Reverse time migration and genetic algorithms Combined for Reconstruction in Transluminal Shear Wave Elastography: An...

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computational Science

Background:

• Transluminal shear wave elastography faces challenges in solving inverse problems for accurate thermal lesion reconstruction.
• High Intensity Focused Ultrasound (HIFU) is used for prostate cancer treatment, generating thermal lesions that require precise identification.
• Existing methods may have limitations in accuracy and computational efficiency for complex inverse problems.

Purpose of the Study:

• To propose and evaluate a new reconstruction approach combining Reverse Time Migration (RTM) and Genetic Algorithms (GAs) for transluminal shear wave elastography.
• To optimize the RTM method for identifying thermal lesions generated by HIFU in prostate cancer treatment.
• To assess the combined RTM-GA approach's accuracy, computational time, and feasibility compared to GA alone.

Main Methods:

• Developed a combined reconstruction approach integrating RTM and GAs to solve the inverse problem in elastography.
• Optimized the RTM method using various cross-correlation techniques and device configurations (emitters/receivers).
• Utilized RTM-identified high-correlation areas to guide the GA in locating HIFU lesions and estimating stiffness/viscosity changes.

Main Results:

• The combined RTM-GA approach demonstrated lower error in reconstructed values and reduced computational time compared to GA alone.
• The best RTM performance was achieved with a novel cross-correlation method and a 3-emitter, 32-receiver configuration.
• Accurate localization of HIFU lesions and stiffness contrast were observed, though viscosity ratio reconstruction had higher errors.

Conclusions:

• The combined RTM-GA approach shows promise for accurate and efficient reconstruction of thermal lesions in HIFU-treated prostate cancer.
• Further validation with diverse scenarios and experimental data is necessary to confirm the approach's full feasibility.
• Optimization of RTM and targeted GA application significantly improve inverse problem-solving in this context.