Description:Field of the Invention
Health information technology (HIT) enables health care companies to take use of ICT breakthroughs to more effectively manage patients' care by using computerized systems. HIT enables the safe utilize and exchange of pertinent healthcare data, which enhances the ability to make high-quality healthcare decisions. The Blockchain technology is seen by some in the IT industry as the fourth industrial revolution, it is inevitable that it will have a significant effect on the world. People believe that blockchain is the ideal technology for the healthcare industry since it can solve challenging and complex problems that exist within the current health system.
Objective of the Invention
The FL training mechanism safeguards patient confidentiality by confining sensitive information to regional organizations and this is achieved through the model's client-server architecture. The creation of a trust model based on blockchain technology and the introduction of a novel consensus approach to maintaining FL nodes, this restriction on relying solely on the protocol will be lifted.
Background of the Invention
Health information technology (HIT) enables health care companies to take use of ICT breakthroughs to more effectively manage patients' care by using computerized systems. HIT enables the safe utilize and exchange of pertinent healthcare data, which enhances the ability to make high-quality healthcare decisions. Patient's information which is related to healthcare, including a catalogue of their medical conditions, lab results, doctor's comments, and radiological results, is contained in their electronic medical records (EMR). A patient may have several electronic medical records (EMR) from the many different hospitals or various doctors clinics. The authenticity that the EHRs comprise patient health records from several healthcare providers sets it apart from the EMR. Greater patient participation, improved efficiency, improved patient healthcare and cost deduction, higher standard of treatment and results, and progressive care coordination are just a few of the advantages of HER.
Blockchain technology is seen by some in the IT industry as the fourth industrial revolution, it is inevitable that it will have a significant effect on the world. People believe that blockchain is the ideal technology for the healthcare industry since it can solve challenging and complex problems that exist within the current health system. There has been some indication of an interest on the part of healthcare practitioners in utilizing blockchain technology. A report on the current state of blockchain technology is compiled annually by Deloitte and published under that company's name. In the 2019 survey, which included over a thousand senior executives, respondents' positive attitudes toward blockchain technology have improved in relation to feasibility, scalability, moving forward with use cases and collaboration with different participants when compared to the same survey conducted in 2018. Comparing 2019 to previous year, executives had a more positive outlook on the organization's ability to overcome problems and address concerns (EP3399501B1). The application of blockchain technology in the medical field holds significant potential to revolutionize healthcare. There has been a boost in the proportion of people interested in putting money into blockchain technology businesses. Blockchain technology enables an improved privacy, higher level of transparency, secured data integrity, improved data sharing, one version of the truth, increased efficiency with cost reduction.
Within the framework of the healthcare industry, this section explores the potential benefits offered by blockchain technology.In traditional, centralized systems via the multiple copies of the similar data as backups in physical servers ensures that a replicated version of the data is always available within the centralized system . In a blockchain, there is only ever going to be a single representation of the truth that is shared and replicated across the network. The user is able to redownload the data via the blockchain network in the event that their own system has been compromised. It has been hypothesized that blockchain, due to the fact that it functions as a decentralized database, is capable of significantly cutting costs while also significantly improving efficiency. The data that you store on a blockchain can be protected from a variety of privacy risks, including those related to data transparency, access control, data ownership and auditability. The use of blockchain technology results in increased operational efficiency. Blockchain have capacity to enable distributed record-keeping and immutability which is certified by community consensus makes it possible to achieve work efficiently (CN111970129B). Data silos are created when companies maintain independent electronic copies of records that are kept in a centralized database. Because human reconciliation is frequently required, data silos are created. Blockchain technology provides a solution to the problem of data silos and encourages effective data sharing.
The existing healthcare system has a problem with the incorrect information or lack of information that is required while the patient treated, thus maintaining the integrity of the data that is collected is extremely important. Unauthorized access via unwanted parties over personal health information is significantly reduced by using blockchain technology. Transactions that cannot be changed contribute to maintaining the integrity of data. Increased confidentiality of information across the network can be achieved through the use of patient data encryption. Existing healthcare information systems put patient data at danger of being stolen or compromised, and there is a significant likelihood that these systems may fail. Furthermore, data corruption is a common occurrence. Because they are concerned about the safety and confidentiality of their medical data, some patients choose not to share certain information with their healthcare professionals. As a result, data protection is at the very top of the list of advantages offered by blockchain technology.
