Human-centric Computing and Information Sciences volume 12, Article number: 55 (2022)
Cite this article 3 Accesses
Distributed transactions in e-Healthcare and the evaluation of medical data have become an active research area of information technology that delivers medical records management and optimization without manually visualizing the computational loss. The increased use of e-Healthcare applications for availing medical services requires efficient computation during the processing of medical transactions and preservation through intelligent measurement analysis. Medical industries often involve and aim for the smooth application of medical transmission of demanding services. Thus, there are significant requirements for calculating loss during optimization and management in the distributed private network. In this paper, we contribute to two different objectives. First, we propose a machine learning-based stochastic gradient descent method for managing medical records and optimizing day-to-day transactions of e-Healthcare applications. This approach evaluates the loss of medical features during computation and enables optimized details of data transmission. Secondly, a blockchain-distributed E-Healthcare novel and a secure serverless architecture are proposed for the medical industry to protect transactions and preserve immutable storage. The simulation result shows the proposed system computations, such as loss = 0.7 (7%), learning-rate = goldilocks, ledger optimization =0.23 (23%), transmission power = -18 dBm, jitter = 32 ms, delay =90 ms, throughput = 170 bytes, duty-cycle and delivery = 0.10(10%), and calculate dynamic response.
Smart Contracts, Blockchain, Machine Learning (ML), Stochastic Gradient Descent (SGD), E-Healthcare, Information Management and Optimization
E-Healthcare applicational medical records management is the process of scheduling, sorting, examining, analyzing, and preserving patients’ sensitive information . It can help to track patients’ causes of diseases, establish efficient monitoring to improve treatment processes and establish effective manufacturing of medicines with accurate record prevention. The current strategy of information transactions, management, and delivery includes medical services requests and availing, services scheduling with cost-efficiency and documenting, recording patients’ emergency complaints, manual diagnosis, counseling, and corresponding treatment evaluated in health information for the medical industry
[2, 33]. Nowadays, the advancement in information technology makes the transmission of digital data electronically through e-Healthcare applications; and so, the big data of medical transactions evolves and needs dynamic, cost-efficient investigation and management problems come into existence . To manage efficient e-Healthcare transactions, the generation of electronic medical data needs efficient run-time processing, classification, and prediction of the futuristic rate of data emergence. Often, these investigational records are required to be shared among different healthcare sectors. It also includes the pharmaceutical and medical device industries, health insurance providers, researchers, pharmacists, and other stakeholders [5, 6]. These challenging aspects pose serious problems in handling day-to-day generated patients’ sensitive data.
The existing processes for capturing health-related data during processing are insecure. The lifecycle of medical data scheduling, processing, organizing, and managing requires consideration. Secondly, to investigate individual aspects of processed information before transmission over the network in terms of integrity, confidentiality, transparency, and provenance [7, 8]. And so, preserve each piece of investigated information in the server-based centralized storage or cloud environment. These investigated records are helpful in the diagnostic, treatment classification, medicine recommendation, and futuristic prediction of data generation and the required level of medical production.
However, the patients of e-Healthcare may receive transfer consulting from one hospital to another hospital during the treatment process . Patients have the right to define access control by stating what type of data will be shared and how long with whom [9, 10]. There is still an uncertain manner derived for stakeholders participating between two different channels, in which authentic users can get control access (read-only) to sensitive health records [11, 12]. Substantially, the process of exchanging the patients’ records between different channels creates a complex environment, such as platform interoperable related issues. With the new trend towards personalized healthcare for efficient treatment, diagnosis, and medicine production, existing e-Healthcare systems that utilize server-based centralized procedures are restricted when it comes to providing a cohesive view and protected shared access control to the medical history of the patient, diagnosis, and health-related transactions with participating stakeholders (authentic). In fact, the e-Healthcare data of the centralized-server approach is vulnerable in terms of alteration, redundancy, and tampering or forgery that can lead to data integrity, transparency, and provenance problems. In addition, all the people who are involved, including the patients, need to trust the central-service experts and be aware of their current security and protection procedures [13, 14].
