Technological advancements have made a modern method to improve human life quality possible [1]. The IoT is an innovative and developing paradigm gaining interest in several application sectors, including smart homes, smart environments, and personal and remote healthcare [2]. Research and technology researchers have identified and evaluated health data sources to learn more and solve health-related challenges [3]. Therefore, creating integrated healthcare technology can boost productivity and increase patient outcomes at every level of the medical system [4]. The world’s largest and fastest-growing industry is the healthcare sector. How healthcare is managed has changed over the past several years from a disease-centered approach to a patient-centered one [5] and a volume-based approach to a value-based strategy of healthcare delivery. The growing drive for patient-centered treatment and value-based healthcare delivery models is guided by the goals of raising public awareness of the excellence of healthcare and reducing costs [6].

By utilizing strong patient safety controls, widespread access to data, remote inpatient monitoring, quick intervention strategies, and decentralized electronic medical records, the creation of new IoT-based healthcare software applications can address some issues associated with conventional healthcare systems [7]. A system created to handle healthcare data is referred to as a medical information system (MIS). This includes the practical administration of a hospital or a system that supports the formulation of healthcare policy, as well as systems that gather, store, handle, and send a patient’s electronic medical record (EMR) [8]. These techniques can increase the quality of life for patients, boost cooperation, boost patient outcomes, lower costs, and boost the overall efficiency of e-healthcare services [9]. Systems that manage data linked to the operations of providers and healthcare groups are also included in the category of health information systems. These could be used in concert to impact research, better patient results, and improve policy and decision making. Because expenditures in extensive data analysis can be significant and create a demand for effective, affordable technology, using the cloud to study big data in healthcare stands to reason [10]. Security is a top priority because medical information systems frequently view, handle, or keep huge amounts of sensitive data. Since the equipment is typically attached to an internal network that is linked to the Internet, it is also susceptible to viruses from devices and other equipment carried into hospitals. Different kinds of malicious attacks can be caused by the attackers [11].

A form of malicious software known as ransomware stops you from reaching your device’s information, infrastructure, or networks and requests a ransom in exchange for their release. These assaults, which have been linked to issues with medical processes, disrupted patient treatment, according to more than half of ransomware victims. The probability of returning to care redirection following an assault was the greatest impact noted. If a hacker seizes control, they can instruct devices to provide false readings, deliver medication drug overdoses, or take other actions that jeopardize the health of patients [12]. Due to the substantial quantity of confidential data that healthcare organizations keep for patient treatment and activities, the sector is seen as a target-rich environment. Consequently, cybercriminals have shifted their focus from the banking industry and retail shops to healthcare facilities due to personal health information potential value being up to 50 times greater than finance data; it can be valuable to attackers. Hospitals are important infrastructure companies that keep, exchange, and use a lot of private information. To provide patients with vital medical treatment, healthcare centers also rely on a number of IoT devices and electronic medical records [13]. This particular combo appeals to cybercriminals as a prize deserving of a hefty ransom that will be rapidly paid. However, becoming a primary target involves more than just motivation and pressure. Hospitals are an excellent target for devastating malware assaults as a result of a number of current occurrences that have combined to create the perfect storm [14].

Identity theft is a major problem for cybercriminals, as it can lead to the theft of personal information such as insurance, names, policy numbers, birth dates, billing data, diagnosis codes, and bank and credit card information [15]. Fraudsters use data from healthcare organizations to create fake IDs, resell medical equipment, and file made-up claims with insurers. Many users are unaware that they have been compromised, leading to unexpected consequences and rampant medical card theft. Medical identity theft is the act of someone using confidential information such as a social security number, without permission, to make false claims to Medicare and other health insurers, which can waste government money and interfere with medical treatment [16]. These identity thefts are correlated to criminal forgery theft. The use of tools, procedures, and measures to defend against cyberattacks on networks, applications, gadgets, systems, and data is known as cyber security. Its objectives are to lower the danger of cyberattacks and safeguard against the unauthorized use of innovations, networks, and platforms [17].

Early in the COVID-19 pandemic, it was unclear how healthcare costs and use would alter globally. Although a pandemic may lead to higher health expenses, spending and use declined [18] due to other considerations. The cost of combating fraud and upholding rules is a factor. Expensive antivirus software must be obtained to shield private patient data from hackers [19]. Due to this, healthcare costs must increase to maintain patient and data security. AI and machine learning have revolutionized healthcare, particularly in medical specialties. The medical disciplines make significant use of computer systems with artificial intelligence, such as remote patient treatment, prescription transcription, enhancing doctor–patient contact, drug research and development from beginning to finish, and patient diagnosis [20]. Modern computer algorithms have recently attained accuracy levels that are comparable to those of human specialists in the field of medical sciences, despite the fact that computer systems frequently perform jobs more quickly than humans do. The goal of separating rhetoric from reality is discussing how AI is reshaping the field of medicine. AI can help healthcare organizations cut costs by deploying more sophisticated technology that is more accurate and well-suited to carry out particular functions [21]. Ensuring that the appropriate care and support are adequately suited to their health objectives might lower the number of necessary diagnostic tests and the readmission rate. It can help physicians by automatically identifying potential issues and alerting medical staff [22]. Additionally, they would lessen the likelihood of misdiagnoses and medical malpractice claims, which can add to costs.

