Secure blockchain enabled Cyber–physical systems in healthcare using deep belief network with ResNet model

Highlights

• Secure intrusion detection with blockchain in the healthcare sector is proposed.

• It involves a series of cyber–physical system and deep learning processes.

• The presented model takes place on NSL-KDD 2015, CIDDS-001, and ISIC dataset.

• Presented DBN model has achieved a detection rate of 98.95% and 98.94% on the applied datasets.

• Effective classification performance with ResNet 101 model achieving maximum sensitivity.

Abstract

Cyber–physical system (CPS) is the incorporation of physical processes with processing and data transmission. Cybersecurity is a primary and challenging issue in healthcare due to the legal and ethical perspective of the patient’s medical data. Therefore, the design of CPS model for healthcare applications requires special attention for ensuring data security. To resolve this issue, this paper proposes a secure intrusion, detection with blockchain based data transmission with classification model for CPS in healthcare sector. The presented model performs data acquisition process using sensor devices and intrusion detection takes place using deep belief network (DBN) model. In addition, the presented model uses a multiple share creation (MSC) model for the generation of multiple shares of the captured image, and thereby achieves privacy and security. Besides, the blockchain technology is applied for secure data transmission to the cloud server, which executes the residual network (ResNet) based classification model to identify the presence of the disease. The experimental validation of the presented model takes place using NSL-KDD 2015, CIDDS-001 and ISIC dataset. The simulation outcome pointed out the effective outcome of the presented model.

Introduction

Recently, Cyber–physical system (CPS) has gained maximum attention which has been assumed as an advanced model. It is the integration of processing and communication abilities globally. CPS is meant to be basic research that has been carried out by the US National Science Foundation (NSF) and preferred by US President’s Council of Advisors on Science and Technology. CPS depends upon sensing, computation, and networking. The current innovations of Wireless Sensor Networks (WSN), medical sensors, and Cloud Computing (CC) tend to make CPS an effective module for clinical applications such as healthcare and home patient care [15]. Such advances are applicable for observing the patient health status in remote site and provide better treatments. A manageable study has been processed on clinical sensors. The sensors are used for collecting the patient details which is highly confidential. The accumulated data is passed to access through the wireless communication medium. Also, wired sensors are applied; hence, WSN offers elasticity and makes it comfortable for physicians and patients. The data gathered by sensors are saved in a server and it can be accessed by physicians [1].

Security is the most significant one where the patient’s details are highly confidential from both legal and ethical sides [6]. Thus, at the time of developing CPS structure medical domains, maxi-mum attentions have to be provided for data security [13], [14], [19]. Database management models have to be effective and scalable. The clinical data is applicable to offer appropriate insight into treatments which is highly essential to save the user’s life, in which the data is available and accessible to legal medical officer whenever required. Additionally, medical applications need maximum computing resources for making smart decisions on the basis of huge patient’s data. Hence, the data which is collected by WSN is highly experienced with limited energy, computing, and memory storage.

The CC model is applicable to offer effective solutions for the problems involved in this study. The recent studies are endeavoring to apply state-of-the-art CC models for CPS that enhances the reliability of the network and activate actual data analysis. Furthermore, CC is said to be a Computing Infrastructure which is applied by a consumer organization whenever required. It offers the service like processing, saved, networking, and software “as a service”. [6] presented CC in the following: “cloud is a parallel and distributed computing (DC) is composed of interlinked as well as virtualized systems are supervised dynamically and projected as diverse computing resources based on the service-level agreements (SLA) that are implemented with a negotiation among service giver and consumers”. Hence, as per the name, it is considered that the cloud is composed of given as follows: (1) self-service, (2) per-usage metering as well as billing, (3) flexibility, and (4) customization.

The 2 factors of CC are highly applicable to end users. Initially, individuals are applying personal information with the help of system with Internet connection. Secondly, software applications are not essential for user’s PC and accessible on the cloud. Additionally, CC offers different software services, APIs, and development devices to researchers. This service activates the users to transform the processing structure to the cloud. The CC is a result of recent models like DC, Internet models such as service based structural design, web services, scheme organization which is autonomic computing, and hardware such as virtualization, multicore chips. The integrated sensor–cloud structure is an interior portion of CPS, in which the cloud maintains the cyber actions and the sensors maintain the external events. A progressive study on sensor–cloud is defined in [2]. Since the CPS method is novel, the problems are resolved effectively. Concurrent computation, data query, memory, and security are some of the problems.