The preserved e-Healthcare patient diagnostic and health-related records need to be utilized for futuristic predictive analysis, management, and optimization . Individual recorded medical entities are investigated in accordance with the rate of quality measurement of patients’ diseases and getting efficient treatments on time, the transmission of medical service deliveries, consultant prescription-related information, etc. For this purpose, machine learning (ML) is used to formulate predictive models. It starts with the features and labels of recorded health data to design a prediction function, as shown in Fig. 1. By training the complete model, there is a need to tune parameters at each stage of data selection and examination. The overall process of predictive futuristic health-related information evaluates the loss of data points in optimization and management, as shown in Fig. 1.
Blockchain e-Healthcare can strengthen security, provide serverless hash-based (SHA-256) encryption performance, and store health records and service delivery transactions on an immutable distributed storage. However, the event of medical node transactions can be stored in a gazette-like chain structure that is connected through different channels in a serverless private network. These business rules are managed and controlled via digital contracts (smart) to achieve distributed automated health service deliveries and emergency responses through the applications in the serverless environment. Many medical industries envisioned the purpose of achieving digital ledger integrity, traceability, provenance, and immutability to enable distributed medical preservation in immutable storage and analysis. The reason for the movement towards a decentralized environment is to protect medical ledgers against a variety of cyber-attacks usually intended for server-based centralized systems [17, 18]. This blockchain decentralized procedure of protected serverless network enables us to enhance the distributed medical nodes’ defense ability with hash-based re-encryption along with the blockchain Hyperledger intrusion detection mechanism and preservation. The design of customized blockchain consensus policies ensures the secure transmission and delivery of health services, as well as immutability, distributed trust, integrity, and transparency in the transactions of each node in the chain.
This paper describes the detailed design of the current process of e-Healthcare related medical transactions classification, scheduling, analysis, organization, management, optimization, and preservation with service delivery protocol-related problems and challenges as well as privacy and protection issues.
We designed and implemented the proposed architecture to examine and analyze preserved day-to-day patients’ medical transactions with the rate of data optimization and predict futuristic logs. The proposed collaborative architecture is efficient and effective when it comes to diagnosis, treatment criteria, data management, and optimization because it uses ML-enabled stochastic gradient descent (SGD).
It shows how to build a new and secure e-Healthcare application and management ledger with blockchain smart contracts and a secure serverless permissioned private network for the medical industry. Finally, the blockchain smart contracts (chaincodes) are designed, implemented, and deployed to automate day-to-day dynamic medical applicational data serverless transmission, preservation, update ledger, and exchange of analyzed futuristic logs (optimization details) among the participating stakeholders in the private network.
|Research methods||Research description||Issues/Challenges/Limitations||Similarity/Weakness|
|Hyperledger Fabric-machine learning (N-gram)-enabled drug-related information management and recommendation system for medical industries ||The authors of this paper proposed two different modules for secure medicine-related data management and provided a recommendation platform for consumers. The Hyperledger fabric is used to deploy a modular architecture for secure data transmission and preservation in the protected private network, whereas the machine learning-enabled N-gram LightGBM model is designed to provide efficient drug-related information to the participating stakeholders.||-A private permissioned network is designed||-Blockchain Hyperledger fabric|
|-Predefined consensus policies of the Hyperledger fabric are used||-N-gram LightGBM model|
|-Recommendation systems suggest only top-rated or highly utilized medicines||-REST API|
