Term of Award
Fall 2025
Degree Name
Master of Science, Electrical and Computer Engineering
Document Type and Release Option
Thesis (open access)
Copyright Statement / License for Reuse
This work is licensed under a Creative Commons Attribution 4.0 License.
Department
Department of Electrical and Computer Engineering
Committee Chair
Mohammad Ahad
Committee Member 1
Sungkyun Lim
Committee Member 2
Rocio Alba-Flores
Abstract
Neuromuscular diseases (NMDs) are heterogeneous group of disorders that affect the peripheral nervous system, muscles, and neuromuscular junctions. They pose significant diagnostic challenges due to overlapping symptoms, complex genetic backgrounds, and reliance on resource-intensive testing. Early and precise diagnosis is critical for guided treatment decisions, at the same time, enabling genetic counseling, and improving patient outcomes. Electrical Impedance Myography (EIM) is a non-invasive neurophysiological method for identifying muscle dystrophy. It evaluates the electrical properties of body tissues using a high-frequency, low-amplitude current through the surface electrodes. To detect neuromuscular diseases EIM has been found to be the most effective technique. But, its measurements can also be affected by factors beyond muscle abnormalities. Variations in muscle thickness, subcutaneous fat (SF), and electrode spacing may influence results causing deviations from expected patterns. This thesis paper presents two separate studies. The first study examines how muscle thickness affects EIM measurements and proposes adjusting electrode size to minimize its impact. Using a finite element method (FEM), results show that while muscle thickness alters EIM parameters, resizing the electrodes reduces these variations. The second study presents an artificial intelligence (AI) driven process of generating synthetic EIM measured parameters using cGAN(conditional generative adversarial network), which are then used to train an artificial neural network (ANN) to predict the condition of the subject. With over 97% classification accuracy, the resulting cGAN-ANN model demonstrates the utility of synthetic data in early AI-driven diagnosis of relatively rare neuromuscular diseases.
Recommended Citation
Alam, Md Didarul, "Electrical Impedance Myography for Neuromuscular Disease Diagnosis: Anatomic Factor Mitigation and cGAN-Based Synthetic Data Generation and Validation" (2025). College of Graduate Studies: Theses & Dissertations. 3042.
https://digitalcommons.georgiasouthern.edu/etd/3042
Research Data and Supplementary Material
No
Included in
Bioelectrical and Neuroengineering Commons, Biomedical Commons, Biomedical Devices and Instrumentation Commons, Other Biomedical Engineering and Bioengineering Commons
