An Effective Secure Medical Image Transmission Using Improved Particle Swarm Optimization and Wavelet Transform
Introduction
Internet of Things (IoT) devices are commonly employed for interconnecting existing medicinal sources and offer a reliable, efficient, and intelligent healthcare service to ageing patients with chronic illnesses. Magnetic resonance imaging (MRI) is filled with various advantages when compared with alternate medical imaging models and plays a vital role in neuroscience study and common neurotic anomaly detection. It contains differential points of interest to apply the imaging in diverse methods since it is capable of showing the action of the human brain [1]. Moreover, it is more secure if it is contrasted with alternate imaging methods. Hence, several approaches are presented to analyze MRI data [2]. The symptomatic evaluations of MRI are developed under the application of programmed as well as the accurate separation of MRI images.
Recently, many developers have projected various methodologies to attain the desired goal. In previous decades, various methodologies have been applied to realize the neurotic brain. In the case of dividing MRI images, the application of patient information should meet the congruity among data usage and computing capable failure due to the unintentional distribution of personal details. This is referred to as a promising application to develop a novel approach for secured Internet of Medical Things (IoMT) [3], w'hich ensures the legal accessing and application of patient’s data. The privacy of a patient’s private data has become endangered by digital health details that are distributed across insecure teams. It provides a negative impact on patients as wfell as legitimate parties. Therefore, the security of patient’s data is a challenging factor for developers to apply these data in a finite analyzing process.
The recently identified quantum process of IoMT in healthcare segments is related to the healthcare of a patient, which is comprised of mobility and makes a vulnerable network for diverse kinds of attacks. Therefore, the security, privacy, trustworthiness, and simultaneously of a patient’s therapeutic as well as sensitive details are named as a potential challenge in the healthcare platform. In [4], a new method to tackle patient’s secured information, a discrete wavelet transform (DWT) technique, has been presented to apply permanence as well as self-sufficiency characteristics of the blockchain. Furthermore, modification of clashing interests to ensure privacy at the time of providing permission to apply data in legit purposes for analyzing process is discussed. To assure the privacy of patients Electronic Health Records (EHRs) and motivate the aforementioned trade-off, only a few methods are mentioned in [5].
Privacy-conserving technologies focus on counteracting divulgence of simple personal data, at the time of developing affected data applied in licit problems. There are diverse experiments carried out for sensitive data integrity that should be transformed to provide a ground-breaking impact on the concerned problem. The application of such models enables the interchange among data security and classification, to provide data usage to limit the malicious breach in data security. The combination of IoT and data transmission makes a vital tool with effective analyzing ability to ensure the security of patient’s data. Since the medical images are enclosed with patients’ personalized information, privacy protection has increased amid transmitting clinical images through the web [6]. Therefore, steganography is applied for providing privacy as well as confidentiality by protecting the private information of patients from therapeutic images.
The study ensures privacy and security of patient’s therapeutic data and finds the data that has not been acquired. On the other hand, it attains a desired classification value, with extra details in the computer-aided diagnosis (CAD) technique. For accomplishing this process, the deployed model uses steganography to cover the personal data of a patient inside an MRI system before undergoing medical analysis. Therefore, it is applied with spatial domain image steganography to incorporate patients’ information and research on a state-of-the-art mechanism where it applies a generic and common steganography system.
The modified image along with a slight color difference is expected to be indistinguishable. Researchers have tried to explore the applicable positions to develop an embed highest invisible method. Yang et al. [7] have employed a least significant bits (LSB) substitution-relied hiding that resolves noise-sensitive areas to attain optimal visual quality as well as the smoothing fact of the stego image (SI). In [8], protective data transmission of sensitive contents across the Internet as attained by dual-level security, which can be reached by cyclicl8 LSB substitution on encrypted images, is discussed. Finally, the simulation outcome of this method offers good imperceptibility and security.
In [9], the projected multi-pixel differencing (MPD) and LSB steganographic procedure to incorporate the private data with edges and smooth blocks is discussed. Muhammad et al. [Ю] implied a secured image steganographic procedure where secret data undergoes encryption with the application of a multi-level encryption model, and embeds from a cover image by applying the stego key directed adaptive LSB substitution. Consecutively, in [11], researchers proposed an approach in LSB substitution to combine more volume of payload on edges of the image.
Muhammad et al. [12] have deployed the Magic LSB Substitution Method (M-LSB- SM)-relied steganographic framework to corporate the secret data on the cover image. The cover image is transformed from red green blue (RGB) to hue saturation intensity (HSI) colorspace to attain encrypted secret details.
Laskar and Hemachandran [13] employed a transposition cipher-based cryptography on payload before forwarding to an LSB substitution steganography on the cover image. Joo et al. [14] projected a pixel-value differencing (PVD) mechanism that applies floor and modulus function to provide optimal security from attackers. Rahim et al. [15] deployed a private data protection module under the application of deep autoencoder (DAE) for rapid image retrieving mobile devices. The developed model ensures easy image retrieval on the application of the latest mobile devices. Only some of the analyzes to use image steganography on security and legitimacy of visual social media are demonstrated. In [16], a discrete cosine transform (DCT) relying on a data covering principle has been deployed for concealing color data from gray-level images. It is evident that the image steganographic method applying the LSB substitution procedure is utilized in many types of research due to enough alarming efficiency. An easy embed method could be used with various data types. The image pixels presented in a spatial domain can be modified to attain secret data. Higher imperceptibility and capacity are fundamental in ensuring secured transmission of secret data. The LSB of the cover image has been embedded with bits of secret message.
This study introduces an improved particle swarm optimization (IPSO)-DWT approach to effectively select the pixels to embed the secret image. The IPSO-DWT algorithm makes use of fitness function (FF), with respect to cost function (CF). The CF evaluates the fitness based on the determination of the edge, entropy, and pixel intensities to validate FF. An elaborate experimental analysis takes place to validate the goodness of the IPSO-DWT algorithm. The results ensured the optimal outcome of the IPSO-DWT algorithm in terms of mean square error (MSE) and peak signal to noise ratio (PSNR) over compared methods.
