Privacy information protection of medical internet based on double chaos encryption algorithm

Abstract

In order to avoid the risk of patients’ private information leakage, this paper puts forward a research on the protection of medical Internet private information based on double chaotic encryption algorithm. This paper analyzes the quantification of risk indicators for privacy information protection of medical Internet, establishes the risk quantification structure of health care big data according to the quantitative calculation results, and puts forward the strategy of controlling access to health care big data, configuring the risk level, describing the attributes of the system database, and realizing the privacy information protection of medical Internet under the double chaotic encryption algorithm. The experimental results show that the real identity of patients is protected to a certain extent in the protection of private information of medical internet after applying this method. Moreover, this method has high storage integrity and small storage standard deviation, and the method in this paper can effectively resist network intrusion. Therefore, it shows that this method has a good effect of protecting private information of medical Internet.

Keywords

Double chaotic encryption algorithm medical internet private information information protection.

1 Introduction

With the rapid development of information technology, medical internet has become an indispensable part of modern medical system. The rise of medical internet provides more convenient and efficient medical services for patients, and also provides more comprehensive and accurate patient data for doctors and medical institutions [1, 2]. However, with the popularization of medical Internet and the expansion of its application scope, the problem of privacy information protection has become increasingly prominent [3, 4]. The private information on the medical internet includes sensitive data, such as personal identity information, medical records, and diagnosis results. The disclosure of this data would seriously affect the personal privacy and safety of patients. Additionally, the medical internet faces security threats, such as hacking, data leakage, and illegal access. Therefore, protecting private information on the medical internet has become one of the important topics in the medical system today.

In the current privacy information protection methods, encryption technology is widely used. Encryption technology ensures the security of data during transmission and storage by transforming sensitive data into incomprehensible ciphertext. Reference [5] proposes a medical information security watermarking scheme based on DCT and DWT. In the first scheme, the combination of DCT and Schur decomposition is implemented. In order to obtain a good compromise between robustness and imperceptibility, integration is performed in the intermediate frequency of the image. In the second scheme, the combination of DWT and Schur decomposition provides a more robust watermark distribution. The experimental results show that the proposed method maintains high-quality watermark images and has strong robustness to some conventional attacks. These schemes allow the protection of patients’ information, thus ensuring the confidentiality of personal data. Reference [6] proposed a key area protection scheme of medical images based on two-dimensional code and reversible data hiding. Firstly, the key areas, that is, the diseased areas of the images, were identified by using the coefficient of variation. Then, other areas are divided into blocks to analyze the texture complexity. A new reversible data hiding algorithm is proposed, which embeds the contents of key areas into high texture areas. On this basis, a fast response code is generated by using the ciphertext of the basic image information to replace the original lesion area. The experimental results show that this method can not only transmit the patient’s sensitive information safely by hiding the pathological contents, but also store the copyright information through the QR code to realize accurate image retrieval. Reference [7] puts forward the influence of publicly-funded medical insurance in India on health care utilization and financial risk protection, and the article provides information on the influence of publicly-funded health insurance on financial risk protection and medical care utilization. Design system review. Cross-sectional studies based on comparison before and after implementation, impact assessment, difference design in differences, before and after implementation, experiments and quasi-random experiments meet the inclusion criteria. Masood et al. [8] proposes a lightweight cryptographic system based on Henon chaotic mapping, Brownian motion, and Chen chaotic system to ensure the security of medical images and prevent unauthorized access. The effectiveness of the system has been demonstrated through histogram analysis, adjacent pixel correlation analysis, contrast analysis, uniformity analysis, energy analysis, NIST analysis, mean square error, information entropy, pixel number change rate, unified average change intensity, peak signal-to-noise ratio, and time complexity. Aggarwal et al. [9] proposes a lightweight joint deep learning architecture while maintaining data privacy constraints for rice leaf disease classification. The distributed client server design of this framework protects the data privacy of all clients and verifies the effectiveness of the joint deep learning model by using independent identically distributed (IID) and non IID data. This study first extracts features from various pre trained models and ultimately selects EfficientNetB3 as the baseline model. Can [10] proposes a medical information privacy protection method based on machine learning. Joint learning is a promising candidate for developing high-performance models while protecting personal privacy. It is a privacy protection solution that stores model parameters rather than data itself. Apply federated learning to medical information collected using smart belts for monitoring different events. The use of federated learning in wearable biomedical monitoring systems based on the Internet of Things has achieved encouraging results by protecting data privacy. Parashar and Shekhawat [11] proposes a medical information privacy protection method based on deep learning, which makes it more difficult to maintain the privacy of gait datasets in deep learning pipelines. One of the popular techniques to stop accessing datasets is to use anonymization, and a reversible gait anonymity pipeline is proposed, which modifies the geometric structure of gait through deformed images, i.e. texture modification. This modified data prevents hackers from using the dataset for adversarial attacks.

