High-Level Modeling and Simulation Tool for Sensor Conditioning Circuit Based on Artificial Neural Networks

In-human testing of a non-invasive continuous low-energy microwave glucose sensor with advanced machine learning capabilities

Continuous glucose monitoring schemes that avoid finger pricking are of utmost importance to enhance the comfort and lifestyle of diabetic patients. To this aim, we propose a microwave planar sensing platform as a potent sensing technology that extends its applications to biomedical analytes. In this paper, a compact planar resonator-based sensor is introduced for noncontact sensing of glucose. Furthermore, in vivo and in-vitro tests using a microfluidic channel system and in clinical trial settings demonstrate its reliable operation. The proposed sensor offers real-time response and a high linear correlation (R2 ∼ 0.913) between the measured sensor response and the blood glucose level (GL). The sensor is also enhanced with machine learning to predict the variation of body glucose levels for non-diabetic and diabetic patients. This addition is instrumental in triggering preemptive measures in cases of unusual glucose level trends. In addition, it allows for the detection of common artifacts of the sensor as anomalies so that they can be removed from the measured data. The proposed system is designed to noninvasively monitor interstitial glucose levels in humans, introducing the opportunity to create a customized wearable apparatus with the ability to learn.

Mobile Phone Usage Detection by ANN Trained with a Metaheuristic Algorithm †

Artificial neural networks (ANN) are widely used to classify high non-linear systems by using a set of input/output data. Moreover, they are trained using several optimization methodologies and this paper presents a novel algorithm for training ANN through an earthquake optimization method. Usually, gradient optimization method is implemented for the training process, with perhaps the large number of iterations leading to slow convergence, and not always achieving the optimal solution. Since metaheuristic optimization methods deal with searching for weight values in a broad optimization space, the training computational effort is reduced and ensures an optimal solution. This work shows an efficient training process that is a suitable solution for detection of mobile phone usage while driving. The main advantage of training ANN using the Earthquake Algorithm (EA) lies in its versatility to search in a fine or aggressive way, which extends its field of application. Additionally, a basic example of a linear classification is illustrated using the proposal-training method, so the number of applications could be expanded to nano-sensors, such as reversible logic circuit synthesis in which a genetic algorithm had been implemented. The fine search is important for the studied logic gate emulation due to the small searching areas for the linear separation, also demonstrating the convergence capabilities of the algorithm. Experimental results validate the proposed method for smart mobile phone applications that also can be applied for optimization applications.