We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Machine Learning-Based Adaptive Moment Gradient for Electrical Impedance Tomography.
- Authors
Idaamar, Soumaya; Louzar, Mohamed; Lamnii, Abdellah; Ben Rhila, Soukaina
- Abstract
Electrical Impedance Tomography (EIT) reconstructs the electrical properties of cellular tissues by taking measurements at their boundaries. Its non-invasive nature, safety, and potential for an extensive variety of therapeutic uses have sparked significant interest. The generation of EIT images involves solving an inverse problem through iterative optimization techniques. As a result, the effectiveness of various optimization strategies for EIT may vary. This study evaluates the efficacy of the Adam optimizer in addressing the EIT inverse problem. Our simulations reveal that Adam demonstrates superior convergence rates, faster reconstruction times, and higher-quality images compared to traditional gradient descentbased optimizers. Specifically, the Adam optimizer produces images of superior quality with improved anomaly localization, all while achieving faster reconstruction speeds and higher convergence rates. Additionally, we provide insights into the optimal parameter configurations for Adam in the context of EIT, offering valuable guidance for future research endeavors in this domain. All things considered, our findings unequivocally demonstrate that the Adam optimizer is a valuable tactic for resolving the EIT inverse problem.
- Subjects
ELECTRICAL impedance tomography; INVERSE problems; MATHEMATICAL optimization
- Publication
IAENG International Journal of Computer Science, 2024, Vol 51, Issue 6, p688
- ISSN
1819-656X
- Publication type
Article