Mutual authentication

Last updated

Mutual authentication or two-way authentication (not to be confused with two-factor authentication) refers to two parties authenticating each other at the same time in an authentication protocol. It is a default mode of authentication in some protocols (IKE, SSH) and optional in others (TLS).

Contents

Mutual authentication is a desired characteristic in verification schemes that transmit sensitive data, in order to ensure data security. [1] [2] [3] Mutual authentication can be accomplished with two types of credentials: usernames and passwords, and public key certificates.

Mutual authentication is often employed in the Internet of Things (IoT). Writing effective security schemes in IoT systems can become challenging, especially when schemes are desired to be lightweight and have low computational costs. Mutual authentication is a crucial security step that can defend against many adversarial attacks, [4] which otherwise can have large consequences if IoT systems (such as e-Healthcare servers) are hacked. In scheme analyses done of past works, a lack of mutual authentication had been considered a weakness in data transmission schemes. [5]

Process steps and verification

Schemes that have a mutual authentication step may use different methods of encryption, communication, and verification, but they all share one thing in common: each entity involved in the communication is verified. If Alice wants to communicate with Bob, they will both authenticate the other and verify that it is who they are expecting to communicate with before any data or messages are transmitted. A mutual authentication process that exchanges user IDs may be implemented as follows:[ citation needed ]

  1. Alice sends an encrypted message to Bob to show that Alice is a valid user.
  2. Bob verifies message:
    1. Bob checks the format and timestamp. If either is incorrect or invalid, the session is aborted.
    2. The message is then decrypted with Bob's secret key, giving Alice's ID.
      1. Bob checks if the message matches a valid user. If not, the session is aborted.
  3. Bob sends Alice a message back to show that Bob is a valid user.
  4. Alice verifies the message:
    1. Alice checks the format and timestamp. If either is incorrect or invalid, the session is aborted.
    2. Then, the message is decrypted with Alice's secret key, giving Bob's ID.
      1. Alice checks if the message matches a valid user. If not, the session is aborted.
  5. At this point, both parties are verified to be who they claim to be and safe for the other to communicate with. Lastly, Alice and Bob will create a shared secret key so that they can continue communicating in a secure manner.

To verify that mutual authentication has occurred successfully, Burrows-Abadi-Needham logic (BAN logic) is a well regarded and widely accepted method to use, because it verifies that a message came from a trustworthy entity. BAN logic first assumes an entity is not to be trusted, and then will verify its legality. [1] [2] [6] [7]

Defenses

Mutual authentication supports zero trust networking because it can protect communications against adversarial attacks, [8] notably:

Man-in-the-middle attack
Man-in-the-middle (MITM) attacks are when a third party wishes to eavesdrop or intercept a message, and sometimes alter the intended message for the recipient. The two parties openly receive messages without verifying the sender, so they do not realize an adversary has inserted themselves into the communication line. Mutual authentication can prevent MITM attacks because both the sender and recipient verify each other before sending them their message keys, so if one of the parties is not verified to be who they claim they are, the session will end. [9]
Replay attack
A replay attack is similar to a MITM attack in which older messages are replayed out of context to fool the server. However, this does not work against schemes using mutual authentication [10] because timestamps are a verification factor that are used in the protocols. [11] [12] If the change in time is greater than the maximum allowed time delay, the session will be aborted. [12] Similarly, messages can include a randomly generated number to keep track of when a message was sent. [11]
Spoofing attack
Spoofing attacks rely on using false data to pose as another user in order to gain access to a server or be identified as someone else. Mutual authentication can prevent spoofing attacks because the server will authenticate the user as well, and verify that they have the correct session key before allowing any further communication and access. [12]
Impersonation attacks
When each party authenticates the other, they send each other a certificate that only the other party knows how to unscramble, verifying themselves as a trusted source. In this way, adversaries cannot use impersonation attacks because they do not have the correct certificate to act as if they are the other party. [6]

Mutual authentication also ensures information integrity because if the parties are verified to be the correct source, then the information received is reliable as well. [6]

mTLS

By default the TLS protocol only proves the identity of the server to the client using X.509 certificates, and the authentication of the client to the server is left to the application layer. TLS also offers client-to-server authentication using client-side X.509 authentication. [13] As it requires provisioning of the certificates to the clients and involves less user-friendly experience, it's rarely used in end-user applications.

