Limited and completely confidential join database queries to protect users' privacy and identity - Patents.com

The method allows secure and private querying of genetic knowledge repositories by using oblivious pseudorandom functions and forgetful decryption to manage query tags, addressing privacy issues and restricting access, thus enhancing patient data security.

JP7583792B2Active Publication Date: 2024-11-14KONINKLIJKE PHILIPS NV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
JP2022512391
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-08-26
Filing Date
2020-08-21
Publication Date
2024-11-14
Estimated Expiration
2040-08-21

AI Technical Summary

Technical Problem

Existing genetic knowledge repositories face challenges in allowing users to query based on combinations of genotypes and/or metadata while addressing privacy concerns, particularly when incorporating data outside a defined scope such as patient phenotypes, due to privacy issues.

Method used

A method involving a client-server interaction using secure communication channels, oblivious pseudorandom function evaluations, and forgetful decryption protocols to generate and manage query tags, ensuring privacy by limiting access rights and maintaining confidentiality of patient data.

Benefits of technology

Enables secure and private querying of patient databases by allowing access to authorized attribute-value pairs while restricting data access, thereby enhancing patient privacy and security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007583792000002
    Figure 0007583792000002
  • Figure 0007583792000003
    Figure 0007583792000003
  • Figure 0007583792000004
    Figure 0007583792000004
Patent Text Reader

Abstract

A method for securely accessing a database containing sensitive data, such as patient clinical information, by a client in a privacy-preserving manner includes: communicating with a server to obtain tags for specific attribute-value pairs when the client is authorized to make a query; imposing a per-client tag assignment and restricting tag generation to authorized query terms with valid digital signatures from a third party; secretly storing the tags and their associated query terms for future queries; transmitting a combination of tags defining combined query terms to a proxy over a secure channel; receiving from the proxy encrypted coefficients of a polynomial whose roots are indexes into query results; decrypting the encrypted coefficients in a first protocol with the server; computing roots of the polynomial based on the decrypted coefficients and discarding all extra roots; obtaining from the proxy an encrypted record associated with the computed root; and decrypting the encrypted record in a second protocol with the server.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE

[0001] Various exemplary embodiments disclosed herein relate generally to limited, completely private, join database queries to protect user privacy and identity. [Background technology]

[0002]

[0002] Genetics knowledge repositories (databases containing structured multi-center patient records and associated genomic abnormalities) can serve as a rich source of information for research and clinical decision support by linking patient phenotypes and clinical outcomes to, among other things, their underlying genotypes. In the case of rare diseases, input queries of patient characteristics and genotypes into a sufficiently large repository can help inform the diagnosis. When no approved treatment can be identified, patients can easily be informed of existing clinical trials. When clinical trials are unavailable, clinicians can engage pharmaceutical companies for new research.

[0003]

[0003] The Beacon Project of the Global Alliance for Genomic Sciences and Health (GA4GH) is an open standard for genomic data discovery that originally served as a linked database aimed at answering questions such as "Do you have information on the allele "A" at position 938294 on chromosome 9?" This feature is useful for clinicians to look for second cases that can inform treatment decisions for their patients. The Beacon Project recently published additional features including tiered metadata access levels (public, registered and curated), where the highest access level can see not only whether the allele exists in the database, but also metadata informing the exact nucleotide change, the reference genome version and associated annotations (such as pathogenicity). Summary of the Invention [Problem to be solved by the invention]

[0004] While useful in their own right, the above models do not allow users to query based on genotype combinations and / or metadata, nor do they incorporate data outside of the defined scope (e.g., a set of patient phenotypes), due in part to the many privacy challenges that face such quality frameworks. [Means for solving the problem]

[0005]

[0004] A summary of various exemplary embodiments is presented below. Some simplifications and omissions may be made in the following summary, but this is intended to highlight and introduce some aspects of various exemplary embodiments, and is not intended to limit the scope of the present invention. A detailed description of the exemplary embodiments sufficient to enable those skilled in the art to utilize the concept of the present invention is provided in the following paragraphs.

[0006]

[0005] Various embodiments relate to a method for securely accessing a patient database by a client, the method including: communicating with a server to convert attribute-value pairs into tags; obtaining tags associated with the attribute-value pairs over a secure communication channel; sending the generated combination of tags defining terms in a conjunctive query to a proxy over a secure communication channel; receiving from the proxy encrypted coefficients of a polynomial whose root is an index to records whose root satisfies the query terms; decrypting the encrypted coefficients in a first protocol with the server; calculating roots of the polynomial based on the decrypted coefficients and discarding all extra roots; obtaining encrypted records associated with the calculated roots; and decrypting the encrypted records in a second protocol with the server.

[0007]

[0006] Various embodiments are described in which the tag is a hash value generated by transforming the attribute-value pair through collaboration between the client and the server using a method that includes permitted oblivious pseudorandom function evaluation.

