Egyptian Informatics Journal 28 (2024) 100569

Contents lists available at ScienceDirect

Egyptian Informatics Journal

journal homepage: www.sciencedirect.com

Full Length Article

A hybrid encryption algorithm based approach for secure privacy

protection of big data in hospitals
Wei Li a,* , Qian Huang b
a
Information Center, Renmin Hospital of Wuhan University, Wuhan 430060, China
b
General Education Department, Wuchang Polytechnic College, Wuhan 430060, China

A R T I C L E I N F O A B S T R A C T

Keywords: Aiming at the hidden danger of information security caused by the lack of medical big data information security
Hospital big data firewall, this paper proposes a security privacy protection method for hospital big data based on hybrid
Security and privacy protection encryption algorithm. First, collect hospital big data including hospital medical business system, mobile wearable
Hybrid encryption algorithm
devices and big health data; Secondly, use byte changes to compress hospital big data to achieve safe trans­
ECC encryption
mission of hospital big data; Then, the hospital sender uses the AES session key to encrypt the hospital big data
AES encryption
Hash value and the ECC public key to encrypt the AES session key, uses SHA-1 to calculate the hash value of the medical big
Signature verification data, and uses the ECC public key to sign the hash value; The hospital receiver uses the ECC private key to verify
the signature, and decrypts the AES session key using the ECC private key. After the AES session key decrypts, the
hospital big data, the hospital big data security privacy protection is completed. The experimental results show
that the method is superior to conventional ECC algorithm or RSA and AES hybrid encryption algorithm in terms
of encryption and decryption time and security strength. The average correlation coefficient of encrypted hos­
pital big data is only 0.0576, and the RL curve value is low and gentle. The encrypted data has good scrambling
effect and low privacy leakage probability, which ensures the confidentiality and integrity of medical data in the
transmission process.

1. Introduction of this method is slow, and it is not suitable for processing hospital big
data with a large amount of data [6]. Suma and Madhumathy used
Hospital informatization construction plays an increasingly signifi­ homomorphic encryption and firefly swarm optimization algorithm in
cant role in ensuring and supporting the smooth progress of medical and hospital big data to protect privacy and perform encrypted data opera­
health system reform and ensuring the sustainable development of tions. Through encryption, public key and private key generation, data
hospitals [1]. It is used as a central source of clinical data to support upload, cloud encryption processing, result return and decryption, they
management decisions. It has high data efficiency, high quality Exten­ realized the access and processing of hospital big data for privacy pro­
sibility and topic oriented features [2,3]. The management of health and tection [7]. However, the calculation cost of this method is high, which
medical data should be guided by corresponding laws and regulations requires a lot of computing resources and time. Jaffar et al. used
[4]. For this reason, it is of great significance to study the security and encryption technology, digital watermarking, access control and other
privacy protection methods of hospital big data for protecting patient means, combined with legal and policy protection, to achieve cloud
privacy, preventing data from being tampered with, improving data based big data security and privacy protection of electronic hospitals
security, promoting data sharing, improving hospital reputation, and [8]. However, the use of this method requires additional storage space
complying with laws and regulations [5]. and computing resources, which will increase the cost of data storage
In response to this problem, Pawar et al. proposed a RSA + AES and processing. Kanna and Vasudevan used the surge ElGamal encryp­
hybrid encryption algorithm to encrypt sensitive data in hospital big tion algorithm to store hospital big data through an improved privacy
data to prevent unauthorized access and theft, and realized the security aware security multi cloud model to achieve data privacy protection and
and privacy protection of hospital big data through the use of firewall secure access [9]. However, this method is prone to data leakage, ma­
and high encryption level VPN network. However, the encryption speed licious attacks and phishing in a cloudy environment.