|The role of distributed e-Healthcare applications for secure health ledgers and maintaining the privacy of medical records using blockchain Hyperledger technology ||This paper indicated the privacy and security issues in the client-server and concerns about e-Healthcare data management, records, and preservation. This paper also highlighted a few potential challenges (such as regulatory compliance, etc.), and demonstrated the impact of distributed ledgers in the medical environment.||-Digital signature-based asymmetric cryptography||-Blockchain consortium network|
|-Proof of work, proof-of-stake, and delegated proof-of-stake are used||-Ethereum|
|-Predefined consensus policies are utilized||-Public channel for electronic medical ledger transaction|
|Blockchain-enabled e-Healthcare protected records preserved in secure distributed storage using KNN training protocols ||Jabarulla and Lee  proposed a secure technique of KNN for a privacy preservation solution. The KNN training over IoT data employs blockchain distributed ledger technology with a partially homomorphic cryptosystem in order to protect participating stakeholders.||-Secure biasing operations||-Rigorous analysis|
|-A secure comparison protocol is derived||-Homomorphic cryptosystem|
|-Secure polynomial operation is applied before data preservation||-Encrypted data sharing via blockchain|
|-Blockchain node scalability evaluation|
|Fitness health-related data management and optimization using the Internet of Things (IoT) enabled blockchain distributed platform ||Frikha et al.  proposed a process of capturing medical data, examining, preserving, analyzing, presenting, managing, and storing it through fitness intelligent devices, which were deployed in connection with the proposed IoT.||-Scope of data and privacy-related challenges||-Blockchain Ethereum|
|-Patient-centric application||-Public permissionless network|
|-For design consensus protocols, the Raspberry Pi 3 is used.||-Secure channel for transactions is deployed|
|Blockchain public permissionless network is designed for secure authentication between participating stakeholders of the e-Healthcare ledger environment ||The authors of this paper proposed a secure blockchain-based distributed architecture tailored specifically to cater to the needs of electronic healthcare applications.||-Platform interoperability issues||-Reliable cryptography with blockchain|
|-Computationally expensive||-Public network|
|-Pure decentralized in nature blockchain||-Predefine consensus protocol|
|Blockchain-enabled digital healthcare preserve system for fourth industry healthcare revolution applications ||Tanwar et al.  proposed an access control policy algorithm to enhance medical data accessibility between stakeholders. Further, assisting in the experimentation of distributed environments to create the Hyperledger fabric-enabled e-Healthcare records sharing system, which is deployed by the use of smart contracts.||-Cross-chaining limitations||-Hyperledger Fabric|
|-Round trip time evaluation.||-Private channel|
|-Streamline data formulation.||-Proof-of-authentication|
|-Distributed ledger preservation|
The e-Healthcare-enabled applicational records are employed by the medical industry using a server-based centralized infrastructure, in which different hospitals retain primary knowledge of patients’ clinical details [19, 20]. The patients’ clinical or health records received from different sources who get treatment from distinct consultants for different diseases are scattered or redundant in the different central servers (active or passive). To address the clinical records filtration, organization, management, and optimization-related challenges, various client-server-enabled centralized e-Healthcare applications of the distinct organization are proposed. However, in this scenario, security and protection are the main areas of concern in the central-server and cloud-enabled environment. Recently, several client-servers-enabled server-based e-Healthcare researchers presented solutions that address the security and privacy concerns. A few experts came up with different ideas, like storing clinical data in the cloud and recording encryption (hash-based) details in a group network. In this context, we examine and analyze several E-Healthcare systems’ data management procedures, evaluate loss functions while optimizing records and management, and tackle secure transmission protocols using blockchain ML models and related literature [20–23] (as shown in Table 1), which are discussed as follows.
