A GRAPH CONVOLUTIONAL-BASED CRYPTOGRAPHIC (GCBC) FRAMEWORK FOR ENHANCED SECURITY AND EFFICIENCY IN HEALTHCARE IMAGE NETWORKS

Abstract

The present invention relates to a Graph Convolutional-Based Cryptographic (GCBC) framework for enhanced security and efficiency in healthcare image networks. The present framework incorporates medical data sourced from Kaggle into a secure architecture. Initial preprocessing is applied to remove noise, followed by the computation of a hash 1 value. Using a generated encryption key, the segmented images undergo an encryption process to safeguard their contents. For user identity verification, the system compares a calculated hash 2 value against the original hash 1 value. Once verification is complete, the data can be decrypted and restored to their original state, granting access only to authorized users. Additionally, to assess the framework’s effectiveness, the performance metrics of the proposed model are compared to a baseline operational copy, with a focus on image privacy. Securing image data within network systems poses significant challenges due to the limitations of traditional embedding systems, which often result in suboptimal performance and increased risks of unauthorized access and data misuse. The existing approaches for image security encountered issues, such as prolonged execution times, further hindering their effectiveness. The present framework named Graph Convolutional-Based Cryptographic (GCBC) Framework addressed the above mentioned issues.

Specification

Description:Field of the Invention
The present invention relates to the technical field of machine learning, more particular to a Graph Convolutional-Based Cryptographic (GCBC) framework for enhanced security and efficiency in healthcare image networks.
Background of the Invention
The invention of the Graph Convolutional-Based Cryptographic (GCBC) Framework framework addresses key challenges in securing medical images within healthcare networks. Traditional security systems for medical data encryption often underperformed due to the reliance on embedding schemes that struggled with high computational times, poor performance on large data sets, and limited payload capacity. Additionally, these traditional methods were prone to security vulnerabilities, risking unauthorized data access, and were less suited to healthcare-specific needs, where efficient processing and high data integrity are crucial.
Previous research has explored diverse approaches for healthcare data security, each with unique advantages and limitations:
1. Ghanem, M.C.; Chen, T.M.; Nepomuceno, E.G. "Hierarchical reinforcement learning for efficient and effective automated penetration testing of large networks." Journal of Intelligent Information Systems 60 (2022): 281-303.
This study explores hierarchical reinforcement learning applied to cybersecurity, partitioning networks into security clusters to handle large, complex networks.
2. Preetha, A.D.; Kumar, T.S.P. "Securing IoT-based healthcare systems from counterfeit medicine penetration using Blockchain." Applied Nanoscale Science 13 (2021): 1263-1275.
This research investigates blockchain-enabled frameworks to prevent unauthorized access in healthcare by using smart contracts for data exchange. The model improves decryption efficiency but faces performance challenges with increased network demand.
Alabbad, M.; Mhaskar, N.; Khedri, R. "Hardening of network segmentation using automated referential penetration testing." Journal of Network and Computer Applications 224 (2024): 103851.
This study presents a framework for network security using referential penetration testing inspired by software engineering techniques. It evaluates network segmentation for secure data transfer, though it requires extensive resources, which can be a limitation for healthcare applications.
3. Hidayanto, B.C.; Akbar, I.A.; Putra, A.Z.D. "Automated Web Security Testing Guide Mapping to Accelerate Process on Penetration Testing." Procedia Computer Science 234 (2024): 1412-1419.
This work introduces a mapping tool to accelerate web security penetration testing by automating test case identification. The automated approach improves accuracy and speed but is focused on web security rather than healthcare-specific data protection.
4. Pandey, M.; Maurya, R.; Tripathi, S. "Blockchain technology for detecting and preventing the distribution of counterfeit medications." IEEE Transactions on Engineering 45 (2023): 87-96.
This paper discusses using blockchain to secure healthcare supply chains, specifically for tracking and authenticating medication distribution. While effective in reducing counterfeit drugs, the approach demands significant computational resources, which can be limiting for real-time applications.
5. Haq, T.U.; Shah, T.; Siddiqui, G.F.; Iqbal, M.Z.; Hameed, I.A.; Jamil, H. "Improved Twofish Algorithm: A Digital Image Enciphering Application." IEEE Access 9 (2021): 76518-76530.
This study enhances the Twofish encryption algorithm for digital image encryption, optimizing security for image-based data. Although effective for secure data storage, it is limited in its application to network-based healthcare image security, where real-time processing is essential.
6. Abu-Faraj, M.; Al-Hyari, A.; Al-Taharwa, I.; Al-Ahmad, B.; Alqadi, Z. "CASDC: A Cryptographically Secure Data System Based on Two Private Key Images." IEEE Access 10 (2022): 126304-126314.
This paper presents a cryptographic system using Twofish and private key images to secure data systems. The model demonstrates high security levels but is computationally intensive and unsuitable for healthcare network scenarios requiring rapid data encryption and decryption.