* Corresponding author.
E-mail address: [email protected] (W. Li).

https://doi.org/10.1016/j.eij.2024.100569
Received 22 July 2024; Received in revised form 29 September 2024; Accepted 21 October 2024
Available online 5 November 2024
1110-8665/© 2024 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Computers and Arti cial Intelligence, Cairo University. This is an open access
article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

AES session key is used to encrypt hospital big data, and its security is Mobile wearable devices play an important role in hospital big data.
the highest among symmetric cryptographic algorithms of the same With the popularization of smartphones and the development of mobile
kind. Therefore, in terms of patient data security, it can resist the attacks medical technology, more and more patients and healthcare workers use
of data eavesdropping and disclosure, and ensure the confidentiality and mobile devices for medical activities. These mobile devices can record
privacy of hospital big data; Using ECC public key to encrypt the session the patient’s health status, vital signs, movement data and other infor­
key of AES can realize the key security management of symmetric mation, providing hospitals with more comprehensive and accurate big
cryptographic algorithm without the hospital data provider sending the data. Such as monitoring patients’ physical condition, improving hos­
key secretly before communication. This method has the advantages of pital medical efficiency, realizing efficient management of medical
high security, convenient key management, simple operation and sup­ equipment and resources, assisting medical research and decision-
porting integrity verification and identity authentication [10]. There­ making, providing personalized health management solutions, and
fore, this article proposes a hospital big data security and privacy realizing telemedicine services.
protection method based on hybrid encryption algorithm. The specific Big health data plays an important role in hospital big data, which
research ideas of this method are as follows: can enhance the precision and personalization of medical services,
optimize the allocation and utilization of medical resources, assist
(1) Collecting hospital big data includes three aspects: hospital medical research and innovation, improve the efficiency of hospital
medical business systems, mobile wearable devices, and big management and operation, as well as realize telemedicine services. Big
health data; health data, including patients’ family medical history, living habits,
(2) Using byte changes to compress hospital big data and achieve environmental factors and other information, can provide hospitals with
secure transmission of hospital big data; Then, the hospital sender more in-depth analysis and prediction, helping hospitals to better un­
encrypts the hospital big data using AES session key, encrypts the derstand the health status of patients and predict the risk of disease.
AES session key using ECC public key, calculates the hash value of Hospital big data includes multiple data sources from hospital
the medical big data using SHA-1, and signs the hash value using medical business systems, mobile wearable devices, and big health data.
ECC public key; By integrating these data, a comprehensive and multidimensional pa­
(3) The hospital recipient uses ECC private key to verify the signature tient profile can be formed, providing a richer information foundation
and decrypts the AES session key using the ECC private key to for medical decision-making. In the data collection stage, advanced data
achieve hospital big data security and privacy protection. anonymization techniques such as k-anonymity and l-diversity are used
(4) Finally, the performance of the encryption algorithm designed in to ensure that sensitive information is not leaked. At the same time,
this paper was verified using encryption and decryption time, differential privacy technology is used to perturb the data to enhance the
data integrity, and average correlation coefficient, and conclu­ privacy protection effect of the data.
sions were drawn. A dynamic compression algorithm based on byte variation is pro­
posed in the article, which automatically adjusts the compression ratio
2. Hybrid encryption algorithm for secure privacy protection of according to the characteristics of different types of data. This method
hospital big data not only reduces the amount of data transmission, but also lowers
transmission latency, improving the overall efficiency of the system.
2.1. Hospital big data collection During the compression process, embedded check codes (such as CRC)
are used to ensure the integrity of the compressed data, avoiding data
Hospital big data covers a variety of aspects, including hospital loss or damage caused by compression.
medical business systems, mobile wearable devices and big health data Therefore, it is necessary to strengthen the data security precautions,
in three aspects, hospital big data structure diagram shown in Fig. 1. clarify the authorization of data use, standardize data sharing manage­
Hospital medical business system is one of the main sources of hos­ ment, and raise the awareness of privacy protection of medical
pital big data, including electronic medical record data, medical image personnel in order to safeguard the privacy and security of hospital big
data, laboratory test data, vital sign data, etc. These data are generated data.
in the medical business system and reflect the patient’s condition,
diagnosis, treatment, nursing and other information. The database of
hospital medical business systems, such as HIS, CIS, LIS and EMR, is 2.2. Hospital big data security privacy protection realization
highly integrated to help medical personnel access the required clinical
data at any time on the system integration platform, so as to make timely 2.2.1. Hospital big data encryption
and effective clinical medical decisions. AES algorithm is a symmetric block cipher algorithm. It performs
iterative operation through Rijndael structure. The packet length is fixed

Fig. 1. Big data structure of hospital.