Fig. 2 presents the proposed e-Healthcare simple architecture for medical industries, which provides minimal functionality to program events of health-related node transactions. This architecture is designed according to the blockchain permissioned network due to the benefits it has over the public infrastructure. The proposed blockchain-enabled e-Healthcare architecture consists of stakeholders, including end-users (patients), authorized members by the patient, consultant, emergency staff, hospital, health authority, and blockchain engineer. These stakeholders’ assets (read-only and write) request for medical services, medical test data, and other health-related transactions, for example, adding new records, updating records, and medical queries in the immutable storage. This designed platform involves different hospitals throughout the region and stakeholders connected to the single secure, permissioned private network. As illustrated in Fig. 2, the health authorities and other medical industries are participating in this ledger to facilitate quick access to and analysis of the futuristic assumptions. In each health-related transaction, there are two different portions highlighted: one is read-only and the other acts as a full node control (administrative environment). The proposed blockchain distributed architecture was designed, implemented, and deployed, and it supports four different types of transactions (as mentioned in contracts 1 and 2), such as AddMedicalLedger(), UpdateMedicalLedger(), SecureTransaction(), and NodeConPreservation(). To perform e-Healthcare transactions, the permission/approval of the blockchain engineer is required, as shown in Fig. 2 and Contract 1. Before every transaction in this ledger, stakeholders get authentication from the blockchain engineer, and then they send the new node transactions to the blockchain network. After the transaction execution, the blockchain engineer generates the node for the transaction and broadcasts it to other participating stakeholders (hospitals/medical industries). The blockchain consensus protocols are used to add new node transactions to the immutable ledger and to update it. For this purpose, we tuned practical byzantine fault tolerance (PBFT) in accordance with the e-Healthcare selection of protocol over the most commonly used proof-of-encrypted transaction because patients get less delay and faster response with reduced throughput. This procedure supports secure communication transmission and ledger maintenance. These day-to-day e-Healthcare transactions are stored in the protected ledger, which is immutable in nature and is used for the analysis of futuristic predictions related to big medical data management and optimization. For this reason, we used ML-based SGD to optimize the big data of the individual patient to release day-to-day storage before preserving it in the immutable system. InterPlanetary File Storage (IPFS) is used as a digital data storage infrastructure that provides a distributed storage environment (active/passive). However, to access this storage, we separate communication channels into off-chain and on-chain. As shown in Fig. 2, an off-chain communication channel is an outer channel to access the blockchain encrypted environment, while an on-chain is the inner one.
In this context, we present an extensive simulation and analysis of the proposed blockchain-enabled medical ledger optimization-related results and secure preservation solution in a distributed serverless network environment. It has been presented that there is a correlation between the day-to-day medical services-related transaction and optimization matrices, as shown in Fig. 6. The loss between the collaborative solution for medical industries is represented through the different parameters, such as a collection of preserved data (primary stored), medical service transmission between stakeholders, required transmission power, modulation level, delay, throughput, jitter, duty cycle, and rate of medical ledger optimization (using SGD) and preservation in the serverless private network, as shown in Fig. 6.
|Research explanation with
|Matrix of the state-of-the-art
|Evaluation matrix of our
scheduling and processing using
|This paper highlighted a few
main features that consume
less computational cost
with limited resources and
increase the rate of
accuracy in terms of
blockchain security. The
main features of this paper are:
allocation and utilization
-Blockchain public network
-Pre-defined hash encryption
-Multimedia-based data processing limitation
|The evaluation criteria of the
paper are discussed as follows:
-Network: Public network
-Consensus: Pre-define with PoW and PoE
-Node size: Default
Batch: Not applicable
-Optimization technique: Not applicable
-Accuracy: 86% success rate
|We proposed a blockchain and
serverless medical data
organizing, and preserving
distributed ledger solution for
medical industries to handle
day-to-day huge amounts of
health information. The
research parameters are
mentioned as follows:
-Blockchain: Permissioned structure
-Network: Private network
-Consensus: PBFT with PoET
-Encryption: Hash (SHA-256)
-Node size: 4 MB fixed size
-Batch: Single batch
-Optimization technique: SGD
-Accuracy: 93.7% (optimize average individual 23%)
This paper discusses the privacy and security-related challenges and limitations running on the traditional e-Healthcare applications along with the medical-ledger organization, management, and optimization limitations. For this purpose, in this paper, we proposed a blockchain ML-enabled collaborative architecture for medical industries to handle medical e-Service transmissions and ledger management. The collaborative approach brings ledger optimization, secure management, protection, integrity, forge-resistance, and control access to the health-related chain of information preservation using blockchain SGD. It also facilitates patients of e-Healthcare distributed applications to request, access, and allow authenticated stakeholders to record their electronic medical sensitive transactions through the e-Healthcare distributed application on the distributed ledger. The blockchain private permissioned platform is designed, implemented, and deployed as it provides a modular infrastructure, which also distinguishes the distributed core system (read/write) and read-only transactions from the applicational environment. For this reason, we designed and created digital (smart) contracts to enable the rules of e-Healthcare governance bodies and accomplish the blockchain consensus based on the predefined policies of PBFT. However, the core systems are evaluated in real-time for fault tolerance and manage the PoET engine to simulate trust events of medical transaction execution for attaining the chain-of-services. The simulation results of the proposed architecture show robustness in terms of efficient performance, including predictive loss = 7%, learning rate = goldilocks (0.5), ledger optimization = 23%, transmission power = -18 dBm, jitter = 32 ms, delay = 90 ms, throughput = 170 bytes, duty-cycle and delivery = 10%, and dynamic serverless responses.
