Ismail Ariffin

A new semantic mining approach for detecting ventricular tachycardia and ventricular fibrillation

Accurately differentiating between ventricular fibrillation (VF) and ventricular tachycardia (VT) episodes is crucial in preventing potentially fatal misinterpretations. If VT is misinterpreted as VF, the patient will receive an unnecessary shock that could damage the heart; conversely, if VF is incorrectly interpreted as VT, the result will be life-threatening. In this study, a new method called semantic mining is used to characterize VT and VF episodes by extracting their significant characteristics (the frequency, damping coefficient and input signal). This newly proposed method was tested using a widely recognized database provided by the Massachusetts Institute of Technology (MIT) and achieved high detection accuracy of 96.7%. The semantic mining technique was capable of completely discriminating between normal rhythms and VT and VF episodes without any false detections and also distinguished VT and VF episodes from one another with a recognition sensitivity of 94.1% and 95.2% for VT and VF, respectively.

A novel FPGA based platform for ultrasound power measurement

Ultrasound power meter is widely used in health care, educational and industrial institutions. It is used to measure the power output of ultrasound machine for diagnostic, therapeutic and non-destructive testing purposes. The current ultrasound machine is high cost and does not have further data processing for analysis. In order to reduce the cost of data processing unit and enable further analysis, a Field-Programmable Gate Array (FPGA) based ultrasound processing platform has been proposed. This platform is able to process the data from two polyvinylidene fluoride (PVDF) sensors and one temperature sensor. It is prepared to measure the ultrasound frequency from 500 kHz to 10 MHz with temperature range from 10 °C to 50 °C and power range from 1mW/cm2 up to 10W/cm2 (based on formula, the resolution is 0.87mW/cm2). For the further analysis, a graphical user interface (GUI) has been developed. The GUI receives the data from the ultrasound processing unit through a serial communication port and display the result in the graph. This FPGA has been tested and can be used as platform for ultrasound power meter. Test result shows that the platform is able to process 10µs ultrasound data with 20ns time-domain resolution and maximum amplitude of 2Vp with 0.12mVpp resolution then display these waveform and calculation result in the GUI.

Design of an A-mode ultrasound amplifier for bone porosity detection

This paper presents the development of an A-mode ultrasound circuit in bone densitometry system. There are other bone densitometry systems available such as X-Ray, MRI and others. However, the system are found to have several disadvantages such as involving ionizing radiation, taking longer time to setup, and so on. In addition, demand for the bone diagnostic is also increasing and due to that the idea to develop ultrasound circuit has emerged. The information regarding the ultrasound circuit has been studied as well as parameters related for bone density measurement. 1MHz frequency has the highest absorption for bone. The ultrasound circuit consists of a pulse generator system, high voltage pulse circuit, transducer for transmitter and receiver part, instrumentation amplifier and oscilloscope. The FPGA is chosen as a pulse generator and output from the FPGA is connected to MOSFET to create a pulse signal using a switching technique and enables voltage amplitude of the signal to be changed. The circuit was tested on human arm and heel to evaluate the signal strength with varying distance between the transducer and voltage signal amplitude. This testing proved that the A-mode ultrasound circuit is successfully transmitted ultrasound signal through human heel.

MATLAB Implementation of Face Identification Using Principal Component Analysis

Face is the greatest superior biometric as the face has a complex, multidimensional and meaningful identity compared from one person to another. Face identification is executed by comparing the characteristics of the face (test image) with those of known individual images in the database. This paper describes the used of the Principal Component Analysis (PCA) algorithm for human face identification based on webcam image. The MATLAB is used as a tool for image processing and analysis. The important decision to identify the person is by the minimum distance of the face images and known face images in face space. From the results, it can be concluded that the work has successfully implemented the PCA algorithm for human face identification through a webcam.