2
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

128b, and it can support 128b, 192b and 256b length keys to jointly 2.2.2. Decrypting big data in hospitals
determine its conversion. Fig. 2 shows the process of 128b AES algo­ (1) The selection of elliptic curves.
rithm encrypting hospital big data. After the first round of key addition In elliptic curve cryptosystem, the selection of the curve is the
and transformation, the first nine rounds are the same. One time, byte deciding factor that affects the security of the system. In the hospital big
replacement, row displacement, column mixing, and round key addition data security privacy protection method, this paper adopts the NIST
are used. The last round is different. The key length and grouping skip recommended curve, using the prime number p192 as the order of a finite
the column mixing operation. field. p192 is a compromise between safety and efficiency, and the cost of
The main function of the byte transformation is that by using S, the curve building is eliminated because the parameters are known.
box completes the mapping of bytes. The byte change can be used to (2) Generation of key.
compress and encrypt the hospital big data, providing a secure trans­ On the elliptic curve Ep (a, b) pick a point on the G(x, y), the order of G
mission method for hospital big data. The box S is a size of 16 × 16 is n (n is a large prime number), public G. In [1, n − 1] randomly deter­
Matrix, completing the mapping of 8b input to 8b output. The input mine an integer Ks , calculation Kp = Ks G, and Kp is an elliptic on curve
( )
value of low 4b is column label, and the input value of high 4b is bit row Ep (a, b). The hospital’s big data key Ks , Kp was determined. Ks is the
label, which is essentially a table lookup process. private key; Kp is the public key.
Row displacement is a left cyclic shift operation on the state matrix (3) ECC decryption process.
based on SubByte. The row displacement operation confuses and re­ Step 1: Hospital A selects an elliptic curve that Ep (a, b), and take a
places the hospital big data, increasing the security and confidentiality point on the elliptic curve, as a base point G(x, y); Hospital A can put
of the hospital big data [11]. After the transformation of AES algorithm elliptic curves Ep (a, b) and base points G(x, y) passed to Hospital B as a
of 128b, the first line of the state matrix will not shift, and the second to public parameter.
fourth lines will rotate 1 ~ 3B to the left, as shown in the following Step 2: Hospital A chooses a private key k, and generate a public key
formula: K = kG; private key k is the secret key of Hospital A, and the public key k
[ ] is the product of the base point G, which can be passed on to Hospital B.
sta[i][j] = sta[i] (j + i)4 , ij ∈ [0, 3] (1)
Step 3: Hospital A will transmit Ep (a, b) and points K, G to Hospital B;
Column obfuscation is the transformation of the columns of the state this information is used to encrypt the data at Hospital B’s end [15].
matrix, each column of the state matrix can be regarded as a polynomial Step 4: Upon receipt of the message, Hospital B encodes the hospital
( )
a(x) with the coefficient from a finite field GF 25 and times less than 4, big data to be transmitted Ep (a, b) up to point M, and generates a random
integer r(r < n); Hospital big data here could be patient data, medical
multiplied by the polynomial c(x) under mode x4 , as the following
records, etc. that need to be shared.
equation:
Step 5: Hospital B calculation point C1 = M + rK; C2 = rG; C1 and C2
c(x) = sta[i][j] × x3 + 01 × x2 + 01 × x + 02 (2) are encrypted hospital big data, which can be passed to hospital A as
( ) encrypted data.
b(x) = c(x) × a(x) mod x4 + 1 (3) Step 6: Hospital B will pass C1 and C2 to hospital A. C1 and C2 is used
for decrypting and verifying the integrity of hospital big data.
Written in matrix form B is:
Step 7: Upon Hospital A receipt of the information, calculates
⎡ ⎤ ⎡ ⎤ ⎡ ⎤
b0 02b0 03 01 01 a0 C1 − kC2 , the result is the point M.
⎢ b1 ⎥ ⎢ 01 02b1 03 01 ⎥ ⎢ a1 ⎥ Because C1 − kC2 = M + rK − k(rG) = M + rK − r(kG) = M.
B = b(x) × ⎣ ⎦ = ⎣
⎢ ⎥ ⎢ ⎥ = ⎢ ⎥ (4)
b2 01 01 02b2 03 ⎦ ⎣ a2 ⎦ After obtaining the point M, hospital A can decode the point M, then
b3 03 01 01 02b3 a3 the original hospital data (plaintext) is decoded.
Like row shifting, column obfuscation also ensures that the AES al­
gorithm obtains a high degree of diffusion properties after multiple 2.3. Digital signature based integrity protection of hospital big data
rounds of computation.
In the encryption process of hospital big data, the method of round Digital signatures can protect the integrity of the message and also
key addition is used [12]. The hospital big data is divided into multiple identify the source of the message and also authentication. Elliptic curve
data groups, and each data group has a corresponding round key. Then, signature algorithm is the DSA (digital signature algorithm) algorithm
the data groups are dissimilar to the corresponding round key to realize ported to the elliptic curve obtained [16]. In the digital signature to be
the encryption of hospital big data [13,14]. used in the security Hash function, in this paper, the scheme of the

