Jim Berilla

A bio-inspired flow sensor

Accurate measurement of the turbulent flow is an important step toward understanding the mechanisms of many unknown phenomena. Turbulence generally can not be easily measured without significantly disturbing the original flow conditions. This paper introduces the efforts that aim to develop a bio-inspired sensor for monitoring turbulent flow. The sensor will consist of an array of micro-pillars or nano-pillars. Piezoelectric elements serve as transductors, which provides a key sensing element in the construction of micro-pillar. A prototype design was fabricated for the micropillar. The performance of sensing principle by this micropillar was evaluated and was found to be sensitive. The micropillar will be further refined into sensing arrays for real time sensing of flow turbulence.

An Innovative Non-invasive ECG Sensor and Comparison Study with Clinic System

A conventional Electrocardiography (ECG) acquisition system uses contact method which requires the electrodes to be in contact with human skin. This study proposes a non-contact sensing platform that can remotely detect the signals in real time. The current sensing system developed in our research was able to detect the ECG signals with an effective distance of 20cm. Digital signal processing algorithms including digital filer, peak identification was developed to decimate the signal noise and compute the period. The heart rate (HR), heart rate variability (HRV), breath frequency can be determined. The performance is compared with conventional ECG system in clinic study. It shows high potential to collect effective information for clinic application.

The Electroencephalographic Response during a Driving Process: Normal Driving, Turning and Collision

The objective of this study is to estimate the drivers cognitive state under different situation and driving scenarios by analyzing EEG signal. The experiment was conducted on a high fidelity driving simulator. Three different scenarios were designed to investigate the drivers state during normally driving (braking and accelerating) and turning (left or right). In the daily-urban scenario, a deer collision occurred. During each driving trial, the EEG signal was recorded by a 32-lead system. The EEG dynamics were analyzed using power spectrum density (PSD), event-related potential (ERP), independent component analysis (ICA) and time-frequency analysis (TFA). In normally driving, the EEG responses are similar during the three different scenarios. The power concentrates on the frontal and back lobes. During turning, the EEG response is somehow different from that in normally driving (only braking and accelerating). In the deer-collision event, a rapid response is apparent in the ERP mapping.

An innovative non-contact ECG sensor for monitoring heart disease

This paper describes the development of a non-contact sensing platform to monitor the ECG signals. The non-contact sensing will be based on capacitive coupling the bioelectricity produced by cardiovascular activities around the heart. High sensitivity sensor and electronics are designed to amplify the signals. Our preliminary study has pointed to the promise of this sensing concept. A sensor prototype was able to clearly detect the ECG signals from 10 cm away from the body. Research tasks continue improving the sensor design to detect the polarization in the ECG signals. The final goal is a non-contact sensing platform for ECG signals and for real time diagnostics of the mental distress and cardiovascular diseases.