Automatic dynamic range adjustment for ultrasound B-mode imaging.

Area of Science:

• Medical Imaging
• Ultrasound Technology
• Image Processing

Background:

• Dynamic Range (DR) is crucial for ultrasound image quality but is often manually adjusted, limiting productivity and usability.
• Current methods for DR adjustment, like histogram matching, can lead to over-contrasted images and are not suitable for 3D ultrasound acquisition.
• Manual DR adjustment in ultrasound is time-consuming, user-dependent, and inconsistent, especially in complex 3D imaging scenarios.

Purpose of the Study:

• To introduce a novel Automatic Dynamic Range Adjustment (ADRA) method for medical ultrasound imaging.
• To enhance image quality and user experience by automating DR optimization in both 2D and 3D ultrasound.
• To overcome the limitations of manual DR adjustment and existing automated methods.

Main Methods:

• The proposed ADRA method adaptively adjusts DR by manipulating input images to match a reference image.
• It utilizes the distance ratio between the log average and extreme values of a reference image for DR calculation.
• Image similarity is quantified using the correlation coefficient (CC) to evaluate ADRA performance.

Main Results:

• In vivo experiments showed significant improvements in CC values for kidney (0.6872 to 0.9870) and liver (0.9274 to 0.9939) data compared to fixed DR.
• The ADRA method outperformed the histogram matching technique for both liver and kidney data.
• In 3D abdominal ultrasound, ADRA improved c-plane image quality and reduced shadow artifacts compared to fixed DR, despite a minor CC increase.

Conclusions:

• The ADRA method effectively enhances image quality in 2D and 3D ultrasound B-mode imaging.
• It improves the similarity between reference and input images while eliminating manual user intervention.
• ADRA offers a robust and automated solution for dynamic range optimization in ultrasound, increasing clinical efficiency.