Fig. 2. AES encryption flow chart.

3
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

signature algorithm using SHA-1 hash function, the algorithm is used in established, the signature is accepted, otherwise the signature is rejected
the ECC. to complete the verification of the identity of the hospital’s big data
(1) Parameter generation access.
Original signature group is comprised of (u1 , u2 , ..., un )n members,
including doctors, nurses, laboratory technicians, administrators and
2.4. Hospital big data security and privacy protection realization
other representatives of various hospital departments. The public key of
ui is Yui = xui G mod n(i = 1, 2, ...n), xui is the private key of ui . The
(1) Cryptographic realization process.
public and private keys represent the keys for data encryption and data
ECC encryption algorithm is used to generate the key pair required
decryption in hospital big data, respectively [17].
by both communication parties (communication between doctors and
Set GF(p) is a singular characteristic finite domains p ∕ = 2, 3, finite
patients or communication between hospitals and other medical in­
domain represents the mathematical domain of data representation and
stitutions), namely public key and private key, and the parameters of
processing in hospital big data, the elliptic curve defined on GF(p)
public key and elliptic curve are disclosed to the public; AES encryption
satisfying the equation y2 = x3 + ax + b ( mod p).
algorithm is used to generate encryption key [18]. Use the security hash
All points of the (x, y) ∈ GF(p)*GF(p) of (a, b ∈ GF(p) ),
function SHA-1 to calculate the medical data to obtain its hash value,
And meet the requirements 4a3 + 27b2 ∕ = 0 with the point O at in­ then use the sender’s ECC private key to digitally sign it, and then use
finity composite of an nonempty set. The elliptic curve discrete loga­ the AES session key to encrypt the medical big data (see Fig. 3).
rithm problem is: An elliptic curve of point the E and point P is known, (2) Decryption realization process.
generating a random integer d, easy to calculate Q = dP, but it would be After signature verification, the receiver ECC private key decrypts to
relatively difficult to calculate d by given Q and P. obtain the medical big data hash value [19]. The receiver uses the same
(2) Signature process. SHA-1 single column algorithm to calculate the decrypted medical big
The elliptic curve domain parameter T on the finite field GF(p) is data hash value, and compares the calculated hash value with the
defined as a hexagon T = (p, a, b, G, n, h). Of these, the meaning of p, a, b received hash value. If the same, the medical data is not tampered with
is the same as above. G is a basis point on an elliptic curve, the G ∕= O, so [20] (see Fig. 4).
that the minimum coefficient n of nG = O is called the order of the point Combining ECC and AES, ECC is used for key exchange and signature
G, recorded as n = ord(G); h is a one-way safe hash function h(⋅). Open verification, while AES is used for data encryption. By optimizing the
Hexadecimal T. key management process, reducing the number of key exchanges, and
User U in the interval [1, n − 1] randomly select an integer d, calcu­ further improving system performance. Dynamically generate keys
late Q = dG mod ⋅n, then the user U its public key is the point Q, the based on data types and transmission environments to improve the
private key is an integer d. The signature process of users U treat randomness and security of keys. Establish a real-time security moni­
signature information m is: toring system to detect and alert for abnormal behavior during data
User U in the zone [1, n − 1] randomly select k, calculate r = transmission. The specific encryption process of combining AES and ECC
XR mod n; in this article is as follows:
Calculate e = h(m),s = k + (r + e)d mod n, the signature pair of (r, Hospital sender:
s) is then sent to the verifier V.
Validation process of validator V: Receive the signature of the pair (r, (1) Generate session key: Generate a temporary AES session key.
s) of m, calculate e = h(m), and then calculate sG − (r ⊕ e)Y = (2) AES encryption: Encrypt hospital big data using AES session keys.
( ʹ ʹ)
XR , YR mod n. Calculate rʹ = XʹR mod n. If the equation r = rʹ (3) ECC encryption: Encrypt AES session keys using ECC public keys.