Although the above research has made some progress, the traditional encryption algorithm has some limitations in the medical Internet. The traditional encryption algorithm is slow to encrypt and decrypt large-scale data, which can’t meet the real-time processing requirements of a large number of data in the medical Internet. At the same time, the key management and distribution mechanism of traditional encryption algorithm is relatively weak, and it is easy to be attacked and cracked. Therefore, this paper proposes a research on privacy information protection of medical Internet based on double chaotic encryption algorithm. As a new encryption technology, double chaotic encryption algorithm has high security and efficiency. The algorithm takes advantage of the randomness and unpredictability of chaotic system to transform plaintext data into ciphertext with high complexity. Compared with the traditional encryption algorithm, the double chaotic encryption algorithm can better cope with the security challenges in the medical Internet. The goal of this study is to provide an efficient, safe and reliable privacy information protection solution for the medical Internet, so as to promote the healthy development of the medical Internet. By protecting patients’ private information, we can improve patients’ trust in the medical system, promote the further optimization of medical services, and make greater contributions to human health.

2 Privacy information protection of medical internet

2.1 Quantification of information protection risk indicators

Due to the behavioral differences among visiting users in protecting private medical internet information, it is necessary to adopt risk adaptability and quantify risk indicators. Assuming that the instability of access behavior is related to the information entropy of user access behavior, the greater the information entropy, the greater the possibility of users snooping on patient information.

Assuming that X represents a random variable and P (X) represents a random distribution function in the process of accessing private information on the medical Internet, the information entropy H (X) of X is:

$H (X) = P (X) \times x_{i}$ (1)

In formula (1), x_i represents the access behavior of the ith user.

The random variable of formula (1) According to the user’s access behavior, X’s information entropy can be divided into two stages, namely, the work target stage and the access to medical records stage, so the quantification of risk indicators also needs to be divided into two stages.

During the stage of user selection of work goals, it is necessary to define the probability of user selecting certain work goals in the calculation process of information entropy. Assuming the ith user u_i is diagnosed and treated for the jth patient v_j, the probability D_m of the user u_i choosing the mth work target P_{D
_m} is:

$P_{D_{m}} = \frac{H (X) \times (u_{i}, v_{j})}{J_{u_{i}, v_{j}} \times K_{D_{m}}}$ (2)

In formula (2), J_{u_i,v_j} represents the set of work targets accessed by user u_i during the diagnosis and treatment of patient v_j; K_{D
_m} represents the number of times the user o_k has selected the work target u_i. So when user v_j is diagnosing and treating patient H (X) ′, the work target information entropy KK accessed is:

$H {(X)}^{'} = P {(X)}^{'} \times x_{i}^{'} \times P_{D_{m}}$ (3)

In formula (3), P (X) ′ represents the random distribution function accessed during the user’s selection of work objectives; $x_{i}^{'}$ represents the access behavior of the ith user accessed during the user’s selection of work objectives stage. The result obtained through formula (3) is the information entropy of the user’s selection of the work target stage [12–14].

The information entropy calculation process is the same as that of the user selecting the work target stage. To calculate the information entropy of accessing medical records, it is also necessary to define the probability P of the user selecting medical records. Assuming that the probability P_{D
_l} of user u_i selecting medical record m_l in item l under work objective D_m is:

$P_{D_{l}} = \frac{H {(X)}^{'} \times (u_{i}, v_{j})}{J_{u_{i}, v_{j}}^{'} \times K_{D_{m}}^{'}}$ (4)

In formula (4), $J_{u_{i}, v_{j}}^{'}$ represents the medical record set accessed by the user when the diagnosis and treatment patient v_j and work goals are determined; $K_{D_{l}}^{'}$ represents the number of times the user has accessed the medical record m_l. So, under user v_i’s work goal D_m, the quantitative results of the risk indicators for medical internet privacy information protection are:

$H (u_{j}) = \frac{H (X) + H {(X)}^{'}}{P_{D_{m}} + P_{D_{l}}}$ (5)

The result calculated by formula (5) is the quantified result of the risk index of privacy information protection of medical Internet.