[0008]

[0007] Various embodiments are described in which the tag is generated only if one or more of the following criteria are met: (i) the total number of tags issued to the client is within a predetermined limit according to the access rights specified in the client's public key certificate; and (ii) each of the query terms has a valid digital signature from a third party.

[0009]

[0008] Various embodiments are described in which the step of decrypting the encrypted coefficients in a first protocol with the server includes oblivious decoding.

[0010]

[0009] Various embodiments are described in which the step of decrypting the encrypted record in a second protocol with the server includes oblivious decryption.

[0011]

[0010] Various embodiments are described that further include steps of limiting the number of tags that can be issued to a client, and secretly storing the tags and query terms associated with the tags on the client for future queries.

[0012]

[0011] Various embodiments are described that further include setting an upper limit on the total number of tags allowed for a client according to the access rights specified in the client's public key certificate, and maintaining a record on the server of the total number of tags issued to the client.

[0013]

[0012] Various embodiments are described which further include creating a user account on a server using the client's authorized public key certificate and associated private key and logging into the user account, and establishing a secure communications channel with the server.

[0014]

[0013] Various embodiments are described which further include steps of secretly storing the tags and query terms associated with the tags for future queries, and establishing a secure communications channel with the proxy.

[0015]

[0014] Further various embodiments relate to a method for establishing secure access to a patient database by a client by a server, the method including: receiving a query tag request including an authorized public key certificate of the client; verifying a digital signature of the client; if verification of the digital signature is successful, establishing a secure communications channel with the client; communicating with the server to convert attribute-value pairs into tags; and secretly storing the authorized tags in the client's account.

[0016]

[0015] Various embodiments are described in which the step of communicating with the server to convert attribute-value pairs into tags further includes a step of performing an oblivious pseudorandom function (OPRF) evaluation with the client.

[0017]

[0016] Various embodiments are described that further include tracking the total number of tags issued to the client.

[0018]

[0017] Various embodiments are described in which the step of verifying the client's digital signature further includes generating and sending a random message to the client.

[0019]

[0018] Further various embodiments relate to a client system configured to securely access a patient database using a proxy system and a server, the client system having: a memory configured to store data and computer instructions; and a processor; configured to: communicate with the server to convert attribute-value pairs into tags; obtain the tags associated with the attribute-value pairs via a secure communications channel; send the generated combination of tags defining terms in a join query to the proxy via a secure communications channel; receive from the proxy encrypted coefficients of a polynomial whose root is an index to records satisfying the query terms; decrypt the encrypted coefficients in a first protocol with the server; calculate roots of the polynomial based on the decrypted coefficients while discarding all extra roots; obtain encrypted records associated with the calculated roots; and decrypt the encrypted records in a second protocol with the server.

[0020]

[0019] Various embodiments are described in which the tag is a hash value generated by transforming the attribute-value pair through collaboration between the client and the server using a method that includes permitted oblivious pseudorandom function evaluation.

[0021]

[0020] Various embodiments are described in which the tag is generated only if one or more of the following criteria are met: (i) the total number of tags issued to the client is within a predetermined limit according to the access rights specified in the client's public key certificate; and (ii) each of the query terms has a valid digital signature from a third party.

[0022] Various embodiments are described in which the act of decrypting the encrypted coefficients in the first protocol with the server includes oblivious decoding.

[0023] Various embodiments are described in which the act of decrypting the encrypted record in a second protocol with the server includes oblivious decryption.

[0024]

[0023] Various embodiments are described in which the processor is further configured to limit the number of tags that can be issued to a client and to secretly store the tags and query terms associated with the tags to the client for future queries.

[0025]

[0024] Various embodiments are described in which the processor is further configured to set an upper limit on the total number of tags allowed for a client in accordance with the access rights specified in the client's public key certificate, and to keep a record on the server of the total number of tags issued to the client.

[0026]

[0025] Various embodiments are described in which the processor is further configured to create a user account on a server using the client's authorized public key certificate and associated private key and log into the user account to establish a secure communications channel with the server.

[0027]

[0026] Various embodiments are described in which the processor is further configured to secretly store the tags and query terms associated with the tags for future queries and to establish a secure communications channel with the proxy.

[0028]

[0027] Further various embodiments relate to a server configured to establish secure access to a patient database by a client, the server including: memory configured to store data and computer instructions; and a processor; configured to: receive a query tag request including an authorized public key certificate of the client; verify the client's digital signature; if verification of the digital signature is successful, establish a secure communications channel with the client; communicate with the client to convert attribute-value pairs into tags; and secretly store the authorized tags in a client account.

[0029]

[0028] Various embodiments are described in which the act of communicating with the client to convert attribute-value pairs into tags further includes the act of performing an oblivious pseudorandom function (OPRF) evaluation with the client.