Fig. 3. Encryption flow chart.

4
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

Fig. 4. Decryption flow chart.

(4) Calculate Hash Value: Use SHA-1 to compute the hash value of
cov(x, y)
medical big data. w = √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ (5)
(5) ECC Signature: Use the ECC public key to sign the hash value. s(x)s(y)

In the formula, the w denotes the correlation coefficient of neigh­

Hospital recipient: boring data in the hospital big data; the cov(x, y) denotes the variance of
the two neighboring data x and y.
(1) Verify signature: Use ECC private key to verify the signature. (2) Record connections (RL).
(2) Decrypt AES session key: Use ECC private key to decrypt AES In order to measure the privacy protection effect of hospital big data
session key. based on the method of this paper, record linkage (RL), RL is used to
(3) AES decryption: Use AES session keys to decrypt hospital big represent the percentage of correctly matched original data records from
data. an anonymized big data set.
∑ ( )
3. Experimental analysis bʹj ∈G Pr bj
ʹ
RL = (6)
n
3.1. Experimental design ( )
n is the number of raw data records for hospital big data, the Pr bʹj is
In order to verify the feasibility and reliability of the hospital big data the record linking probability of the anonymized record, which is
security privacy protection method based on the hybrid encryption al­ expressed as follows.
gorithm proposed in this paper, the database of a large hospital is ⎧
⎨ 0, bj ∕
ʹ∈G
selected as the experimental object. The database is a relational data­ ( ) ⎪
base. The hospital database contains rich patient information, with a Pr bj =
ʹ
1 (7)
total data volume of 111.45 GB, including image information, text in­ ⎩ , bʹj ∈ G
⎪
|G|
formation, video information and other types of data. These data record
the patient’s medical history, diagnosis results, treatment plans and G is the set of original records with the smallest distance with bʹj , if
other information. At the same time, the database also contains the in­ the correct original record bʹj exist in G, then calculate the probability Pr
formation of doctors, nurses and other staff, as well as related medical ( ) ( )
management information, such as drug management, medical equip­ bʹj = |G|
1
of bʹj in the guess G. Otherwise. Pr bʹj = 0. Record links
ment management, etc. The method in this paper is used to store and measure actual privacy in terms of privacy attacks, and thus, the lower
manage the information in the database safely, and two traditional the value RL, the lower the probability of privacy leakage, and the better
methods are selected for comparison. These two traditional methods are the privacy of data released after anonymization.
security and privacy protection methods based on blockchain and
extensible markup language respectively.
3.3. Analysis of experimental results