2.2 Establishing the risk quantification structure of health care big data

Based on the quantitative calculation results of the medical internet privacy information protection risk indicators mentioned above, assuming that the input user access request input vector is X = [x₁, x₂, x₃] ^T, where T represents the access cycle transposition; At this point, the input variables of access requests are divided into four categories: high, medium, low, and very low. Based on the distribution characteristics of risk indicators [15, 16], a dual chaotic encryption algorithm is used to establish a user access request function. Therefore, let the mean of the membership function be c_ij; If the standard deviation of the membership function is $σ_{ij}$ , the $u_{i}^{j}$ expression of the membership function for each variable is calculated as:

$u_{i}^{j} = e^{\frac{{(x_{i} - c_{ij})}^{2}}{σ_{ij}^{2}}} \times H (u_{j})$ (6)

In Formula (6), the natural constant is represented; At this time, the final output variable of Formula (6) is Y, and then:

$Y = u_{k}^{1} \times β_{k} \times {\bar{β}}_{k} \times y_{k}$ (7)

In formula (7), $u_{k}^{1}$ represents the total number of rules; β_k represents the output result of the predecessor network; ${\bar{β}}_{k}$ represents the average output value of the predecessor network, and y_k represents a successor rule, which constitutes the risk quantification structure of health care Big data, as shown in Fig. 1.

Fig. 1

Quantitative structure diagram of health care big data risk.

As can be seen from Fig. 1, the risk quantification structure of health care big data established this time takes the user request information as the input vector, and through the superposition of various layers, the input user request information is superimposed and learned [17–19]. At this time, the structure will automatically adjust the learning parameters until the training error is within the set range, thus completing the establishment of the risk quantification structure of health care big data. Based on the risk quantification structure designed for healthcare big data, the risk control can be implemented, and the access control strategy for healthcare big data can be adjusted according to the level of access risk.

2.3 Control the access strategy of health care big data

Based on the risk quantification structure of health care big data, the process of controlling health care big data access is designed, as shown in Fig. 2.

Fig. 2

Controlling the access process of health care big data.

As can be seen from Fig. 2, according to the quantitative results of risk indicators, the information entropy of the user’s choice of work target stage and the information entropy of the user’s access to medical records stage are obtained, and the access risk is controlled through the risk quantification structure of health care big data. During the control process, it is necessary to assign access roles, assess the access risk of the individual requesting access, and adjust the risk quota based on the calculation results of the access risk. This means that when there’s a remaining risk quota, the corresponding access rights should be allocated accordingly; When the risk limit is exhausted, access to health care big data is blocked [20]. In the process of controlling the access to health care big data as shown in Fig. 2, it is also necessary to determine the risk quota, risk level and reduce the risk quota. The calculation process is as follows:

Step 1: Determine the risk limit. Assuming that the user u who requests access to health care big data has an available risk limit of Q_u during the access period, and the total number of users who have made access to health care big data during this time period is |U|, there are:

$A_{k} = \frac{Q_{u} (o)}{| U | \times E_{rfjghkl}} \times Y$ (8)

In formula (8), o represents the o time period; E_rfjghkl represents the average consumption of risk limit; According to the average value of risk limit consumption in o time period, the allocated risk limit is determined.

Step 2: Determine the access risk level. Based on the risk quantification model of health care big data, the risk level is determined based on the calculation results of risk value. There are five levels, and the risk quota reduction ratio corresponding to each level is different. The risk level determined this time and its corresponding quota reduction ratio are shown in Table 1.

Table 1

Risk level configuration table

Risk level	Grade strength	Value at risk range	Quota reduction ratio	Risk assessment score	Frequency of risk discovery
1	Safe	0 < R ⩽ 0.2	0.5	100	200
2	Safer.	0.2 < R ⩽ 0.4	0.4	80	180
3	Medium	0.4 < R ⩽ 0.6	0.3	50	150
4	More dangerous	0.6 < R ⩽ 0.8	0.2	30	120
5	Danger	0.8 < R ⩽ 1	0.1	10	80

Note: R stands for value at risk.