[0030]

[0029] Various embodiments are described in which the processor is further configured to track the total number of tags issued to the client.

[0031]

[0030] Various embodiments are described in which the act of verifying the client's digital signature further includes the act of generating and sending a random message to the client.

[0032]

[0031] For a better understanding of various exemplary embodiments, reference is made to the accompanying drawings. [Brief description of the drawings]

[0033] [Figure 1] FIG. 1 shows how to set up a database and add new samples to it. [Diagram 2] FIG. 2 illustrates a method for setting a limited number of permissible query terms for authorized anonymous clients. [Diagram 3]

[0034] FIG. 3 shows a method for completely confidential queries with query term restrictions specific to each client. [Figure 4]

[0035] FIG. 4 shows an exemplary hardware diagram for implementing the client, server or proxy. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034]

[0036] In the figures, for ease of understanding, the same reference numbers are used to denote components having substantially the same or similar configurations and / or substantially the same or similar functions.

[0035]

[0037] The description and drawings illustrate the principles of the present invention. Thus, it will be understood that those skilled in the art can devise various configurations that embody the principles of the present invention and are included within the scope of the present invention, although not explicitly described or shown herein. Furthermore, all examples described herein are intended primarily for instructional purposes to help the reader understand the principles of the present invention and the concepts provided by the inventor to facilitate the art, and should not be construed as being limited to such specifically described examples and conditions. Furthermore, the term "or" as used herein refers to a non-exclusive or (i.e., and / or) (e.g., "or other" or "or alternative") unless otherwise indicated. Also, the various embodiments described herein are not necessarily mutually exclusive, as some embodiments can be combined with one or more other embodiments to form new embodiments.

[0036]

[0038] Critical to the ability to utilize the genetics knowledge warehouse for rare diseases (and more common illnesses) is the ability to pose queries related to each patient's condition and genotype. Equally important is the ability to maintain the privacy of the patients present in the knowledge warehouse and the patient data contained in the queries. Existing search engines (e.g., beacon networks, etc.) that house pure genomic data have relatively simple means (data aggregation, information budgeting, secure access, etc.) to partially mitigate re-identification, since re-identification is difficult with genomic data alone. However, for scenarios that require knowledge of patient characteristics such as phenotype, diagnosis, demographics, and other aspects of the patient journey, the barrier for privacy is necessarily set higher. The combination of genotypic and phenotypic data makes it much easier to identify a particular patient, and therefore much more difficult to protect the privacy of such patients. Any solution that promises to utilize combined clinical and genomic data must address this privacy dilemma. Below we describe an embodiment that satisfies these concerns using a set of methods for database initialization and expansion, completely private query construction, and client-targeted query restriction.

[0037]

[0039] Genetics knowledge repositories are also useful tools for creating and optimizing clinical pathways. Clinical pathways are an important tool for oncology care and represent a standard set of treatment guidelines for a given patient's medical history. Clinical pathways are generally created and improved with data residing at individual institutions due to data sharing / ownership barriers. However, privacy-preserving mechanisms are likely to enable cross-institutional genetic knowledge repositories and provide additional data as a means to optimize existing clinical pathways.

[0038]

[0040] In the following, an embodiment of the system is described that uses a set of security protocols to improve secure access to patient data. First, a third-party protocol for completely private conjunctive database queries is described. This third-party protocol is described in more detail in Private Database Queries Using Somewhat Homomorphic Encryption by Boneh D., Gentry C., Halevi S., Wang F., Wu D.J. (2013) and in Applied Cryptography and Network Security. ACNS 2013. Lecture Notes in Computer Science, vol 7954. Springer, Berlin, Heidelberg, which are incorporated herein by reference as if fully incorporated by reference.

[0039]

[0041] The protocol involves three parties: a server, a proxy, and a client. The server prepares an encrypted inverted index during a pre-processing step and sends it to the proxy, after which the server keeps only the keys used to generate this inverted index, including some hash keys and the secret key of the homomorphic encryption (SWHE) scheme.

[0040]

[0042] For every attribute-value pair (a, v) in the database, the inverted index contains the record (tg, Enc(A(x))), where tg is a tag computed as tg=Hash("a=v"), and A(x) is a polynomial whose root is exactly the index r of the record containing that attribute-value pair.

[0041]

[0043] The proxy stores an inverted index and encrypted records for the database, such that each tag tg is associated with a particular data record in the database. The basic assumption is that there is no collusion between the server and the proxy.

[0042]

[0044] The client has a public key certificate and an associated private key, and the client obtains query tags by performing an oblivious pseudorandom function (OPRF) evaluation with the server. The client secretly stores the query tags and their associated query terms, and initiates a combined query by sending a selected set of such tags to a proxy over a secure communication channel, ensuring that the query remains secret.

