Perceptibility curve test for digital radiographs before and after application of various image processing algorithms

Area of Science:

• Radiology
• Medical Imaging
• Digital Image Processing

Background:

• Digital radiography systems like Digora are widely used.
• Image quality in digital radiographs can be influenced by various factors, including exposure settings and post-processing techniques.
• Digital image enhancement methods are applied to improve the diagnostic quality of radiographs.

Purpose of the Study:

• To compare the effectiveness of different digital image enhancement techniques on Digora radiographs.
• To evaluate image quality using a perceptibility curve (PC) test.
• To determine the optimal enhancement method for improving detail perception.

Main Methods:

• Radiographs of an aluminum test object with varying hole sizes were acquired using the Digora system.
• Original digital images underwent several enhancement methods: colour-coded, contrast/brightness, histogram equalization, and negative modes.
• Eight observers assessed image perceptibility using a modified PC test, with data analyzed using repeated-measures ANOVA and Bonferroni/Dunn tests.

Main Results:

• Histogram equalization yielded the highest number of perceptible details across optimal and wide exposure ranges.
• Significant differences in detail perception were observed with various enhancement modalities at low and middle exposure ranges.
• No significant differences in detail perception were found among enhancement methods at high exposure ranges.

Conclusions:

• Histogram equalization provides subjectively superior image quality for Digora radiographs.
• Contrast and brightness enhancement are the next most effective methods for improving image quality.
• Digital image enhancement, particularly histogram equalization, is crucial for optimizing diagnostic information in digital radiography.