Step 3: Reduce the risk quota. Assuming that the user u requesting access to health care big data has a risk value of R_(u,i) at the i time, there are:

$A_{(u, i)} = R_{(u, i)} \times A_{k}$ (9)

In formula (9), A_(u,i) represents the deduction amount of risk limit.

According to the above formulas (8) and (9), as well as the risk level determined in the access control process of health care big data and the corresponding quota reduction ratio, it can be known that in order to ensure the smooth acquisition of health care big data, the risk amount will be allocated to each role regularly and quantitatively. Therefore, the distribution principle of risk limit is: under the premise of ensuring the smooth access of legal users, the risk limit of legal users should be higher than the consumption of risk value within a certain time range. Therefore, in order to prevent illegal users from accessing health and medical big data, the risk limit of illegal users should be lower than the consumption of illegal users within a certain time range.

3 Realize the protection of private information of medical internet

Because the database of hospital medical information system is mostly composed of documents, the database tree is used to describe the medical information system. In the hospital medical information system, the system database is regarded as a hierarchical set of elements, attributes and texts, and the tree diagram shown in Fig. 3 is used to describe the system database.

Fig. 3

Tree description of system database.

In Fig. 3, the ID can determine the value of the goal and the destination. According to the tree description of the system database shown in Fig. 3, if the system database is defined as a sextuple λ, then:

$λ = (A_{1}, A_{2}, A_{3}, B_{1}, B_{2}, B_{3})$ (10)

In formula (10), A₁ represents a set of finite nodes of a document; A₂ represents the mapping on the attribute label; A₃ represents the mapping on the string; B₁ represents the mapping on the element tag; B₂ represents a mapping on a set of types; B₃ represents the root of database tree description λ.

Based on the system database elements and database operation mode determined by the aforementioned contents, identity authentication technology is adopted to verify the user’s identity for accessing the system database, thereby preventing the forging of system database access identities. According to the tree description and database definition of the system database shown in Fig. 3, the identity authentication technology is adopted to authenticate the access user of the system database according to the six steps of information sending, receiving, encryption, decryption, random number generation and certificate verification. Then the identity authentication process of the system database designed in this study is as follows:

Step 1: Connect the system database authentication client and the user identity authentication module, and send information to the user identity authentication module.

$M_{HV} = \frac{λ \times G_{H} \times K_{H} \times U_{H}}{D_{V}} A_{(u, i)}$ (11)

In formula (11), G_H stands for plaintext message; K_H stands for user identification number; U_H stands for user certificate; D_V represents the random number generated by the user authentication module.

Step 2: After receiving the plaintext message shown in Formula (11), the user identity authentication module will use the public key of the authentication authority to verify the legality of Z_u. The authentication authority will obtain the user’s public key Z₁, and the user identity authentication module can encrypt N₁ with Z₁ and return the information to the authentication client.

$S_{D} = \frac{M_{HV} \times Z_{1} \times N_{1}}{F_{GH}}$ (12)

In formula (12), Z₁ represents the certificate of the user identity authentication module; N₁ represents the private key of the user authentication module; F_GH represents the random number generated by the authentication client.

Step 3: After receiving the reply message shown in formula (12), the authentication client will use the public key of the authentication authority to verify the authenticity of N₁. At the same time, it will also obtain the public key R₂ of the user identity authentication module and use Z₂ to verify Q_GH. If Z₂ can be obtained, it is considered that the sender of the information is the user identity authentication module. At this point, the user can use the private key to decrypt S_B to obtain N₁ and authenticate the information sent by the client:

$S_{B} = \frac{M_{RZ} \times R_{2} \times Z_{2}}{Q_{GH}}$ (13)

Step 4: After receiving the sending information shown in formula (13), the user identity authentication module decrypts the R₂ using the private key Z₁, and then decrypts the Z₂ using Z₁. If S_B is obtained, it indicates that the user applying for access is a legitimate user. At this time, the user’s identity is verified and can access the system database.

