Image quality assessment metric for frame accumulated image

Area of Science:

• Medical Imaging
• Image Quality Assessment
• Signal Processing

Background:

• Medical image quality is crucial for accurate diagnosis.
• Current objective metrics inadequately assess images from frame accumulation technology.
• Gray-scale resolution is often overlooked, focusing primarily on spatial resolution and display limitations.

Purpose of the Study:

• To propose a novel metric, mean signal-to-noise ratio (MSNR), for evaluating frame-accumulated medical images.
• To address the limitations of existing metrics in assessing gray-scale resolution and precision beyond standard display capabilities.

Main Methods:

• Developed the mean signal-to-noise ratio (MSNR) metric based on signal-to-noise principles.
• Utilized the mean of multiple images as a reference image for quality assessment.
• Calculated MSNR for image sets with varying frame accumulation levels under constant illumination.

Main Results:

• The proposed MSNR metric provides a more reasonable evaluation of frame-accumulated medical image quality.
• Experimental results demonstrate MSNR's effectiveness in assessing images exceeding original gray scale and precision.
• Compared to other methods, MSNR is simpler and more suitable for frame-accumulated images.

Conclusions:

• MSNR is a promising metric for objective quality assessment of frame-accumulated medical images.
• The metric overcomes limitations of traditional methods by considering enhanced gray-scale resolution.
• MSNR offers a more effective and straightforward approach to medical image quality evaluation.