3.2. Design of experimental indicators 3.3.1. Comparison of correlation coefficients of neighboring data
The results of the statistical correlation coefficients of the three
(1) The correlation coefficient of neighboring data. neighboring data are shown in Table 1.
The experiments use the correlation coefficient of neighboring data As can be seen from Table 1, the correlation coefficient of neigh­
in the encrypted hospital big data files as the performance evaluation boring data in hospital big data using the security and privacy protection
index of the above three privacy protection methods. The correlation method in this paper is low, the minimum value of correlation coeffi­
coefficient of neighboring data in the encrypted hospital big data can cient is 0.024, the maximum value of correlation coefficient is 0.120,
reflect the encryption effect of hospital big data, the lower the correla­ and the average correlation coefficient is 0.0576, which is low and
tion coefficient is, the better the disruption effect of hospital big data is, basically negligible, which indicates that the method in this paper has a
and the more difficult it is to decipher the encrypted information, which better effect of disrupting the big data in hospitals with a higher level of
means that the security of hospital big data is higher. The experiment security. The correlation coefficients of adjacent data in hospital big
randomly selects 10 encrypted big data, and calculates the correlation data are relatively high when using the blockchain-based security and
coefficient of neighboring data in hospital big data, that is. privacy protection method and the extensible markup language-based

5
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

Table 1 hospital and encrypts the privacy of the patients, the ciphertext trans­
Correlation coefficients of adjacent data in hospital big data of the three mission is directly interacted with the file sharing method, and
methods. encryption and decryption tests are carried out, and basic information
Hospital big data Textual Blockchain Extensible markup such as the birth of the patients and the purchase of the drugs are shown
grouping serial number method approach language method in Table 2 and Table 3 and the information is inputted into the
1 0.049 0.701 0.649 encryption and decryption test system of the hospital’s big data is
2 0.079 0.759 0.891 encrypted, and the experimental results are shown in Fig. 6.
3 0.039 0.831 0.756 As can be seen from Fig. 6, this paper encrypts the data that learns
4 0.071 0.739 0.525
the basic information such as the birth of the patient and the purchase of
5 0.031 0.831 0.754
6 0.120 0.919 0.689 medication based on the method of ECC + AES hybrid encryption al­
7 0.049 0.769 0.746 gorithm. The algorithm successfully encrypted 512 MB of data with its
8 0.056 0.554 0.839 excellent performance and processing capability. The process was
9 0.058 0.645 0.569 completed in only 87 ms. This not only proves the efficiency of the al­
10 0.024 0.365 0.784
gorithm, but also meets the requirements of hospitals’ big data pro­
cessing speed. This encryption process ensures the security of patient
security and privacy protection method. The maximum values of the data and prevents unauthorized access and leakage. At the same time,
correlation coefficients of the blockchain-based method and the exten­ due to the superior processing speed, the method can also effectively
sible markup language-based method in the present experiment are cope with the processing needs of large amounts of data. This provides a
0.919 and 0.839, respectively; the minimum values of the correlation secure and efficient data processing solution for hospitals to better
coefficients are 0.365 and 0.525, and the average values of the corre­ manage and protect patient information. Overall, the method in this
lation coefficients are 0.6813, 0.713 and 0.725, respectively, which are paper successfully solves the problems of security and processing speed
much higher than the method of this paper. The experimental results of patient data in hospital big data by using ECC + AES hybrid
show that the method in this paper is better than the blockchain method encryption algorithm. This method not only ensures the security of data,
and the scalable markup language method in encrypting big data in but also meets the demand of data processing speed, which provides a
hospitals. strong support for the development of hospital informationization.
In summary, this method based on ECC + AES hybrid encryption
3.3.2. Record connections (RL) comparison algorithm is applied to hospital big data and encrypts the patient’s
In order to verify the privacy of this paper’s method, the experi­ private information, which can effectively protect the patient’s privacy
mental data are selected from the hospital database, and different and ensure the security and confidentiality of the data. At the same time,
quantities of data are used as the experimental sample space to compare the encryption algorithm in this paper has efficient encryption and
the privacy of this paper’s method, the blockchain method and the decryption performance, which can meet the needs of practical
extensible markup language method, and the experimental results are applications.
shown in Fig. 5.
As can be seen from Fig. 5, the RL values of all three methods in­ 3.3.4. Data integrity verification
crease as the hospital dataset becomes larger. However, compared with To further verify the security effect of data encryption, the hybrid
the RL curves of the other two methods, it can be seen that the RL curves encryption algorithm (AES and ECC encryption algorithms) proposed in
of this paper’s method are smoother and the RL values are lower, which this paper is compared with the currently mainstream algorithms: AES
have higher privacy. encryption algorithm and ECC encryption algorithm. SHA-256 hash
values are calculated for the original data, encrypted data, and hybrid
3.3.3. Security and privacy protective fruit encrypted data, respectively. The hash values of the data before and
In order to verify the effectiveness of the method of this paper in the after encryption are compared to ensure that the data has not been
hospital big data security privacy protection, the ECC + AES hybrid tampered with during the encryption process, and the size changes of
encryption algorithm of this paper is applied to the big data of this the data files before and after encryption are recorded.
According to Table 5, the data size of AES method increases by about
16 bytes per record, ECC method increases by about 32 bytes per record,
and hybrid encryption algorithm increases by about 8 bytes per record.
The hash values of AES, ECC, and hybrid encryption algorithms are
consistent, indicating that data integrity is guaranteed. The AES method
is suitable for scenarios that require fast encryption and decryption,
especially when dealing with large amounts of data. The increase in data
size is relatively small, but key management is more complex. Although
the ECC method has a slower encryption speed, it provides higher se­
curity, especially when the key length is shorter. The data size has
increased significantly, making it suitable for scenarios with extremely