According to the attributes of the database, the double chaotic encryption algorithm is used to correlate the privacy configuration, and the user security level is adaptively adjusted to remind users of privacy security. Double chaotic encryption algorithm is an encryption method based on chaos theory, which uses more than two chaotic systems for encryption operation. Chaotic system refers to a kind of nonlinear dynamic system, which is highly sensitive and unpredictable.

The basic idea of double chaotic encryption algorithm is to transform plaintext into ciphertext by mixing plaintext data with chaotic sequence to realize data protection and privacy security.

Specifically, the double chaotic encryption algorithm usually includes the following steps:

Step 1: Initialization: Select appropriate initial parameters and initialize the chaotic system.

Step 2: Generate chaotic sequences: According to the characteristics of chaotic systems, generate a group of highly irregular and complex chaotic sequences.

Step 3: key generation: the chaotic sequence is transformed into a key sequence through a certain mapping relationship.

Step 4: data encryption: XOR the plaintext data and the key sequence or other confusing operations to obtain ciphertext data.

Step 5: Decryption of data: Decryption of ciphertext data by using the same key sequence to restore plaintext data.

The advantage of double chaotic encryption algorithm lies in its high confidentiality and anti-attack. Because of the randomness and unpredictability of chaotic sequences, it is difficult for attackers to crack keys and obtain plaintext information. In addition, the double chaotic encryption algorithm also has good scalability and flexibility, and the parameters can be adjusted and improved according to specific needs.

According to the double chaotic encryption algorithm, the privacy information protection of medical internet is realized, and the specific steps are as follows:

Step 1: Process the privacy information of medical Internet that needs to be protected, remove unnecessary identifiers or personal identity information, and desensitize it to protect privacy. Select the appropriate chaotic system and initialize its parameters to ensure that the chaotic system is in the initial state.

Step 2: According to the selected chaotic system and initialization parameters, a group of highly irregular and complex chaotic sequences are generated. Assume that the expression f (T) of chaotic sequence is as follows:

$f (T) = r_{j} \times E_{j} \times E_{h} \times S_{B}$ (14)

In formula (14), r_j represents the j th natural number and E_j represents the number of natural numbers; E_h indicates the period.

Step 3: Transform the chaotic sequence into a key sequence through a certain mapping relationship, and these keys will be used for subsequent data encryption and decryption operations. The preprocessed privacy information of medical Internet is calculated with the generated key sequence to obtain ciphertext data.

If the ciphertext authorization is divided into three types: η₁, infinite η₀, and restricted η₂, there are:

$\begin{matrix} W_{ER} = & \frac{(η_{1} + η_{0} + η_{1}) \times P \times Q \times f (T)}{N} \\ \times (t_{max} - t_{min}) \end{matrix}$ (15)

In formula (15), N represents the number of operations; P represents access permissions; Q represents authorization determination; t represents the time interval within the [t_min, t_max] range, where t_min represents the lower limit and t_max represents the upper limit.

Step 4: Transfer the encrypted medical internet privacy information to the desired target location. Ensure the security of ciphertext during transmission. If ciphertext data needs to be stored on a server or database, it is necessary to ensure the security of the storage and take corresponding security measures such as access control and encrypted storage. Use the same key sequence to decrypt the received ciphertext data to restore the original medical internet privacy information.

$λ_{η} (u, q, p) = \frac{V_{B} \times W_{ER}}{G_{1} \times N_{2} \times M_{3}}$ (16)

In formula (16), V_B represents an access request; G₁ represents authorization tuple; N₂ represents illegal access; M₃ indicates that the authorization function returns an authorization.

Step 5: Further process or analyze the decrypted data, extract useful medical indicators, generate reports, etc., thereby completing the research on privacy information protection methods for medical internet.

The privacy information protection process of medical internet based on dual chaotic encryption algorithm is shown in Fig. 4.

Fig. 4

Medical internet privacy information protection process based on double chaos encryption algorithm.