Mutual TLS authentication (mTLS) is more often used in business-to-business (B2B) applications, where a limited number of programmatic and homogeneous clients are connecting to specific web services, the operational burden is limited, and security requirements are usually much higher as compared to consumer environments.

mTLS is also used in microservices-based applications based on runtimes such as Dapr, via systems like SPIFFE. [8]

Lightweight schemes vs. secured schemes

While lightweight schemes and secure schemes are not mutually exclusive, adding a mutual authentication step to data transmissions protocols can often increase performance runtime and computational costs. [2] This can become an issue for network systems that cannot handle large amounts of data or those that constantly have to update for new real-time data (e.g. location tracking, real-time health data). [2] [10]

Thus, it becomes a desired characteristic of many mutual authentication schemes to have lightweight properties (e.g. have a low memory footprint) in order to accommodate the system that is storing a lot of data. [4] Many systems implement cloud computing, which allows quick access to large amounts of data, but sometimes large amounts of data can slow down communication. Even with edge-based cloud computing, which is faster than general cloud computing due to a closer proximity between the server and user, [6] lightweight schemes allow for more speed when managing larger amounts of data. One solution to keep schemes lightweight during the mutual authentication process is to limit the number of bits used during communication. [4]

Applications that solely rely on device-to-device (D2D) communication, where multiple devices can communicate locally in close proximities, removes the third party network. This in turn can speed up communication time. [14] However, the authentication still occurs through insecure channels, so researchers believe it is still important to ensure mutual authentication occurs in order to keep a secure scheme. [14]

Schemes may sacrifice a better runtime or storage cost when ensuring mutual authentication in order to prioritize protecting the sensitive data. [2] [12]

Password-based schemes

In mutual authentication schemes that require a user's input password as part of the verification process, there is a higher vulnerability to hackers because the password is human-made rather than a computer-generated certificate. While applications could simply require users to use a computer-generated password, it is inconvenient for people to remember. User-made passwords and the ability to change one's password are important for making an application user-friendly, [15] so many schemes work to accommodate the characteristic. Researchers note that a password based protocol with mutual authentication is important because user identities and passwords are still protected, as the messages are only readable to the two parties involved. [16]

However, a negative aspect about password-based authentication is that password tables can take up a lot of memory space. [15] One way around using a lot of memory during a password-based authentication scheme is to implement one-time passwords (OTP), which is a password sent to the user via SMS or email. OTPs are time-sensitive, which means that they will expire after a certain amount of time and that memory does not need to be stored. [17]

Multi-factor authentication

Recently, more schemes have higher level authentication than password based schemes. While password-based authentication is considered as "single-factor authentication," schemes are beginning to implement smart card (two-factor) [15] or biometric-based (three-factor) authentication schemes. Smart cards are simpler to implement and easy for authentication, but still have risks of being tampered with. [15] Biometrics have grown more popular over password-based schemes because it is more difficult to copy or guess session keys when using biometrics, [7] but it can be difficult to encrypt noisy data. [17] Due to these security risks and limitations, schemes can still employ mutual authentication regardless of how many authentication factors are added. [7]

Certificate based schemes and system applications

Mutual authentication is often found in schemes employed in the Internet of Things (IoT), where physical objects are incorporated into the Internet and can communicate via IP address. [11] Authentication schemes can be applied to many types of systems that involve data transmission. [14] As the Internet's presence in mechanical systems increases, writing effective security schemes for large numbers of users, objects, and servers can become challenging, especially when needing schemes to be lightweight and have low computational costs. Instead of password-based authentication, devices will use certificates to verify each other's identities.

Radio networks

Mutual authentication can be satisfied in radio network schemes, where data transmissions through radio frequencies are secure after verifying the sender and receiver. [12] [18]

Radio frequency identification (RFID) tags are commonly used for object detection, which many manufacturers are implementing into their warehouse systems for automation. [19] This allows for a faster way to keep up with inventory and track objects. However, keeping track of items in a system with RFID tags that transmit data to a cloud server increases the chances of security risks, as there are now more digital elements to keep track of. [19] A three way mutual authentication can occur between RFID tags, the tag readers, and the cloud network that stores this data in order to keep RFID tag data secure and unable to be manipulated. [19]

Similarly, an alternate RFID tag and reader system that assigns designated readers to tags has been proposed for extra security and low memory cost. [20] Instead of considering all tag readers as one entity, only certain readers can read specific tags. With this method, if a reader is breached, it will not affect the whole system. Individual readers will communicate with specific tags during mutual authentication, which runs in constant time as readers use the same private key for the authentication process.

