UltraFlex: Iterative Model-Based Ultrasonic Flexible-Array Shape Calibration.

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Biomedical Engineering

Background:

• Flexible and wearable ultrasonic arrays offer potential for advanced medical imaging.
• Accurate calibration of array shape is crucial for optimal ultrasound image quality.
• Existing calibration methods may not be robust or accurate enough for complex array geometries.

Purpose of the Study:

• To evaluate the performance of an iterative model-based ultrasonic flexible-array shape calibration framework (UltraFlex).
• To assess various image quality metrics for their effectiveness in array shape calibration.
• To determine the most robust and accurate metrics for flexible ultrasonic array calibration.

Main Methods:

• Utilized an iterative model-based framework (UltraFlex) with automatic differentiation for array shape calibration.
• Evaluated image quality metrics including speckle brightness, envelope entropy, coherence factor, lag-one coherence, common-midpoint correlation coefficient (CMCC), and common-midpoint phase error (CMPE).
• Tested on simulated phantoms, experimental phantoms, and in vivo liver datasets with transducers of known geometries.

Main Results:

• Lag-one coherence, CMCC, and CMPE metrics enable more accurate element position estimations and improved ultrasound image focusing.
• Models based on CMCC and phase error demonstrated robustness against additive white noise.
• Achieved median mean Euclidean errors (MEEs) of 3.7 μm (simulation), 29.7 μm (phantom), and 69.0 μm (in vivo liver data).

Conclusions:

• CMCC and CMPE are superior metrics for ultrasonic flexible-array shape calibration, providing accurate element positioning and robust performance.
• The UltraFlex framework shows significant promise for the development of advanced flexible and wearable ultrasonic arrays.
• This work advances the field of ultrasound calibration, paving the way for next-generation ultrasound devices.