Use and application of improved golden sine optimization algorithm in data hiding

Data hiding plays a crucial role in preserving individual privacy and ensuring the security of critical information in the modern world. The confidentiality of medical imaging data, in particular, has drawn significant attention due to the requirements of patient privacy and data security. In this context, the security of images in the Digital Imaging and Communications in Medicine (DICOM) format is vital not only for the protection of individual rights but also for enhancing the reliability of medical data sharing and storage processes. In this study, an innovative approach called the Binary Improved Golden Sine Optimization Algorithm (BGoldSA-II) was developed and applied for data hiding in DICOM images. The proposed method combines the aesthetic and mathematical advantages of the golden ratio with advanced optimization techniques to achieve high data embedding capacity and superior image quality. This hybrid algorithm aims to outperform traditional methods by offering lower error rates and higher security levels during the data hiding process. Experimental studies have demonstrated that the proposed method exhibits competitive performance in both theoretical and practical contexts. In 512×512-sized DICOM images, quality metrics for embedding 1000 characters were measured as follows: Peak Signal-to-Noise Ratio (PSNR) of 71.36 dB, Mean Squared Error (MSE) of 0.004, Root Mean Squared Error (RMSE) of 0.066, Structural Similarity Index (SSIM) of 0.99, and Universal Image Quality Index (UIQ) of 0.99. For embedding 5000 characters in the same image size, the quality metrics were PSNR 62.62 dB, MSE 0.033, RMSE 0.182, SSIM 0.97, and UIQ 0.99. In 256×256-sized DICOM images, the quality metrics for embedding 1000 characters were PSNR 65.25 dB, MSE 0.018, RMSE 0.135, SSIM 0.98, and UIQ 0.99, while for embedding 5000 characters, they were PSNR 56.54 dB, MSE 0.136, RMSE 0.36, SSIM 0.89, and UIQ 0.99. These results confirm that BGoldSA-II offers an effective solution for steganography applications and introduces significant innovations compared to existing methods in the literature. This study makes two notable contributions to the literature. First, it expands the use of meta- heuristic algorithms like BGoldSA-II in steganography processes, broadening their application in the field of data security. Second, the proposed method for ensuring DICOM data confidentiality provides a practical framework and serves as a guide for future research. In this regard, BGoldSA-II aims to open new avenues in the field of information security, both in academic research and practical applications.