4 Experimental analysis

4.1 Experimental preparation

In order to verify the effect of research on privacy information protection of medical internet based on double chaotic encryption algorithm, experiments were carried out. The experiment prepares the medical Internet privacy information data set, ensures that the data set contains sensitive information, installs the required programming language and encryption library, builds the experimental environment on the computer, and preprocesses the medical Internet privacy information data set, that is, cleans the data and removes irrelevant information. Using the selected programming language, the double chaotic encryption algorithm is implemented to ensure that the realized algorithm can encrypt and decrypt data. The data set is divided into training set and test set. The training set is used to train the double chaotic encryption algorithm, and the data in the test set is encrypted by the double chaotic encryption algorithm. The encrypted data is decrypted and compared with the original data. To compare the effects of encryption algorithms and evaluate their effectiveness in protecting private information of medical Internet. The experimental parameters are shown in Table 2.

Table 2
Experimental parameter setting table

Serial number Parameter Content

1 Encryption algorithm Double chaos encryption algorithm

2 Data set Medical internet privacy information dataset

3 Experimental platform Computers, servers

4 Experimental tools Python, c++programming languages

Serial number	Parameter	Content
1	Encryption algorithm	Double chaos encryption algorithm
2	Data set	Medical internet privacy information dataset
3	Experimental platform	Computers, servers
4	Experimental tools	Python, c++programming languages

According to the experimental parameter configuration, 6 patients from the hospital were selected as experimental samples, and the original information of the patients’ visits is shown in Table 3.

Table 3

Original information of patient visits

Patient	Full name	Age	Gender	Disease	Case number
1	Chen Qiangdong	25	Male	Emphysema	2023199825001
2	Guo Xiaoxiang	40	Male	Infectious diarrhea	2023198340002
3	Wang Xiang	52	Male	Constipation	2023197152003
4	Xu Tianjiao	41	Female	Colitis	2023198241004
5	Liu Jiehui	45	Female	Duodenal ulcer	2023197845005
6	Li Li	47	Female	Lung cancer	2023197647006

Applying the method presented in this article, an anonymous publication table for protecting patient medical internet privacy information was obtained, and the results are shown in Table 4.

Table 4

Patient medical internet privacy information protection form

Patient	Full name	Age	Gender	Disease	Case number
1	Chen * Dong	20	Male	Emphysema	2023**25*
2	Guo * Xiang	29	Male	Infectious diarrhea	2023**40*
3	Wang*	32	Male	Constipation	2023**52*
4	Xu * jiao	40	Female	Colitis	2023**41*
5	Liu * Hui	22	Female	Duodenal ulcer	2023**45*
6	Li*	24	Female	Lung cancer	2023**47*

In Table 4, the patient’s name is partially masked with an asterisk (*), and the year and part of the case number are also blurred. By comparing Table 3 and Table 4, it can be seen that after applying the method in this paper, the two attribute information, name and case number, are anonymous. In the privacy information protection of medical internet, the true identity of patients is protected to a certain extent, so as to avoid speculating the personal information of users through the relevance of information, so as to protect the privacy and sensitive information of patients and realize the privacy information protection of medical internet. At the same time, the age in the patient’s medical record number is still reserved, so that doctors can track and manage medical records according to the patient’s age.

4.2 Analysis of experimental results

When protecting private information of medical Internet, data loss often occurs, resulting in low storage capacity. In view of this defect, the method of reference [5], the method of reference [6] and the proposed method are used to compare and test the amount of protected data respectively, and the storage integrity of the three methods for medical Internet privacy information protection data is verified. The specific test results are shown in Fig. 5.

Fig. 5

Comparison results of data storage of medical Internet privacy information protection under different methods.

According to Fig. 5, it can be known that the amount of medical Internet privacy information protection data to be stored in this experiment is 5000. For storing 5000 data, the method of reference [5] can completely store the medical Internet privacy information protection data every time, which shows that the method of reference [5] has not encountered the problem of data loss and proves that the storage integrity of this method is the highest. However, the method used in this experiment can be completely realized when storing 1000 data, but with the continuous increase of data, there is a problem of data loss, which leads to the smallest storage result among the three methods. Therefore, it can be concluded that the method in reference [5] has the largest storage capacity, while the proposed method has the smallest storage capacity.

Storage standard deviation belongs to the average of data deviation from the average distance, which can effectively reflect the dispersion degree of data set. Therefore, the amount of traceable storage data of medical devices determines the change of storage standard deviation, which indicates the load balance of cloud storage network. The smaller the standard deviation, the better the load balance effect. According to this feature, the load balance comparison test is carried out by three methods, and the specific test results are shown in Table 5.