Summary of the Invention
Federated learning is a game-changer for machine learning because it allows for greater scalability, security, and privacy. Secure data cooperation for its cloud computing ecosystem was further attained by the scheme replicated relying cryptographic cloud environment. The FL training mechanism safeguards patient confidentiality by confining sensitive information to regional organizations and this is achieved through the model's client-server architecture. Also, it's easy to implement, extensible and compatible with other systems. Through effective broker management, cloud computing service providers increase the benefits for healthcare users and a decentralised network that is both scalable and powered by Blockchain technology makes it possible for potentially unreliable nodes to communicate with one another. The success of Blockchain and FL technology is essential for end-device privacy. However, with the creation of a trust model based on blockchain technology and the introduction of a novel consensus approach to maintaining FL nodes, this restriction on relying solely on the protocol will be lifted.
Brief Description of Drawings
Figure 1: Federated Learning Architecture for Healthcare
Figure 2: FL - Cloud based Blockchain Strategy system architecture
Detailed Description of the Invention
Connecting various items through the Internet of Things makes it possible can gather data that could be utilized to improve productivity, effectiveness, and human health. Using an IoT-based observing network for patient health is one of the most promising new ways to close the global health gender equity gap. These IoT technologies are also known as IoMT. People can acquire information on their lives, carnal and psychological efficacy, and living circumstances, between additional belongings, by linking their existences toward the connection. This enables medical professionals toward remotely and continuously checks on patients' health. Transferring patients from their homes to hospitals for standard checkups is very challenging in today's environment. Queue up, transportable period, and the risks of peoples catching different infections though passing over this dirty location are just a few of the complications. As a result, the healthcare business is concentrating on in-home healthcare services, which allow patients to undergo screening inside the privacy of their own homes. A smart health monitoring system is intended to assist patients who live in distant places in contacting doctors in urban areas.
The advancement of using blockchain technology opens up fresh avenues for addressing important privacy, security, and ethical concerns in a smart healthcare system. Blockchain technology provides an open and responsible platform for data protection. The implementation of systems based on the deployment of block chain and integrated "cloud-like" PC set-ups can solve these problems. FL is a distributed training the machinery framework along with IoT that allows several strategies toward cooperatively develop training the machinery prototypes deprived of actually transferring any real records. FL's architecture is seen in Figure 1. This improves the intelligent healthcare system by preventing patient information from leaking. Figure 1 depicts a FL-based hospital in which fixed instruments collect health evidence since healthcare sources, numerous control strategies cooperate scheduled FL algorithms, while machines culture methods evaluate the people’s comfort and, uncertainty necessary, pursue urgent cloud-based support. Due to the tremendous assurance, it offers for analysing fragmented sensitive material, federated learning is a newly popular paradigm. So instead merging facts since different fonts or using the outdated "find and copy" method, it lets you train a consistent international prototype on a central server while keeping the data in the organisations that need it. Individual sites can collaborate to train a global model using this technique, called federated learning. FL is termed as the technique of combining preparation facts since several causes towards create a universal prototypical deprived of right swapping datasets.
Comprehensive and precise diagnosis, remote telemonitoring & detection, remote surgery, and intelligent treatment, which includes online training for anxiety sufferers are all concepts covered under intelligence in healthcare 5.0. The term "artificial intelligence" (AI) refers to a broad category of technological advances in intelligence. It denotes to the capacity of ML techniques toward forecast outcomes deprived of involving humans. The rapid growth and improvement of technology in health care has led to the creation of the term "smart health system." Many solutions have been put up for application in a variety of productions, comprising intelligent homes, the Industrial Internet of Things (IIoT), and intelligent medical filed, in light of these characteristics. However, as privacy risks become more sophisticated, a number of challenges remain when using blockchain-based FL in healthcare: The parameters of the model that are kept in the blockchain can still be used by adversaries to figure out the original confidential clinical data. Certain clinical data from medical devices might be fabricated in order to trick the FL process. Medical equipment is not compelled to offer facts and dispensation authority toward FL.
Towards overcome the above-mentioned issues, this training combines FL with advanced encryption towards deliver a secure and privacy-preserving healthcare 5.0 system. The most important contributions made by this study are as follows: We introduce a Hy-FL based blockchain with 5th generation healthcare that not only builds an accurate collaborative model based on multiple edge devices but also controls the entire training process. They present a reward selection-based technique, which helps to equalize privacy and data accuracy metrics in order to alter the noise involved in the training phase. Several medical organizations are taken into account with in conceptual scheme, which makes the model trained locally models of other medical organizations may improve system with the functionality of the healthcare system for global data being shared in the model. Toward deploy FL in the field of medical field, the training method gets enhanced for medical facts by nearby teaching prototypes. Near offer a clever mixture strategy that will improve safe message then efficient observing the health of the persons. Create an Intrusion Detection System (IDS) in the field of medical that improves safety and confidentiality by identifying interruptions and occurrence behaviours.
