidates for real-time applications.
While the findings are encouraging, more research is needed to validate the results in human patients and address the practical challenges of implementing these devices. Nonetheless, this work contributes to the ongoing efforts to improve the lives of those suffering from debilitating nerve disorders.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
🧠
Intelligent Neural Interfaces: An Emerging Era in Neurotechnology
Mahsa Shoaran, Uisub Shin, MohammadAli Shaeri
0
0
Integrating smart algorithms on neural devices presents significant opportunities for various brain disorders. In this paper, we review the latest advancements in the development of three categories of intelligent neural prostheses featuring embedded signal processing on the implantable or wearable device. These include: 1) Neural interfaces for closed-loop symptom tracking and responsive stimulation; 2) Neural interfaces for emerging network-related conditions, such as psychiatric disorders; and 3) Intelligent BMI SoCs for movement recovery following paralysis.
6/3/2024
🧠
Brain-Inspired Spiking Neural Networks for Industrial Fault Diagnosis: A Survey, Challenges, and Opportunities
Huan Wang, Yan-Fu Li, Konstantinos Gryllias
0
0
In recent decades, Industrial Fault Diagnosis (IFD) has emerged as a crucial discipline concerned with detecting and gathering vital information about industrial equipment's health condition, thereby facilitating the identification of failure types and severities. The pursuit of precise and effective fault recognition has garnered substantial attention, culminating in a focus on automating equipment monitoring to preclude safety accidents and reduce reliance on human labor. The advent of artificial neural networks (ANNs) has been instrumental in augmenting intelligent IFD algorithms, particularly in the context of big data. Despite these advancements, ANNs, being a simplified biomimetic neural network model, exhibit inherent limitations such as resource and data dependencies and restricted cognitive capabilities. To address these limitations, the third-generation Spiking Neural Network (SNN), founded on principles of Brain-inspired computing, has surfaced as a promising alternative. The SNN, characterized by its biological neuron dynamics and spiking information encoding, demonstrates exceptional potential in representing spatiotemporal features. Consequently, developing SNN-based IFD models has gained momentum, displaying encouraging performance. Nevertheless, this field lacks systematic surveys to illustrate the current situation, challenges, and future directions. Therefore, this paper systematically reviews the theoretical progress of SNN-based models to answer the question of what SNN is. Subsequently, it reviews and analyzes existing SNN-based IFD models to explain why SNN needs to be used and how to use it. More importantly, this paper systematically answers the challenges, solutions, and opportunities of SNN in IFD.
6/5/2024
Listen to the Waves: Using a Neuronal Model of the Human Auditory System to Predict Ocean Waves
Artur Matysiak, Volker Roeber, Henrik Kalisch, Reinhard Konig, Patrick J. C. May
0
0
Artificial neural networks (ANNs) have evolved from the 1940s primitive models of brain function to become tools for artificial intelligence. They comprise many units, artificial neurons, interlinked through weighted connections. ANNs are trained to perform tasks through learning rules that modify the connection weights. With these rules being in the focus of research, ANNs have become a branch of machine learning developing independently from neuroscience. Although likely required for the development of truly intelligent machines, the integration of neuroscience into ANNs has remained a neglected proposition. Here, we demonstrate that designing an ANN along biological principles results in drastically improved task performance. As a challenging real-world problem, we choose real-time ocean-wave prediction which is essential for various maritime operations. Motivated by the similarity of ocean waves measured at a single location to sound waves arriving at the eardrum, we redesign an echo state network to resemble the brain's auditory system. This yields a powerful predictive tool which is computationally lean, robust with respect to network parameters, and works efficiently across a wide range of sea states. Our results demonstrate the advantages of integrating neuroscience with machine learning and offer a tool for use in the production of green energy from ocean waves.
4/16/2024
🤿
On the Feasibility of Deep Learning Classification from Raw Signal Data in Radiology, Ultrasonography and Electrophysiology
Szilard Enyedi
0
0
Medical imaging is a very useful tool in healthcare, various technologies being employed to non-invasively peek inside the human body. Deep learning with neural networks in radiology was welcome - albeit cautiously - by the radiologist community. Most of the currently deployed or researched deep learning solutions are applied on already generated images of medical scans, use the neural networks to aid in the generation of such images, or use them for identifying specific substance markers in spectrographs. This paper's author posits that if the neural networks were trained directly on the raw signals from the scanning machines, they would gain access to more nuanced information than from the already processed images, hence the training - and later, the inferences - would become more accurate. The paper presents the main current applications of deep learning in radiography, ultrasonography, and electrophysiology, and discusses whether the proposed neural network training directly on raw signals is feasible.
4/16/2024