Real-Time Edge Neuromorphic Tasting From Chemical Microsensor Arrays

Abstract

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.

Person identification using deep neural networks on physiological biomarkers during exercise

“…Ablation studies show that the accuracy decreases if any of the sensory input is absent. The networks can be deployed on mobile platforms for real-time processing of these timevarying signals [21], [22]. Results show that we can achieve 22 inferences/sec when running the GRU-200 model on the Jetson Nano using the low power mode of 5 W. This shows that we can easily integrate this module in a wearable device.…”

Analytical Assessment of Sodium Isfet Based Sensors for Sweat Analysis

“…The ISFETs showed good performance Na + quantification in sweat samples, being feasible their integration in compact electronic wearable devices [42,43]. As future work, the ISFETs will be integrated in a CMOS fabricated chip for real-time, continuous monitoring of analytes on-body, allowing correction of interferences using deep learning approaches [44].…”

End-to-End Prediction of Sodium Concentration from Uncalibrated Sodium ISFETs

“…This work was partially funded by SNSF-Sinergia WeCare project (N°CRSII5 177255). However, large datasets are difficult to collect from ISFET sensors for DNN training because of the long time constants of chemical reactions [11]. This work proposes an end-to-end prediction network which is first trained on a large dataset of simulated Na + ISFET recordings.…”