The proposed Hy-Fl system based blockchain architecture is based on a cloud-based platform with IoT and Blockchain integration. The architecture is divided into Sensor Networks and private data center's that store the data produced by Internet of Things devices. The data will move to the data monitoring and other medical devices after going through some data validation and processing steps in the cloud network. Whenever a given service is begun, the cloud server processes & transmits all data needed according to the monitoring networks, including such temperature, pulse, ECG, sensor, etc. The specific parameters, such as CPU, computing power, GPU, etc., are demanded by the services and are chosen depending on demand. The design connects the cloud server and information to the blockchain, whose primary function is to establish secure connection and efficiently process data. These blockchain data are kept in a way that ensures data authentication by collecting and tying together data from various sources. When processing data, the user applies the targeted data, which is then processed to produce the desired results. The secure communication is established based on the data's permissions and controls, and the data is labelled with its quality level after being trained and processed. Next, the labelled trained data is sent to the FL model. In FL, the data results are collected from the data center's data users. Then, using a data analyzer prediction process based on the recent and historical data lines, computational logic is applied. This prediction process works to enhance system outcomes. In FL, the data results are collected from the data center's data users. Then, using a data analyzer prediction process based on the recent and historical data lines, computational logic is applied. This prediction process works to enhance system outcomes.
The following are the main components of the system: The user interface displayed various Internet of Things (IoT) devices, as well as a blockchain cloud network where data is validated, processed, and distributed to a third party, such as monitors, MRI machines, smartphones, heart rate and other phenomenon measuring gadgets, and so on. Body data such as temperature, heart rate, blood pressure, electrocardiogram, and many other types of data were displayed on the user interface. The rest web service provides a simple interface for sensors to communicate with the cloud and retrieve the data they require. The cloud server provides high-powered machines with plenty of CPU, computation power, RAM, GPU, and network bandwidth to allow data to be managed quickly and globally via a variety of interfaces (such as PCs, TVs, and smartphones). To make sure that computation and communication are safe, a Blockchain cloud layer sits on top of the control layer's public data. In order to collect data and work on it together, blockchain technology keeps track of how data is used and makes sure it is still valid. When a data user applies it, our system takes care of collecting the necessary data, processing it, and returning the results. As data ownership can be separated and permissions can be granted, data rotation can be made manageable and safe. A model's ability to perform well when labels are applied to training data is heavily influenced by the calibre of the training data. In order for the FL model to learn and compile the raw data, it must be labelled with the appropriate identifiers. The three label assignment techniques are as follows: Adding data items to a dataset with labels already assigned, assigning labels to data items using a console and adding labels to data items manually. Uniform information acquisition with FL and multi-lateral computing is completed by using the universal data centre. While user data is securely stored in a private data centre, FL ensures confidentiality by using data calculations rather than raw data. Through advanced analysis and real-world applications, computational logic makes it easier to study linguistic tensions. The data analyst also looks at both recent and historical healthcare data to forecast patterns and lines of data and improve outreach, as shown in Figure 3. , Claims:The scope of the invention is defined by the following claims:

Claim:
1. The System/Method for Enhancing Healthcare services using Federated Learning and Blockchain Technology comprising the steps of:
a) A method is designed to build accurate collaborative model based on multiple edge devices
b) A method is designed to equalize privacy and data accuracy metrics in order to alter the noise involved in the training phase.
c A method is designed in the field of medical that improves safety and confidentiality.
2. The System/Method for Enhancing Healthcare services using Federated Learning and Blockchain Technology as claimed in claim1, led to the design of Hy-FL based blockchain with 5th generation healthcare system.
3. The System/Method for Enhancing Healthcare services using Federated Learning and Blockchain Technology as claimed in claim1, reward selection-based technique is used for equalizing privacy and data accuracy.
4. The System/Method for Enhancing Healthcare services using Federated Learning and Blockchain Technology as claimed in claim1, an Intrusion Detection System (IDS) is created in the field of medical that improves safety and confidentiality by identifying interruptions and occurrence behaviours.