Table 5

Storage table under three methods 5 Storage load comparison test under three methods 5 Storage load comparison test under three methods Storage load comparison test

Data volume/piece	Storage standard deviation
	Proposed method	Reference [5] method	Reference [6] method
500	16	29	31
1000	28	45	54
1500	31	54	58
2000	32	68	69
2500	40	79	80
3000	43	81	86
3500	51	92	94
4000	66	102	109
4500	77	125	128
5000	89	139	146

According to Table 5, with the increase in the amount of medical internet privacy information protection data, the storage standard deviation of all three methods has increased. The storage standard deviation of the proposed method is relatively small, which proves that the proposed method performs better in data load balancing compared to the methods in references [5] and [6]. However, the storage standard deviation of the methods in references [5] and [6] is relatively large, which verifies the poor performance of these two methods in load balancing.

In order to evaluate the security of medical internet privacy information, understand the performance of different methods in real attack scenarios and their defense capabilities against different types of attacks, identify potential vulnerabilities and weaknesses in medical internet platforms, reveal security vulnerabilities, and strengthen corresponding defense measures. Further enhance the security protection level of privacy information, improve and reinforce existing defense mechanisms, and compare and verify the method proposed in this paper with the methods in reference [5] and [6] based on the success rate of attack defense. The success rates of attack defense for the three methods are shown in Table 6.

Table 6

Attack defense success rate results

Data volume/piece	Attack defense success rate/%
	Proposed method	Reference [5] method	Reference [6] method
500	96.8	70.5	74.5
1000	97.5	71.7	71.2
1500	97.1	69.8	74.4
2000	95.8	72.3	74.2
2500	96.7	68.9	69.1
3000	97.6	70.8	64.5
3500	94.5	74.6	69.3
4000	96.3	71.5	61.8
4500	95.4	68.4	78.4
5000	96.1	71.4	72.4

By analyzing the data in the table, it can be concluded that the method proposed in this paper has advantages over the methods in reference [5] and [6] in terms of attack defense success rate. From the data results, it can be seen that the proposed method exhibits a high success rate in attack defense under all data volumes, significantly surpassing the methods in reference [5] and [6]. For example, under 500 data volumes, the attack defense success rate of our method reached 96.8%, while the methods in reference [5] and [6] were 70.5% and 74.5%, respectively. In addition, the method proposed in this paper maintains a relatively stable attack defense success rate under different data volumes, demonstrating a certain degree of robustness. Compared with the method in reference [6], our method achieved higher success rates in attack defense on most data volumes. For example, on 4500 and 5000 data volumes, our method achieved success rates of 95.4% and 96.1%, respectively, while the method in reference [6] achieved success rates of 68.4% and 72.4%, respectively. Therefore, from the perspective of attack defense success rate, the method proposed in this paper has obvious advantages in protecting medical internet privacy information, which can maintain high defense effectiveness and have a certain degree of stability under different data volumes. This further verifies the superiority and feasibility of the method proposed in this paper in protecting medical internet privacy information.

5 Conclusion

The aim of this study is to explore a medical internet privacy information protection method based on dual chaos encryption algorithm, and verify its feasibility and effectiveness in practical applications. By conducting in-depth research on the principles and characteristics of the dual chaos encryption algorithm, and combining it with the actual needs of the medical internet, a privacy information protection method suitable for the medical internet has been designed. The experimental results show that after applying the method proposed in this article, the true identity of patients is protected to a certain extent in the protection of medical internet privacy information; With the increase of data volume for privacy information protection in medical internet, the storage standard deviation of the proposed method has been improved, and the storage standard deviation is relatively small, which has a good protection effect. However, there are still some shortcomings in this study. This study only provides a preliminary exploration of the application of the dual chaotic encryption algorithm in the medical internet, and further research and practice are needed. Future research can consider introducing more encryption technologies and algorithms into the medical internet to enhance the effectiveness and performance of privacy information protection.

References

Lin

T.W.

and Hsu

C.L.

, FAIDM for medical privacy protection in 5g telemedicine systems, Applied Sciences 11(3) (2021), 1155–1159.

Wang

, Fan

, Yang

, et al., A new RFID ultra-lightweight authentication protocol for medical privacy protection in smart living, Computer Communications 186(15) (2022), 121–132.