Conceptualization, AAK, AAL, MS. Investigation and methodology, AAK, OC, MS, WA, HH, ZAH. Writing of the original draft, AAK. Writing of the review and editing, AAK, AAL, MS. Software, AAK, OC, MS, WA, HH, ZAH. Formal analysis, AAK, OC, MS, WA, HH, ZAH.
This work was supported by National Natural Science Foundation of China (Grant No. 62250410365) and Taif University Researchers Supporting (Project No. TURSP-2020/107), Taif University, Taif, Saudi Arabia.
The authors declare that they have no competing interests.
Name : Abdullah Ayub Khan
Affiliation : Department of Computer Science, Sindh Madressatul Islam University, Karachi, Sindh, Pakistan
Biography : Abdullah Ayub Khan is currently pursuing a Ph.D. degree with the Department of Computer Science, Sindh Madressatul Islam University Karachi. He has published around twenty research articles in well-reputed journals (such as IEEE Access, MDPI, Elsevier, Spring, Wiley) in the domain of digital forensics, cyber security, blockchain, hyperledger technology, and artificial intelligence.
Name : Asif Ali Laghari
Affiliation : Department of Computer Science, Sindh Madressatul Islam University, Karachi, Sindh, Pakistan
Biography : Asif Ali Laghari received the B.S. degree in Information Technology from the Quaid-e-Awam University of Engineering Science and Technology Nawabshah, Pakistan, in 2007 and Master degree in Information Technology from the QUEST Nawabshah Pakistan in 2014. From 2007 to 2008, he was a Lecturer in the Computer and Information Science Department, Digital Institute of Information Technology, Pakistan. In 2015, he joined the school of the Computer Science & Technology, Harbin Institute of Technology, where he was a PhD student. Currently he is Assistant professor in Sindh Madressatul Islam University, Karachi, Pakistan. He has published more than 60 technical articles in scientific journals and conference proceedings. His current research interests include Machine Learning, Computer networks, cloud computing, IoT, Fog computing and multimedia QoE management.
Name : Muhammad Shafiq
Affiliation : Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China
Biography : Muhammad Shafiq received the Ph.D. degree at the School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China, in 2018. He completed his Post doctorate in 2020 at Cyberspace Institute of Advanced Technology, Guangzhou University, Guangzhou, China. He is currently Distinguished Associate Professor at the Cyberspace Institute of Advance Technology, Guangzhou University, Guangzhou, China. He is Associate Editor at Wireless Communications & Mobile Computing Journal. He become IEEE Senior Member in 2021. He has guest editor in several Special Issues in reported SCI journals. He has invited speaker in well-known conferences. He is a member of several well-known conferences (CP, PC and TPC member). His current research areas of interest include IoT Security, IoT anomaly and intrusion traffic classification, IoT management, Network Traffic Classification, Cloud Computing and Network Security. He has authored more than over 50 peer-reviewed articles on topics related to cybersecurity. He has received multiple awards for Academic Excellence and University Contribution. He is the reviewer of many prominent journals, including, IEEE Transaction on Network Science and Engineering, Transactions on Multimedia Computing, Communications, and Applications, IEEE Transaction on Parallel and Distributed Systems, IEEE Transaction on Cloud Computing, IEEE Transaction on Service Computing, IEEE Transaction on Industrial Informatics, IEEE Internet of Things, IEEE Access, Concurrency and Computation: Practice and Experience (Wiley), Journal of Parallel and Distributing Computing. In 2020, he received Certificate of Appreciation from Journal Concurrency and Computation: Practice and Experience by Wiley.