Table 2
Basic information of patients.
Patient ID Date of birth Area code Sex

P1 7/18/87 19,130 F
P2 2/17/86 90,031 M
P3 5/07/77 94,107 M
P4 1/5/76 18,181 F
P5 8/4/82 94,177 M
P6 3/9/79 90,101 M
P7 4/5/65 15,203 M
P8 2/6/82 15,247 F
Fig. 5. Influence of data set size change on privacy.

6
W. Li and Q. Huang Egyptian Informatics Journal 28 (2024) 100569

Table 3 encryption of patient privacy.

Information of drugs purchased by patients.
Drug ID Drug name Subclass Main category CRediT authorship contribution statement
D1 Citalopram SSRIs Antidepressant
D2 Phenelhydrazine Monoamine oxidation inhibitors Antidepressant Wei Li: Writing – original draft, Validation, Software, Project
D3 Erythromycin Macrolides Antibiotic administration, Conceptualization. Qian Huang: Writing – review &
D4 Selegiline Monoamine oxidation inhibitors Antidepressant editing, Resources, Methodology, Formal analysis, Data curation.
D5 Azithromycin Macrolides Antibiotic
D6 Cephalosporin Beta-lactam Antibiotic
D7 Penicillin Beta-lactam Antibiotic Declaration of competing interest
D8 Fluoxetine SSRIs Antidepressant

The authors declare that they have no known competing financial

interests or personal relationships that could have appeared to influence
the work reported in this paper.

