Deep learning in ultrasound elastography imaging: A review.

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Artificial Intelligence

Background:

• Altered tissue mechanical properties are linked to disease onset and progression.
• Ultrasound elastography assesses tissue stiffness via strain or shear wave measurements.
• Deep learning is increasingly utilized in medical imaging research.

Purpose of the Study:

• To review deep learning frameworks applied to ultrasound elastography.
• To summarize recent advances in ultrasound elastography using deep learning.
• To discuss future directions for deep learning in this field.

Main Methods:

• Description of common deep learning frameworks (MLP, CNN, RNN).
• Review of recent studies applying deep learning to ultrasound elastography.
• Analysis of algorithm development and clinical diagnostic applications.

Main Results:

• Deep learning models show promise in enhancing ultrasound elastography accuracy.
• Applications span various diagnostic areas, improving disease detection.
• Integration of deep learning advances algorithm development and clinical utility.

Conclusions:

• Deep learning offers significant potential to advance ultrasound elastography.
• Further research is needed to address current challenges and optimize clinical translation.
• Future developments may lead to more precise disease diagnosis and monitoring.