Published Paper Details:

Abstract: In the changing landscape of digital healthcare, protecting the integrity and confidentiality of accurate results depend on the quality of medical imagery becoming increasingly important. Watermarking in medical imaging serves as a robust technique to embed critical patient related information through medical imaging to confirm the integrity of medical content, verify ownership, and detect tampering. It is presented as a multi-format watermarking system, capable of digitally marking medical images types: DICOM, PNG, and JPGand also maximizing the trade-offs of watermark capacity and image quality. We executed and evaluated three popular techniques: A tool to separate important features from a dataset, a method to convert image data into frequency components and principal Component Analysis (PCA). The research incorporated the invisibly watermarking the part of non-Interest, allowing the diagnostic integrity of the relevant area to be preserved and retained. The approach consists of preprocessing the images, watermark embedding and extraction using the different algorithms, and evaluating the results using objective measures such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Mean Squared Error (MSE). The results of the experiment indicate that SVD performs better than both PCA and DCT for most cases in terms of imperceptibility and robustness, meaning SVD showed higher PSNR and SSIM across all image formats. PCA showed similar performance to SVD but slower computation time and DCT showed a faster computation time but more distortion. The comparison across different methods indicates strengths and weaknesses in how each method performs on variations of different image formats and watermark payloads. This research provides a complete and scalable system to protect medical images watermarking along with guidelines for implementation in telemedicine, cloud-based electronic health record systems, and medical archives. The approach minimizes visual degradation while addressing concerns related to secure data storage and the security, authenticity, and traceability of images in digital healthcare systems.