Many e-Healthcare systems that remotely monitor patient health data use wireless body area networks (WBAN) that transmit data through radio frequencies. [12] This is beneficial for patients that should not be disturbed while being monitored, and can reduced the workload for medical worker and allow them to focus on the more hands-on jobs. However, a large concern for healthcare providers and patients about using remote health data tracking is that sensitive patient data is being transmitted through unsecured channels, [9] so authentication occurs between the medical body area network user (the patient), the Healthcare Service Provider (HSP) and the trusted third party.

Cloud based computing

e-Healthcare clouds are another way to store patient data collected remotely. [2] Clouds are useful for storing large amounts of data, such as medical information, that can be accessed by many devices whenever needed. Telecare Medical Information Systems (TMIS), an important way for medical patients to receive healthcare remotely, can ensure secured data with mutual authentication verification schemes. [14] Blockchain is one way that has been proposed to mutually authenticate the user to the database, by authenticating with the main mediBchain node and keeping patient anonymity. [21]

Fog-cloud computing is a networking system that can handle large amounts of data, but still has limitations regarding computational and memory cost. [22] Mobile edge computing (MEC) is considered to be an improved, more lightweight fog-cloud computing networking system, [22] and can be used for medical technology that also revolves around location-based data. Due to the large physical range required of locational tracking, 5G networks can send data to the edge of the cloud to store data. An application like smart watches that track patient health data can be used to call the nearest hospital if the patient shows a negative change in vitals. [6]

Fog node networks can be implemented in car automation, keeping data about the car and its surrounding states secure. By authenticating the fog nodes and the vehicle, vehicular handoff becomes a safe process and the car’s system is safe from hackers. [10]

Machine to machine verification

Many systems that do not require a human user as part of the system also have protocols that mutually authenticate between parties. In unmanned aerial vehicle (UAV) systems, a platform authentication occurs rather than user authentication. [2] Mutual authentication during vehicle communication prevents one vehicle's system from being breached, which can then affect the whole system negatively. For example, a system of drones can be employed for agriculture work and cargo delivery, but if one drone were to be breached, the whole system has the potential to collapse. [2]

Related Research Articles

<span class="mw-page-title-main">HTTPS</span> Extension of the HTTP communications protocol to support TLS encryption

Hypertext Transfer Protocol Secure (HTTPS) is an extension of the Hypertext Transfer Protocol (HTTP). It uses encryption for secure communication over a computer network, and is widely used on the Internet. In HTTPS, the communication protocol is encrypted using Transport Layer Security (TLS) or, formerly, Secure Sockets Layer (SSL). The protocol is therefore also referred to as HTTP over TLS, or HTTP over SSL.

Kerberos is a computer-network authentication protocol that works on the basis of tickets to allow nodes communicating over a non-secure network to prove their identity to one another in a secure manner. Its designers aimed it primarily at a client–server model, and it provides mutual authentication—both the user and the server verify each other's identity. Kerberos protocol messages are protected against eavesdropping and replay attacks.

<span class="mw-page-title-main">Public-key cryptography</span> Cryptographic system with public and private keys

Public-key cryptography, or asymmetric cryptography, is the field of cryptographic systems that use pairs of related keys. Each key pair consists of a public key and a corresponding private key. Key pairs are generated with cryptographic algorithms based on mathematical problems termed one-way functions. Security of public-key cryptography depends on keeping the private key secret; the public key can be openly distributed without compromising security.

<span class="mw-page-title-main">Authentication</span> Act of proving an assertion, often the identity of a computer system user

Authentication is the act of proving an assertion, such as the identity of a computer system user. In contrast with identification, the act of indicating a person or thing's identity, authentication is the process of verifying that identity. It might involve validating personal identity documents, verifying the authenticity of a website with a digital certificate, determining the age of an artifact by carbon dating, or ensuring that a product or document is not counterfeit.

In law, non-repudiation is a situation where a statement's author cannot successfully dispute its authorship or the validity of an associated contract. The term is often seen in a legal setting when the authenticity of a signature is being challenged. In such an instance, the authenticity is being "repudiated".