In CPS, the concatenation of active user input like smart feedback scheme, electronic details of patient information, and passive user input like biosensors and smart tools in clinical platforms supports the data acquisition for making effective decisions. The integration of data acquisition and decision making scheme has been explored crucially for medical applications and these integration gains maximum attention. The options of applying CPS in medical sector encloses the establishment of coordinated interoperation of independent and applicable tools, also the novel approaches to manage and operate healthcare physical systems under the application of calculation and manage, reduced implantable smart tools, body area networks, systematic tools, and novel fabrication concepts.

Though massive CPS structures are existed, only less CPS architectures are available for medical domains. Hu et al. [11] have presented a service-oriented architecture (SOA) relied medical CPS method; hence, it suffers from incomplete structure. Wang et al. [26] implied a trusted CPS structural design for medical sector that applies WSN-CC combined approach. Banerjee et al. [5] applied a modeling and investigation of medical CPS. Thus, it experiences the issues of security and privacy. Rajkumar [18] stated that the challenges of CPS can be resolved in future where the outcome in a science of CPS as well as technological solutions can be provided. Similarly, real-time, affordable, and consistent CPS can be placed in massive applications. In Rajhans et al. [17] deployed an abstraction and verification of this strategy, and a structural approach for CPS which has been projected by applying structural as well as semantic mappings for assuring reliability and activate scheme-level validation. In Vamvoudakis et al. [23] introduced a game-based method for evaluating binary arbitrary parameter according to the sensor values without any corruption. The CPS management can be processed, using wireless manage and event-based control, Demirel et al. [8] carried out a study in wireless control loops along with sensor readings that are unwanted and energy-based multihop wireless network. Finally, a modular design model has been deployed and optimized the data transmission strategies as well as managing rules.

In [21], the event-oriented state evaluation issues are known in which sensors are stimulated by a fusion model when it is a better prediction value and should not exceed the applied threshold. Also, the cases of CPS are identified. Then, CC is extensively employed recently. Thus, it can accomplish with best outcomes. Baliga et al. [4] offered a comprehensive power application diagnosis for CC. It assumes the public and private clouds with power utilization by switching, transmission, data computation as well as data storage. Warneke and Kao [27] applied dynamic resource assignment to effective parallel information procedure in CC. Khazaei et al. [12] examined the pool management method to CC centers. In [12], network virtualization as well as software-based networking for CC has been consolidated.

Xiao et al. [28] used an attribute-based approach for the study of security and privacy issues in CC. Also, the well-known CC environments like Amazon, Google, and IBM, has been applied in diverse applications like smart transportation systems, developing organization. In [25], a context-aware vehicular CPS with cloud support has been projected and 2 important service elements like vehicular social networks as well as context-aware vehicular security have been deployed. In Wan et al. [24] provided a novel structure named as VCMIA by combining vehicular CPS with mobile CC, and offers portable services for effective users like drivers and passengers to use the mobile traffic clouds. A new RGB based share creation model using elliptic curve cryptography (ECC) method has been presented in [20] to attain privacy and safety. The presented model generates a set of shares for an individual image which further undergoes encryption and decryption using ECC [3].

To resolve cybersecurity issue in CPS, this paper introduces secure intrusion detection with blockchain based data transmission with classification model for CPS in healthcare sector [9]. The presented model comprises different stages namely data acquisition, intrusion detection, data transmission and classification. In the beginning, the medical data is acquired by the use of sensor devices. Next, deep belief network (DBN) model based intrusion detection process takes place to identify the anomaly data in the gathered data. Simultaneously, multiple share creation (MSC) model is applied, which creates a set of shares of the captured image. Then, the blockchain technology is employed for secure data transmission to the cloud server, which executes the residual network (ResNet) based classification model for the identification of the existence of the disease. The experimental validation of the presented model takes place using skin lesion dataset and the simulation outcome pointed out the effective outcome of the presented model. In short, the paper contributions are summarized as follows.

• Propose a new secure intrusion detection with blockchain based data transmission with classification model for CPS in healthcare sector

• Involves several processes namely data acquisition, intrusion detection, encryption, data transmission and classification.

• Employ DBN based intrusion detection process for the identification of anomalies

• Introduce MSE with blockchain technology for secured data transmission

• Finally, ResNet based classification model is applied for disease diagnosis