, Tian

, Li

, et al., Improving prediction for medical institution with limited patient data: leveraging hospital-specific data based on multicenter collaborative research network, Artificial Intelligence in Medicine 113(21) (2021), 1–13.

Liu

, Ni

, Wang

, et al., Fog computing solution of privacy protection of the information in mobile medical device, China Medical Equipment 19(1) (2022), 138–142.

Fares

, Khaldi

, Redouane

, et al., DCT & DWT based watermarking scheme for medical information security, Biomedical Signal Processing and Control 66(4) (2021), 838–846.

Zhang

, Xu

, Ma

, et al., Medical image key area protection scheme based on QR code and reversible data hiding, Security and Communication Networks 7(7) (2021), 1–11.

Reshmi

, Unnikrishnan

, Rajwar

, et al., Impact of public-funded health insurances in India on health care utilisation and financial risk protection: a systematic review, BMJ Open 11(12) (2021), 1–11.

Masood

, Driss

, Boulila

, et al., A lightweight chaos-based medical image encryption scheme using random shuffling and XOR Operations, Wireless Personal Communications 127(19) (2021), 1405–1432.

Aggarwal

, Khullar

, Goyal

, et al., Lightweight federated learning for rice leaf disease classification using non independent and identically distributed images, Sustainability 15(16) (2023), 12149–12168.

10.

Can

S.E.C.

, Privacy-preserving federated deep learning for wearable IoT-based biomedical monitoring, ACM Transactions on Internet Technology 21(1) (2021), 1561–1572.

11.

Parashar

and Shekhawat

R.S.

, Protection of gait data set for preserving its privacy in deep learning pipeline, IET Biometrics 11(6) (2022), 557–569.

12.

Sun

, Zhang

and Feng

, Do individuals disclose or withhold information following the same logic: a configurational perspective of information disclosure in social media, Aslib Journal of Information Management 74(4) (2022), 710–726.

13.

Zhang

S.B.

, Yuan

L.J.

, Mao

X.J.

, et al., Privacy protection method for k-modes clustering data with local differential privacy, Acta Electronica Sinica 50(9) (2022), 2181–2188.

14.

Zhou

Y.P.

and Wang

G.C.

, Research on security structure retrieval simulation of sensitive information in mobile internet, Computer Simulation 39(9) (2022), 451–455.

15.

, Du

H.N.

and Li

, Blockchain supervisable privacy protection scheme based on group signature and attribute encryption, Computer Engineering 48(6) (2022), 132–138.

16.

Chen

, Meng

, Zhou

, et al., A blockchain-based medical data sharing mechanism with attribute-based access control and privacy protection, Wireless Communications and Mobile Computing 26(5) (2021), 1–12.

17.

, Zhang

, Wang

, et al., MNSSp3: Medical big data privacy protection platform based on Internet of things, Neural Computing & Applications 34(14) (2022), 11491–11505.

18.

Wang

M.S.

, Yao

, Gao

F.X.

, et al., Study on differential privacy protection for medical set-valued data, Computer Science 49(4) (2022), 362–368.

19.

Zhang

S.Q.

, Li

X.H.

, Jiang

X.Y.

, et al., AUR-Tree differential privacy data publishing algorithm for medical data, Application Research of Computers 39(7) (2022), 2162–2166+2171.

20.

Liu

, Peng

Y.F.

, Tian

, et al., A survey on medical information privacy protection based on blockchain, Journal of Zhengzhou University (Natural Science Edition) 53(2) (2021), 1–18.

Privacy information protection of medical internet based on double chaos encryption algorithm

Abstract

Keywords

1 Introduction

2 Privacy information protection of medical internet

2.1 Quantification of information protection risk indicators

4.1 Experimental preparation

Table 2 Experimental parameter setting table Serial number Parameter Content 1 Encryption algorithm Double chaos encryption algorithm 2 Data set Medical internet privacy information dataset 3 Experimental platform Computers, servers 4 Experimental tools Python, c++programming languages

References

Table 2
Experimental parameter setting table

Serial number Parameter Content

1 Encryption algorithm Double chaos encryption algorithm

2 Data set Medical internet privacy information dataset

3 Experimental platform Computers, servers

4 Experimental tools Python, c++programming languages