Password Authentication Protocol (PAP) is a password-based authentication protocol used by Point-to-Point Protocol (PPP) to validate users. PAP is specified in RFC 1334.

In cryptography and computer security, a man-in-the-middle (MITM) attack or on-path attack is a cyberattack where the attacker secretly relays and possibly alters the communications between two parties who believe that they are directly communicating with each other, as the attacker has inserted themselves between the two parties.

Transport Layer Security (TLS) is a cryptographic protocol designed to provide communications security over a computer network. The protocol is widely used in applications such as email, instant messaging, and voice over IP, but its use in securing HTTPS remains the most publicly visible.

<span class="mw-page-title-main">Public key infrastructure</span> System that can issue, distribute and verify digital certificates

A public key infrastructure (PKI) is a set of roles, policies, hardware, software and procedures needed to create, manage, distribute, use, store and revoke digital certificates and manage public-key encryption. The purpose of a PKI is to facilitate the secure electronic transfer of information for a range of network activities such as e-commerce, internet banking and confidential email. It is required for activities where simple passwords are an inadequate authentication method and more rigorous proof is required to confirm the identity of the parties involved in the communication and to validate the information being transferred.

Wi-Fi Protected Access (WPA), Wi-Fi Protected Access 2 (WPA2), and Wi-Fi Protected Access 3 (WPA3) are the three security certification programs developed after 2000 by the Wi-Fi Alliance to secure wireless computer networks. The Alliance defined these in response to serious weaknesses researchers had found in the previous system, Wired Equivalent Privacy (WEP).

An authentication protocol is a type of computer communications protocol or cryptographic protocol specifically designed for transfer of authentication data between two entities. It allows the receiving entity to authenticate the connecting entity as well as authenticate itself to the connecting entity by declaring the type of information needed for authentication as well as syntax. It is the most important layer of protection needed for secure communication within computer networks.

A replay attack is a form of network attack in which valid data transmission is maliciously or fraudulently repeated or delayed. This is carried out either by the originator or by an adversary who intercepts the data and re-transmits it, possibly as part of a spoofing attack by IP packet substitution. This is one of the lower-tier versions of a man-in-the-middle attack. Replay attacks are usually passive in nature.

Internet security is a branch of computer security. It encompasses the Internet, browser security, web site security, and network security as it applies to other applications or operating systems as a whole. Its objective is to establish rules and measures to use against attacks over the Internet. The Internet is an inherently insecure channel for information exchange, with high risk of intrusion or fraud, such as phishing, online viruses, trojans, ransomware and worms.

Extensible Authentication Protocol (EAP) is an authentication framework frequently used in network and internet connections. It is defined in RFC 3748, which made RFC 2284 obsolete, and is updated by RFC 5247. EAP is an authentication framework for providing the transport and usage of material and parameters generated by EAP methods. There are many methods defined by RFCs, and a number of vendor-specific methods and new proposals exist. EAP is not a wire protocol; instead it only defines the information from the interface and the formats. Each protocol that uses EAP defines a way to encapsulate by the user EAP messages within that protocol's messages.

In cryptography, a password-authenticated key agreement (PAK) method is an interactive method for two or more parties to establish cryptographic keys based on one or more party's knowledge of a password.

In cryptography, forward secrecy (FS), also known as perfect forward secrecy (PFS), is a feature of specific key-agreement protocols that gives assurances that session keys will not be compromised even if long-term secrets used in the session key exchange are compromised, limiting damage. For HTTPS, the long-term secret is typically the private key of the server. Forward secrecy protects past sessions against future compromises of keys or passwords. By generating a unique session key for every session a user initiates, the compromise of a single session key will not affect any data other than that exchanged in the specific session protected by that particular key. This by itself is not sufficient for forward secrecy which additionally requires that a long-term secret compromise does not affect the security of past session keys.

Multi-factor authentication is an electronic authentication method in which a user is granted access to a website or application only after successfully presenting two or more pieces of evidence to an authentication mechanism. MFA protects personal data—which may include personal identification or financial assets—from being accessed by an unauthorized third party that may have been able to discover, for example, a single password.