These studies highlight a range of existing methods and challenges in healthcare data security, such as high computational requirements, limited applicability to real-time needs, and vulnerability to complex cyber-attacks.
None of the prior art indicated above either alone or in combination with one another disclose what the present invention has disclosed.
The present GCBC framework, by integrating graph convolutional networks with Twofish encryption, aims to address these issues by improving efficiency and enhancing security specifically tailored to healthcare applications
Drawings
Figure 1. Difficulties with the traditional method.
Figure 2. Proposed GCBC architecture.
Figure 3. The process of the GCBC.
Figure 4. A) input image, B) encrypted image, and C) decrypted image.
Figure 5. Encryption time.
Figure 6. Decryption time.
Figure 7. MSE.
Figure 8. PSNR.
Figure 9. Error rate
Detailed Description of the Invention
The following description includes the preferred best mode of one embodiment of the present invention. It will be clear from this description of the invention that the invention is not limited to these illustrated embodiments but that the invention also includes a variety of modifications and embodiments thereto. Therefore, the present description should be seen as illustrative and not limiting. While the invention is susceptible to various modifications and alternative constructions, it should be understood, that there is no intention to limit the invention to the specific form disclosed, but, on the contrary, the invention is to cover all modifications, alternative constructions, and equivalents falling within the spirit and scope of the invention as defined in the claims.
In any embodiment described herein, the open-ended terms "comprising," "comprises," and the like (which are synonymous with "including," "having" and "characterized by") may be replaced by the respective partially closed phrases "consisting essentially of," consists essentially of," and the like or the respective closed phrases "consisting of," "consists of, the like. As used herein, the singular forms "a", "an", and "the" designate both the singular and the plural, unless expressly stated to designate the singular only.
The present Graph Convolutional-Based Cryptographic (GCBC) Framework introduces several unique features tailored for healthcare data security. It integrates graph convolutional neural networks with the Twofish encryption algorithm, enhancing efficiency and allowing rapid processing of large datasets.
The framework's dual-hash authentication mechanism (hash 1 and hash 2) adds an extra layer of security, ensuring user identity verification before data access.
With encryption and decryption times as low as 0.225 s and 0.231 s, respectively, it enables real-time data handling. The high PSNR value (24.7) ensures minimal quality loss for encrypted medical images, crucial for accurate diagnostics. Additionally, its low error rate (0.0021%) supports data integrity and reliability, while scalability across distributed environments makes it adaptable for diverse healthcare applications
The Graph Convolutional-Based Cryptographic (GCBC) Framework secures healthcare images through a multi-step methodology integrating a graph convolutional neural network (GCN) and the Twofish encryption algorithm.
• The process begins with data preprocessing to remove noise from medical images sourced from Kaggle. Then, a hash value (hash 1) is calculated to represent the initial state of the image data.
• Following this, images are encrypted using the Twofish algorithm with a generated key, transforming the images into an unreadable format for security.
• A second hash value (hash 2) is computed from the encrypted data for user identity verification. The user is granted access only if hash 1 matches hash 2.
• If authenticated, the system decrypts the images, restoring them to their original form for authorized users. This dual-hash approach ensures both data integrity and secure access.
The performance of the framework is as follows:
a) Low Encryption and Decryption Times: Achieved encryption and decryption times of 0.225 seconds and 0.231 seconds, respectively, allowing for fast data processing crucial for real-time applications.
b) High Image Quality (PSNR): The framework maintained a peak signal-to-noise ratio (PSNR) of 24.7, ensuring minimal quality loss for encrypted and decrypted medical images, essential for diagnostic accuracy.
c) Low Mean Squared Error (MSE): Achieved an MSE of 4.245, indicating high precision in data reconstruction and minimal loss of detail in image quality.
d) Low Error Rate: Demonstrated a low error rate of 0.0021%, supporting data integrity and ensuring accurate information transfer without corruption.
e) High Throughput: Attained a throughput rate of 71.225%, indicating efficient encryption, transmission, and decryption processes.
f) Robust Security through Dual-Hash Authentication: The dual-hash verification process (hash 1 and hash 2) successfully secured data access, allowing only authorized users to decrypt and access the original images
Table 1. Overall comparison.
Method Encryption Time Decryption Time Throughput MSE PSNR Confidentiality Score (%)
CASDC 509.66 s 1802 s 3237.2 6.7064 22.7171 77%
MC 0.3340 s 0.3340 s 5279.7 7.6702 21.2168 72%
GCBC 0.225 s 0.231 s 71.225 4.245 24.721 98%
Table 2. Performance statistics of GCBC.
Metrics Performance
Encryption time 0.225 s
Decryption time 0.231 s
MSE 4.245
PSNR 24.721
Confidentiality score 98%
Throughput 71.225%
Error 0.0021%

