The Optimal Model for Copy-Move Forgery Detection in Medical Images.

Area of Science:

• Digital image forensics
• Medical imaging analysis
• Computer vision

Background:

• Digital devices facilitate the creation of counterfeit medical images.
• Copy-move forgery detection (CMFD) in medical imaging is critical due to potential misuse in resource-limited settings.
• Image forgery directly impacts clinical decision-making and patient care.

Purpose of the Study:

• To develop an optimal method for detecting copy-move forgery in medical images.
• To enhance the reliability and authenticity of digital medical records.

Main Methods:

• The proposed method utilizes an evolutionary algorithm for robust fake block detection.
• Discrete Cosine Transform (DCT) is applied for signal-level image processing.
• Discrete Wavelet Transform (DWT) segments the image, with the low-low band analyzed.
• An equilibrium optimization algorithm searches blocks for forgery detection.

Main Results:

• The method achieved high performance metrics: 90.07% precision, 92.34% recall, and 91.56% F1-score.
• The proposed approach demonstrated superiority over existing methods when tested on medical images.
• The technique effectively identifies duplicated image regions indicative of forgery.

Conclusions:

• The developed method offers a more accurate solution for copy-move forgery detection in medical images.
• This advancement contributes to the integrity of medical diagnostic information.
• The findings support the use of evolutionary algorithms in securing medical image data.