Name : Omar Cheikhrouhou
Affiliation : CSE Laboratory, National School of Engineers of Sfax, University of Sfax, Sfax, Tunisia
Biography : Omar Cheikhrouhou is currently an Assistant Professor at Higher Institute of Computer Science of Mahdia, University of Monastir, Tunisia. He was an Assistant Professor at College of Computer and Information Technology, Taif, KSA. He is also a member of CES Lab (Computer and Embedded System), University of Sfax, National School of Engineers. Dr. Omar Cheikhrouhou has received his Ph.D. degrees in Computer Science from the National School of Engineers of Sfax in March 2012. His Ph.D. deals with security of Wireless Sensor Networks and more precisely in “Secure Group Communication in Wireless Sensor Networks”. Currently, his research interests span over several areas related to Wireless Sensor Networks, CyberSecurity, Edge Computing, Blockchain, Multi-Robot System Coordination, Smart and Secure Healthcare, etc. Dr. Omar has several publications in several high-quality international journals and conferences. He has received some awards, including the “Governor Prize” from the Governor of Sfax in 2005.
Name : Habib Hamam
Affiliation : Faculty of Engineering, Moncton University, Moncton, Canada
Biography : Habib Hamam obtained the B.Eng. and M.Sc. degrees in information processing from the Technical University of Munich, Germany 1988 and 1992, and the PhD degree in Physics and applications in telecommunications from Université de Rennes I conjointly with France Telecom Graduate School, France 1995. He also obtained a postdoctoral diploma, “Accreditation to Supervise Research in Signal Processing and Telecommunications”, from Université de Rennes I in 2004. He was for 10 years (2006-2016) a Canada Research Chair holder in “Optics in Information and Communication Technologies”. He is currently a full Professor in the Department of Electrical Engineering at Université de Moncton. He is OSA senior member, IEEE senior member and a registered professional engineer in New-Brunswick. He is among others editor in chief in CIT-Review and associate editor of the IEEE Canadian Review. His research interests are in optical telecommunications, Wireless Communications, diffraction, fiber components, RFID, information processing, data protection, COVID-19, and Deep learning.
Name : Wajdi Alhakami
Affiliation : Department of Information Technology, College of Computers and Information Technology, Taif University, Taif, Saudi Arabia
Biography : WAJDI ALHAKAMI received the B.Sc. degree in Computer Science, Saudi Arabia, in 2007. the M.Sc. degree in Computer Network, and the Ph.D. degree in Network Security from the University of Bedfordshire, United Kingdom in 2011 and 2016 respectively. He is currently an Associate Professor with department of Information Technology, College of Computers and Information Technology, Taif University, Taif, Saudi Arabia. His research interests include the Internet of Things, Cyber Security, and Computer Networking.
Abdullah Ayub Khan1,2, Asif Ali Laghari1, Muhammad Shafiq3,*, Omar Cheikhrouhou4, Wajdi Alhakami5, Habib Hamam6, and Zaffar Ahmed Shaikh2, Healthcare Ledger Management: A Blockchain and Machine Learning-Enabled Novel and Secure Architecture for Medical Industry, Article number: 12:55 (2022) Cite this article 3 AccessesDownload citation
Anyone you share the following link with will be able to read this content:
Provided by the Springer Nature SharedIt content-sharing initiative