Application of machine learning to predict ultrasound wave propagation in biphasic fluid–solid media
Date of Award
2025
Document Type
Thesis
Degree Name
Master of Science in Engineering (MSE)
Department
Mechanical and Aerospace Engineering
Committee Chair
Sarma L. Rani
Committee Member
Kader Frendi
Committee Member
Sivaguru S. Ravidran
Research Advisor
Sarma L. Rani
Subject(s)
Ultrasonic waves, Machine learning, Algorithms, Diagnostic ultrasonic imaging
Abstract
The objective of this thesis is to investigate the application of Machine Learn ing (ML) algorithms to predict ultrasound wave propagation in the cartilage tissue of the knee joint. The learning data necessary for the ML algorithm has been generated through finite-element method (FEM)-based simulations for solving the Biot theory equations governing the propagation of continuous ultrasound through the cartilage. Specifically, we computed the ultrasound-induced dilatations and displacements in the microscale cartilage that is represented as consisting of four zones, namely the chondrocyte cell and its nucleus, the pericellular matrix (PCM) that forms a layer around the chondrocyte, and the extracellular matrix (ECM). The chondrocyte–PCM complex, referred to as the chondron, is embedded in the ECM. The top surface of the ECM layer is subjected to specified amplitude and frequency of continuous ul trasound. The learning data for the ML algorithm was generated at the ultrasound frequencies of 0.5, 0.75, 2, 3, 4, and 5 MHz. This data was used to train the ML algorithm and then used to predict ultrasound propagation at a test frequency of 1 MHz. It was found that the ML algorithm predictions showed excellent agreement with the FEM simulation data at the test frequency. The errors between the FEM and ML results at 1 MHz were than less than 1 %
Recommended Citation
Dandamudi, Jahnavi, "Application of machine learning to predict ultrasound wave propagation in biphasic fluid–solid media" (2025). Theses. 752.
https://louis.uah.edu/uah-theses/752