Medical imaging correction: a comparative study of five contrast and brightness matching methods

Area of Science:

• Medical Imaging
• Image Processing
• Computer Vision

Background:

• Medical image analysis often requires consistent contrast and brightness for accurate comparisons over time.
• Acquisition process variations can lead to significant dissimilarities in medical image data.
• Existing methods range from simple filters to complex recursive techniques for image correction.

Purpose of the Study:

• To comprehensively compare five distinct methods for matching contrast and brightness in medical image pairs.
• To evaluate the performance of these methods across diverse contrast, brightness, and combined variations.
• To identify the most effective technique for enhancing medical image comparability.

Main Methods:

• Five methods were compared: Contrast Stretching, Ruttimann's Robust Film Correction, Boxcar Filtering, Least-Squares Approximation, and Histogram Registration.
• A dataset of 100 medical image pairs, divided into five sets, was used for evaluation.
• Both qualitative (visual assessment, digital subtraction) and quantitative (statistical evaluation, metrics) analyses were performed.

Main Results:

• Histogram Registration demonstrated superior performance in systematically matching contrast and brightness across most evaluated image pairs.
• Qualitative and quantitative analyses confirmed the effectiveness of Histogram Registration over the other four methods.
• The study provides a robust comparison of image enhancement techniques for medical applications.

Conclusions:

• Histogram Registration is the recommended method for contrast and brightness matching in medical imaging applications requiring accurate comparisons.
• The findings support the use of Histogram Registration for improving the reliability of longitudinal medical image analysis.
• This research contributes to the field by providing empirical evidence for the efficacy of different image enhancement strategies.