In cryptography, the Salted Challenge Response Authentication Mechanism (SCRAM) is a family of modern, password-based challenge–response authentication mechanisms providing authentication of a user to a server. As it is specified for Simple Authentication and Security Layer (SASL), it can be used for password-based logins to services like LDAP, HTTP, SMTP, POP3, IMAP and JMAP (e-mail), XMPP (chat), or MongoDB and PostgreSQL (databases). For XMPP, supporting it is mandatory.

Network eavesdropping, also known as eavesdropping attack, sniffing attack, or snooping attack, is a method that retrieves user information through the internet. This attack happens on electronic devices like computers and smartphones. This network attack typically happens under the usage of unsecured networks, such as public wifi connections or shared electronic devices. Eavesdropping attacks through the network is considered one of the most urgent threats in industries that rely on collecting and storing data. Internet users use eavesdropping via the Internet to improve information security.

Application Layer Transport Security (ALTS) is a Google-developed authentication and transport encryption system used for securing Remote Procedure Call (RPC) within Google machines. Google started its development in 2007, as a tailored modification of TLS.

References

  1. 1 2 Chen, Yulei; Chen, Jianhua (2020). "A secure three-factor-based authentication with key agreement protocol for e-Health clouds". The Journal of Supercomputing. 77 (4): 3359–3380. doi:10.1007/s11227-020-03395-8. ISSN   0920-8542. S2CID   221146362.
  2. 1 2 3 4 5 6 7 8 Chen, Liquan; Qian, Sijie; Lim, Ming; Wang, Shihui (2018). "An enhanced direct anonymous attestation scheme with mutual authentication for network-connected UAV communication systems". China Communications. 15 (5): 61–76. doi:10.1109/CC.2018.8387987. S2CID   49333360.
  3. ↑ Sannasiddappa, Thippeswamy H.; Costabile, Adele; Gibson, Glenn R.; Clarke, Simon R. (2011). "The Influence of Staphylococcus aureus on Gut Microbial Ecology in an in Vitro Continuous Culture Human Colonic Model System". PLOS ONE. 6 (8): e23227. Bibcode:2011PLoSO...623227S. doi: 10.1371/journal.pone.0023227 . PMC   3153491 . PMID   21858036.
  4. 1 2 3 Jan, Mian Ahmad; Khan, Fazlullah; Alam, Muhammad; Usman, Muhammad (2019). "A payload-based mutual authentication scheme for Internet of Things". Future Generation Computer Systems. 92: 1028–1039. doi:10.1016/j.future.2017.08.035. hdl: 10453/117906 . S2CID   57380203.
  5. ↑ Amin, Ruhul; Islam, SK Hafizul; Vijayakumar, Pandi; Khan, Muhammad Khurram; Chang, Victor (2018). "A robust and efficient bilinear pairing based mutual authentication and session key verification over insecure communication". Multimedia Tools and Applications. 77 (9): 11041–11066. doi:10.1007/s11042-017-4996-z. S2CID   13674284.
  6. 1 2 3 4 5 Chen, Chin-Ling; Chiang, Mao-Lun; Hsieh, Hui-Ching; Liu, Ching-Cheng; Deng, Yong-Yuan (2020). "A Lightweight Mutual Authentication with Wearable Device in Location-Based Mobile Edge Computing". Wireless Personal Communications. 113: 575–598. doi:10.1007/s11277-020-07240-2. S2CID   218934756.
  7. 1 2 3 Sahoo, Shreeya Swagatika; Mohanty, Sujata; Majhi, Banshidhar (2020). "Improved Biometric-Based Mutual Authentication and Key Agreement Scheme Using ECC". Wireless Personal Communications. 111 (2): 991–1017. doi:10.1007/s11277-019-06897-8. S2CID   208125038.
  8. 1 2 "Mutual TLS: Securing Microservices in Service Mesh". The New Stack. 2021-02-01. Retrieved 2021-02-20.