The Graph Convolutional-Based Cryptographic (GCBC) Framework presents several technical breakthroughs and advantages, especially in terms of efficiency, security, and cost-effectiveness:
1. Improved Efficiency in Encryption and Decryption: The GCBC model significantly reduces encryption and decryption times compared to traditional methods. For example, the encryption time was recorded at 0.225 seconds and decryption time at 0.231 seconds, both substantially faster than methods like CASDC (509.66 s for encryption and 1802 s for decryption). This efficiency is crucial for healthcare networks where data must be processed in real-time.
2. Enhanced Data Security through Dual-Hash Verification: By using two hash values (hash 1 and hash 2), the GCBC framework provides an added layer of security by verifying user identity before granting access to data. This verification step helps to prevent unauthorized access and data breaches, which are common challenges in healthcare data management.
3. High Image Quality Post-Encryption: The framework achieves a peak signal-to-noise ratio (PSNR) of 24.7, indicating minimal image degradation after encryption and decryption. This is a notable improvement over other models, ensuring that the quality of medical images remains high even after encryption, which is essential for accurate diagnostics and patient care.
4. Reduced Error Rates and Improved Data Integrity: The GCBC model demonstrates a low error rate of 0.0021%, which is an advantage in maintaining the integrity of sensitive healthcare data during transmission. This low error rate also enhances the framework's robustness against data corruption and unauthorized modifications.
5. Cost and Resource Efficiency: The streamlined processing times for encryption and decryption contribute to lower computational and resource costs. Additionally, the model's high throughput rate (71.225%) and reduced mean squared error (MSE) of 4.245 offer better performance metrics at a lower resource demand compared to traditional models like QITSA, QICSA, and QIGA.
Overall, the GCBC framework achieves surprising results by combining a graph convolutional approach with Twofish encryption, resulting in enhanced security, reduced processing times, and improved data quality, making it both effective and cost-efficient for healthcare network security.


, Claims:1. A Graph Convolutional-Based Cryptographic (GCBC) Framework, comprises of:
• data preprocessing to remove noise from medical images sourced from Kaggle;
• a hash value (hash 1) is calculated to represent the initial state of the image data;
• images are encrypted using the Twofish algorithm with a generated key, transforming the images into an unreadable format for security;
• A second hash value (hash 2) is computed from the encrypted data for user identity verification. The user is granted access only if hash 1 matches hash 2; and
• If authenticated, the system decrypts the images, restoring them to their original form for authorized users.
2. The Graph Convolutional-Based Cryptographic (GCBC) Framework as claimed in the claim 1, wherein framework incorporates medical data sourced from Kaggle into a secure architecture.
3. The Graph Convolutional-Based Cryptographic (GCBC) Framework as claimed in the claim 1, wherein framework achieved
• encryption and decryption times of 0.225 seconds and 0.231 seconds, respectively;
• a peak signal-to-noise ratio (PSNR) of 24.7;
• MSE of 4.245, indicating high precision in data reconstruction and minimal loss of detail in image quality;
• a low error rate of 0.0021%, supporting data integrity and ensuring accurate information transfer without corruption; and
• a throughput rate of 71.225%, indicating efficient encryption, transmission, and decryption processes.
4. The Graph Convolutional-Based Cryptographic (GCBC) Framework as claimed in the claim 1, wherein the dual-hash verification process (hash 1 and hash 2) successfully secured data access, allowing only authorized users to decrypt and access the original images.

Documents