References

[1] Zhang SB, Yuan LJ, Mao XJ, Zhu GM. Privacy protection method for k-modes
clustering data with local differential privacy. Acta Electron Sin 2022;50(9):
2181–8.
[2] Ebert I, Wildhaber I, Adams-Prassl J. Big data in the workplace: privacy due
diligence as a human rights-based approach to employee privacy protection. Big
Data Soc 2021;8(1):474–80.
[3] Gupta EAS. Privacy protection in big data: a survey. Turk J Comput Math Educat
(TURCOMAT) 2021;12(2):562–8.
[4] Sharma EASK. A framework of big data as service platform for access control &
privacy protection using blockchain network. Turk J Comput Math Educat
(TURCOMAT) 2021;12(11):476–85.
[5] Makri EL, Georgiopoulou Z, Lambrinoudakis C. Utilizing a privacy impact
assessment method using metrics in the healthcare sector. Inform Comput Secur
2020;28(4):503–29.
[6] Pawar AB, Ghumbre SU, Jogdand RM. Privacy preserving model-based
authentication and data security in cloud computing. Internat J Pervasive Comput
Commun 2023;19(2):173–90.
[7] Suma MR, Madhumathy P. Brakerski-gentry-vaikuntanathan fully homomorphic
Fig. 6. Main page of the encryption and decryption test system. encryption cryptography for privacy preserved data access in cloud assisted
internet of things services using glow-worm swarm optimization. Trans Emerg
Telecommun Technol 2022;33(12):4641–61.
[8] Jaffar D, Sajid A, Jahangeer A, Wagan A. Cloud based e-health care data security
Table 5
and privacy by using advanced encryption standard and watermarking technique.
Data integrity verification. Malay E Comm J 2020;4(1):1–4.
Encryption method Change in data size (bytes/ Hash value [9] Kanna GP, Vasudevan V. An improved privacy aware secure multi-cloud model
with proliferate elgamal encryption for big data storage. Int J Inf Comput Secur
record) consistency
2022;17(2):1–20.
AES +16 consistent [10] Li P, Xu C, Xu H. Data privacy protection algorithm based on redundant slice
ECC +32 consistent technology in wireless sensor networks. Internat J Inform Secur Privacy (IJISP)
Hybrid encryption +8 consistent 2021;15(1):190–212.
algorithm [11] Yu J. Research on cloud computing user privacy protection based on dynamic
adaptive ant colony algorithm. Internat J Autonom Adapt Commun Syst: IJAACS
2020;13(2):178–94.
[12] Prabhu J, Dass AS. Hybrid coherent encryption scheme for multimedia big data
high security requirements. The hybrid encryption algorithm combines management using cryptographic encryption methods. Int J Grid Util Comput
the advantages of AES and ECC, maintaining fast encryption and 2020;11(4):496–508.
decryption speeds while improving the security of key management. The [13] Lee RR, Peters S, Mcdonagh JE, Farre A, Cordingley L, Rapley T. Data protection,
information governance and the potential erosion of ethnographic methods in
increase in data size is minimal, making it suitable for application sce­ health care. Sociol Health Illn 2022;44(1):211–7.
narios that require a balance between speed and security. [14] Khin NN, Win T. A novel hybrid encryption method based on honey encryption and
advanced DNA encoding scheme in key generation. J Comput Commun 2022;10
(9):22–36.
4. Conclusion [15] Ampavathi A, Pradeepini G, Saradhi TV. Optimized deep learning-enabled hybrid
logistic piece-wise chaotic map for secured medical data storage system. Int J Inf
This article designs and studies a hospital big data security and pri­ Technol Decis Mak 2023;22(5):1743–75.
[16] Uma MS, Vasanthanayaki C. Secure medical health care content protection system
vacy protection method based on hybrid encryption algorithms to pre­ (SMCPS) with watermark detection for multi cloud computing environment.
vent hospital big data from being eavesdropped and tampered with. This Multimed Tools Appl 2020;79(6):4075–97.
article proposes a hybrid encryption algorithm based on ECC and AES, [17] Hui KJ. User privacy protection behavior and information sharing in mobile health
application. Internat J Adv Trends Comput Sci Eng 2020;9(4):5250–8.
which encrypts sessions through the hybrid encryption algorithm and
[18] Searle R, Gururaj P. Protecting patient confidentiality in the internet of medical
uses ECC technology to achieve decryption, completing the security things through confidential computing. J Data Protect Privacy 2023;5(4):347–62.
protection of hospital big data. The results indicate that the method [19] Sun LH. Hadoop-based smart grid time series big data processing method. Comput
proposed in this article is suitable for processing hospital big data and Simul 2020;37(12):67–71.
[20] Awotunde JB, Oladipo ID, Abdulraheem M, Balogun GB, Tomori AR. An iomt-
has a faster running time. The method proposed in this article has been based steganography model for securing medical information. Int J Healthc
applied to a hospital big data platform and successfully achieved Technol Manag 2022;19(4):218–36.