scholarly journals Real-Time Edge Neuromorphic Tasting From Chemical Microsensor Arrays

2021 ◽  
Vol 15 ◽  
Author(s):  
Nicholas LeBow ◽  
Bodo Rueckauer ◽  
Pengfei Sun ◽  
Meritxell Rovira ◽  
Cecilia Jiménez-Jorquera ◽  
...  

Liquid analysis is key to track conformity with the strict process quality standards of sectors like food, beverage, and chemical manufacturing. In order to analyse product qualities online and at the very point of interest, automated monitoring systems must satisfy strong requirements in terms of miniaturization, energy autonomy, and real time operation. Toward this goal, we present the first implementation of artificial taste running on neuromorphic hardware for continuous edge monitoring applications. We used a solid-state electrochemical microsensor array to acquire multivariate, time-varying chemical measurements, employed temporal filtering to enhance sensor readout dynamics, and deployed a rate-based, deep convolutional spiking neural network to efficiently fuse the electrochemical sensor data. To evaluate performance we created MicroBeTa (Microsensor Beverage Tasting), a new dataset for beverage classification incorporating 7 h of temporal recordings performed over 3 days, including sensor drifts and sensor replacements. Our implementation of artificial taste is 15× more energy efficient on inference tasks than similar convolutional architectures running on other commercial, low power edge-AI inference devices, achieving over 178× lower latencies than the sampling period of the sensor readout, and high accuracy (97%) on a single Intel Loihi neuromorphic research processor included in a USB stick form factor.

Sensors ◽  
2019 ◽  
Vol 19 (8) ◽  
pp. 1841 ◽  
Author(s):  
Anup Vanarse ◽  
Adam Osseiran ◽  
Alexander Rassau

Recent studies in bioinspired artificial olfaction, especially those detailing the application of spike-based neuromorphic methods, have led to promising developments towards overcoming the limitations of traditional approaches, such as complexity in handling multivariate data, computational and power requirements, poor accuracy, and substantial delay for processing and classification of odors. Rank-order-based olfactory systems provide an interesting approach for detection of target gases by encoding multi-variate data generated by artificial olfactory systems into temporal signatures. However, the utilization of traditional pattern-matching methods and unpredictable shuffling of spikes in the rank-order impedes the performance of the system. In this paper, we present an SNN-based solution for the classification of rank-order spiking patterns to provide continuous recognition results in real-time. The SNN classifier is deployed on a neuromorphic hardware system that enables massively parallel and low-power processing on incoming rank-order patterns. Offline learning is used to store the reference rank-order patterns, and an inbuilt nearest neighbor classification logic is applied by the neurons to provide recognition results. The proposed system was evaluated using two different datasets including rank-order spiking data from previously established olfactory systems. The continuous classification that was achieved required a maximum of 12.82% of the total pattern frame to provide 96.5% accuracy in identifying corresponding target gases. Recognition results were obtained at a nominal processing latency of 16ms for each incoming spike. In addition to the clear advantages in terms of real-time operation and robustness to inconsistent rank-orders, the SNN classifier can also detect anomalies in rank-order patterns arising due to drift in sensing arrays.


2015 ◽  
Vol 24 (6) ◽  
pp. 1703-1711 ◽  
Author(s):  
Rosana Alves Dias ◽  
Filipe Serra Alves ◽  
Margaret Costa ◽  
Helder Fonseca ◽  
Jorge Cabral ◽  
...  

2018 ◽  
Author(s):  
J. I. Alvarez Claramunt ◽  
P. E. Bizzotto ◽  
F. Sapag ◽  
E. Ferrigno ◽  
J. L. Barros ◽  
...  

2017 ◽  
Vol 10 (2) ◽  
pp. 169-178 ◽  
Author(s):  
Shouhei Kidera ◽  
Luz Maria Neira ◽  
Barry D. Van Veen ◽  
Susan C. Hagness

Microwave ablation is widely recognized as a promising minimally invasive tool for treating cancer. Real-time monitoring of the dimensions of the ablation zone is indispensable for ensuring an effective and safe treatment. In this paper, we propose a microwave imaging algorithm for monitoring the evolution of the ablation zone. Our proposed algorithm determines the boundary of the ablation zone by exploiting the time difference of arrival (TDOA) between signals received before and during the ablation at external antennas surrounding the tissue, using the interstitial ablation antenna as the transmitter. A significant advantage of this method is that it requires few assumptions about the dielectric properties of the propagation media. Also the simplicity of the signal processing, wherein the TDOA is determined from a cross-correlation calculation, allows real-time monitoring and provides robust performance in the presence of noise. We investigate the performance of this approach for the application of breast tumor ablation. We use simulated array measurements obtained from finite-difference time-domain simulations of magnetic resonance imaging-derived numerical breast phantoms. The results demonstrate that our proposed method offers the potential to achieve millimeter-order accuracy and real-time operation in estimating the boundary of the ablation zone in heterogeneous and dispersive breast tissue.


2017 ◽  
Vol 5 (5) ◽  
pp. 320-325
Author(s):  
Ahmad T. Jaiad ◽  
Hamzah Sabr Ghayyib

Water is the most precious and valuable because it’s a basic need of all the human beings but, now a day water supply department are facing problem in real time operation this is because less amount of water in resources due to less rain fall. With increase in Population, urban residential areas have increased because of this reasons water has become a crucial problem which affects the problem of water distribution, interrupted water supply, water conservation, water consumption and also the water quality so, to overcome water supply related problems and make system efficient there is need of proper monitoring and controlling system. In this project, we are focusing on continuous and real time monitoring of water supply in IOT platform. Water supply with continuous monitoring makes a proper distribution so that, we can have a record of available amount of water in tanks, flow rate, abnormality in distribution line. Internet of things is nothing but the network of physical objects embedded with electronics, sensors, software, and network connectivity. Monitoring can be done from anywhere as central office. Using Adafruit as free sever data continuously pushed on cloud so we can see data in real time operation. Using different sensors with controller and raspberry pi as Mini computer can monitor data and also control operation from cloud with efficient client server communication.


Sign in / Sign up

Export Citation Format

Share Document