AI applications can deal with the enormous amounts of data generated in the medical field and discover valuable knowledge that would otherwise be hidden in big medical data. Healthcare stakeholders may use AI-based computational tools to harness the power of data to review historical data, anticipate prospective outcomes, and identify the optimal actions for the current context. As a result, AI is becoming more essential to healthcare stakeholders in decision-making [23]. When putting privacy protection measures in place inside a specific system, this service represents a possible privacy breach that must be considered. End users are now more concerned than ever with the privacy of their health data due to increasing awareness among them [24]. New types of cyber-attack will be made possible by advances in AI. These attacks may use AI systems to do specific tasks more effectively than humans could or exploit flaws in AI systems that humans cannot control.

Additionally, AI systems regulate elements of malware and robot behavior that are impossible for humans to hold [25] manually. In the past, several security measures were put out to protect the transmission of patient data to hospitals ^{[26][27][28][29]}[26,27,28,29]. However, the high cost and lengthy process prevent the best option from being implemented. Therefore, this research proposes a new cost-effective security algorithm for an intelligent hospital management system for COVID-19 data transmission.

As new devices proliferate, they often integrate the Internet of Things (IoT), generating and exchanging a massive quantity of data in the process. As a result, providing protection in an IoT setting is more difficult than expected. Properties such as secrecy, integrity, authorization, privacy, permission, and availability must all be ensured in order to ensure security in the IoT. Following is a summary of specific recent articles related to this research: Thilagam, K. et al. [30] offered IoT-based deep learning techniques based on privacy protection and a data analytics system. The health-related data are examined in the cloud using a convolutional neural network (CNN), free of user privacy data. As a result, a safe access control component is introduced for the IoT–Healthcare system based on user attributes. Furthermore, a higher user count enables an accuracy of about 98%. Experimental research shows that the suggested solution is reliable and efficient in terms of little privacy leakage and good data integrity.

Ali, Aitizaz et al. [31] created a novel deep-learning strategy-based secure searchable blockchain that functions as a distributed database and uses homomorphic encryption to allow users to access data safely via search. Using an IoT dataset, this study evaluated and compared the recommended access control mechanisms to reference models. The hyper ledger tool’s smart contracts implement the suggested algorithms.

Deep learning (DL) methods were combined with authorized blockchain and intelligent contracts by Kumar, Randhir et al. [32] to create the unique, safe, and effective data-sharing model PBDL. To be more precise, PBDL has a blockchain-based system to register, authenticate (using zero-knowledge evidence), and verify the communicating parties before employing an innovative contract-based agreement method. The healthcare data are encoded or transformed into a new format using stacked sparse variational autoencoding (SSVAE) in this technique. In addition, the attack detection mechanism is identified and enhanced using self-attention linked bidirectional long short-term memory (SA-BiLSTM).

Kute, Shruti Suhas et al. [33] provided a study of cutting-edge research involving the IoT in healthcare, particularly on obesity, overweight, and persistent degenerative illnesses. Secrecy, integrity, authentication, access, trust, validation, information management, and storage and availability issues must be resolved for IoT in real-world applications. A description of the security, privacy, and trust problems in IoT-based machine learning depending on healthcare systems is also provided in this study.

Using a hybrid deep neural network system, Ali, Aitizaz et al. [34] proposed a new group theory (GT) that depended on the binary spring search (BSS) technique. The blockchain was presented as a distributed database to guarantee secure tracking and keyword-based access to the dataset. The proposed methodology also offered a secure critical revocation method, and various policies were updated accordingly. The security of patient healthcare information access systems incorporating blockchain and a confidence chain addressed the efficiency and safety difficulties in the existing schemes for exchanging both forms of digital healthcare data.

One such IoT and cloud computing application was the topic of a study by Anuradha, M. et al. [35]. This work’s primary goal was to develop a cancer prediction system utilizing the Internet of Things after extracting the specifics of blood results to determine whether they were normal or abnormal. Additionally, the blood results of cancer patients were encrypted and stored in the cloud for easy Internet access by doctors and nurses who needed to handle patient data discreetly. This focused on improving the calculations and processing in the healthcare industry. To offer authentication and security when dealing with patients with cancer, encryption and decryption were performed using the AES method.

Initial emphasis was placed on the fundamental security requirements for a Body Sensor Network (BSN)-based contemporary healthcare system. As a result, BSN-Care was proposed, a successful IoT-based healthcare system that enabled BSN to effectively meet these requirements Satyanarayan et al. [36].

The Authentication, Authorization, and Audit Logs (AAA) services were achieved by FBASHI, a system built on blockchain technology and fuzzy logic Zulkifl, Z. and Khan et al. [37]. This work provided a heuristic method for conducting driven flexible security, offering AAA services for medical care IoTs and networks based on the blockchain. It also suggested an approach for action driven flexible security using fuzzy logic.

For IoT-enabled hospitals, a reciprocal authentication method that protects privacy was suggested by Das, S. and Namasudra in order to accomplish quick and efficient network device verification [38]. This suggested authentication method was built using lightweight cryptographic primitives, such as XOR, combination, and hash operation, to accommodate the computing power of the IoT devices. The suggested strategy could block unwanted devices from accessing healthcare networks by establishing a safe connection between an approved device and a gateway.