Explainable Machine Learning based Control Charts for High-Dimensional Non-Stationary Time Series Data in IoT Systems: Challenges, Methods, and Future Directions
Aamir Saghir, Kim Duc Tran, Kim Phuc Tran · 2025
One of the most prevalent cardiovascular disorders and a leading cause of death and disability worldwide is brain stroke. It has been recognised as a significant issue with healthcare resources throughout the last few decades. Although the currently available technologies, such as magnetic resonance imaging, computed tomography, positron emission tomography, ultrasound, and X-ray imaging, are very expensive, large, and difficult for patients in remote hospitals to acquire and afford. Due to its quick, non-ionizing, light weight, affordable, and high image resolution capabilities, an electromagnetic (EM) head imaging system is a reliable method that can play a fundamental role in diagnosing strokes inside the human head. In order to identify a stroke inside the brain using an internet of things-enabled EM head imaging system, this chapter designs and fabricates wideband compact metamaterial (MTM) loaded three-dimensional (3D) antennas with tissue-mimicking head phantoms. First, 3D antennas with MTM loads are planned, analysed, and made. The antennas are made up of two slotted dipole components with a directional radiation pattern achieved by a folded parasitic resonator and a finite array of MTM unit cells. Second, a specially designed wideband frequency-dispersive dielectric head phantom with tissue-like characteristics is created. The phantom is made from materials that simulate fake tissue and is assembled afterwards layer by layer into a 3D head model, greatly enhancing the accuracy of the experimental validation procedure for EM head imaging system.