  9. 1 2 Sasikaladevi, N.; Malathi, D. (2019). "Energy Efficient Lightweight Mutual Authentication Protocol (REAP) for MBAN Based on Genus-2 Hyper-Elliptic Curve". Wireless Personal Communications. 109 (4): 2471–2488. doi:10.1007/s11277-019-06693-4. S2CID   204084523.
  10. 1 2 3 Dewanta, Favian; Mambo, Masahiro (2019). "A Mutual Authentication Scheme for Secure Fog Computing Service Handover in Vehicular Network Environment". IEEE Access. 7: 103095–103114. Bibcode:2019IEEEA...7j3095D. doi: 10.1109/ACCESS.2019.2931217 . S2CID   199509951.
  11. 1 2 3 Melki, Reem; Noura, Hassan N.; Chehab, Ali (2020). "Lightweight multi-factor mutual authentication protocol for IoT devices". International Journal of Information Security. 19 (6): 679–694. doi:10.1007/s10207-019-00484-5. S2CID   209340123.
  12. 1 2 3 4 5 6 Narwal, Bhawna; Mohapatra, Amar Kumar (2020). "SEEMAKA: Secured Energy-Efficient Mutual Authentication and Key Agreement Scheme for Wireless Body Area Networks". Wireless Personal Communications. 113 (4): 1985–2008. doi:10.1007/s11277-020-07304-3. S2CID   216529906.
  13. ↑ Dierks, Tim (August 2008). "The Transport Layer Security (TLS) Protocol Version 1.2". tools.ietf.org. Retrieved 2016-04-22.
  14. 1 2 3 4 g. Lopes, Ana Paula; Gondim, Paulo R. L. (2020). "Mutual Authentication Protocol for D2D Communications in a Cloud-Based E-Health System". Sensors. 20 (7): 2072. Bibcode:2020Senso..20.2072G. doi: 10.3390/s20072072 . PMC   7181216 . PMID   32272675.
  15. 1 2 3 4 Karuppiah, Marimuthu; Saravanan, R. (2015). "Cryptanalysis and an Improvement of New Remote Mutual Authentication Scheme using Smart Cards". Journal of Discrete Mathematical Sciences and Cryptography. 18 (5): 623–649. doi:10.1080/09720529.2015.1013693. S2CID   62591965.
  16. ↑ Karuppiah, Marimuthu; Das, Ashok Kumar; Li, Xiong; Kumari, Saru; Wu, Fan; Chaudhry, Shehzad Ashraf; Niranchana, R. (2019). "Secure Remote User Mutual Authentication Scheme with Key Agreement for Cloud Environment". Mobile Networks and Applications. 24 (3): 1046–1062. doi:10.1007/s11036-018-1061-8. S2CID   64720667.
  17. 1 2 Sharma, Mohit Kr; Nene, Manisha J. (2020). "Two-factor authentication using biometric based quantum operations". Security and Privacy. 3 (3). doi: 10.1002/spy2.102 .
  18. ↑ Choudhary, Karanjeet; Gaba, Gurjot Singh; Butun, Ismail; Kumar, Pardeep (2020). "MAKE-IT—A Lightweight Mutual Authentication and Key Exchange Protocol for Industrial Internet of Things". Sensors. 20 (18): 5166. Bibcode:2020Senso..20.5166C. doi: 10.3390/s20185166 . PMC   7570918 . PMID   32927788.
  19. 1 2 3 Anandhi, S.; Anitha, R.; Sureshkumar, Venkatasamy (2020). "An Authentication Protocol to Track an Object with Multiple RFID Tags Using Cloud Computing Environment". Wireless Personal Communications. 113 (4): 2339–2361. doi:10.1007/s11277-020-07330-1. S2CID   219070999.
  20. ↑ Guo, Fuchun; Mu, Yi; Susilo, Willy; Varadharajan, Vijay (2017). "Privacy-Preserving Mutual Authentication in RFID with Designated Readers". Wireless Personal Communications. 96 (3): 4819–4845. doi:10.1007/s11277-017-4430-x. S2CID   207264759.
  21. ↑ Liu, Xiaoxue; Ma, Wenping; Cao, Hao (2019). "MBPA: A Medibchain-Based Privacy-Preserving Mutual Authentication in TMIS for Mobile Medical Cloud Architecture". IEEE Access. 7: 149282–149298. Bibcode:2019IEEEA...7n9282L. doi: 10.1109/ACCESS.2019.2947313 . S2CID   204863294.
  22. 1 2 Liu, Xiaoxue; Ma, Wenping; Cao, Hao (2019). "NPMA: A Novel Privacy-Preserving Mutual Authentication in TMIS for Mobile Edge-Cloud Architecture". Journal of Medical Systems. 43 (10): 318. doi:10.1007/s10916-019-1444-9. PMID   31522286. S2CID   202570185.
This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.