[0043]

[0045] To impose restrictions on the query terms allowed for authorized but anonymous clients, each client is only allowed to obtain up to a maximum number of query tags that depend on the access rights specified in the client's public key certificate. During the tag generation process, the server keeps a record of the total number of tags already issued and stops generating any more tags if the maximum number is reached for the client. The client then secretly stores the tags issued by the server and their associated query terms for future queries. As a result, a client can only form queries using a limited number of query terms. Duplicate tag requests for the same query term lead to a waste of tag allocation and must therefore be avoided by clients.

[0044]

[0046] In a more stringent setting, where a client is only allowed to query specific attribute-value pairs that have been individually digitally signed by a third party (e.g., a laboratory that generates clinical reports containing information about the patient's phenotype and genomic abnormalities), the client obtains the tag by performing an oblivious PRF evaluation with the server, where the evaluation is successful only if the client has a valid signature. The corresponding approved oblivious PRF protocol is described in De Cristofaro, Emiliano, Yanbin Lu, and Gene Tsudik, "Efficient techniques for privacy-preserving sharing of sensitive information," International Conference on Trust and Trustworty Computing, Springer, Berlin, Heidelberg, 2011. At the end of the protocol, if the signature is valid, the client obtains the tag, while the server does not obtain any new information. If the signature is invalid, neither the client nor the server obtains any new information.

[0045]

[0047] FIG. 1 shows how to set up the database and add new samples to it. To add a new sample to the database, the server creates a searchable attribute-value pair (a i , v ij For each tg, an inverted index record is generated (110) including a tag, and an encrypted polynomial function is generated (110). i =v ij ”). The encrypted polynomial function above has a root Enc(A ij (x)), where Enc() is a SWHE function, and if the attribute-value pair already exists in the database, Enc(A ij(x)) may be newly generated or updated by incorporating the index of a new matching sample as an additional root to the encryption polynomial of an existing sample. In the context of clinical and genomic databases, the attribute-value pairs may include some demographic / clinical data, e.g., "HDL=High", or genomic abnormalities, e.g., "BRAFV600E=True". The server then sends the inverted index to the proxy, retaining only the hashing and encryption keys (115). The proxy then stores the inverted index and the encrypted record (120) and the process ends (125).

[0046]

[0048] FIG. 2 shows how to request a limited number of allowed query tags from an authorized anonymous client. To prevent malicious clients from abusing the covert query mechanism by performing large searches to access data for illicit purposes, query-level access control is imposed to limit access to the minimum number of operations and data required. This can be achieved through a query tag request process in which an anonymous client with an authorized public key certificate accesses (210) its account at the server by signing (215) a random message generated by the server with its private key. Upon receiving the digital signature, the server verifies the signature using the client's public key, and if the digital signature is correct (220), a secure communication channel can be established (225) with a secret session key shared between the server and the client. Thus, the client can obtain (230) tags corresponding to a list of attribute-value pairs by performing an oblivious pseudorandom function (OPRF) evaluation with the server. These attribute-value pairs are the tags, tg=PRF S(a=v) (230), where PRF is a pseudorandom function such as a hash function, and S is a key associated with the PRF. In a more stringent setting where the client is only allowed to query certain attribute-value pairs that have been individually digitally signed by a third party, the client obtains tags by performing an oblivious PRF evaluation with the server, which is successful only if the client has a valid signature. The client then secretly stores the generated tags and their associated query terms for future queries (235). The server keeps a record of the total number of tags issued to the client, and imposes an upper bound on the total number of query terms allowed for each client (235'). This process can be repeated until the number of tags reaches the maximum allowed for the client. The maximum number of tags can be set according to the client's access rights specified in the authorized public key certificate.

[0047]

[0049] Figure 3 shows the method for a completely confidential query with query term restrictions specific to each client. The client query “SELECT * FROM database WHERE a1=v1AND … AND a n = v n " and the individual attribute-value pair tags tg k Assuming that has been obtained from the server as described in the tag request process above, the steps involved in the perfectly secret query process are as follows: 1. The client and the proxy establish (310) a secure communication channel to protect the query tags. 2. The client uses the tag tg for the join query. k to the proxy (315). 3. The proxy obtains the encrypted coefficients of the polynomial A1(x), ..., An(x), whose roots are the record indices corresponding to the tags tg1, ..., tgn of the attribute-value pairs a1 = v1, ..., an = vn. The proxy then:

number

[0048]

[0050] We now present some examples of applications of the embodiments described herein.

[0049]

[0051] In a first example, a clinician queries an existing database for registered patients. An automated (hospital-initiated, third-party application-initiated) or manual (clinician-initiated) query can be sent to query existing registered pediatric patients in an existing genetics knowledge repository. The clinician initially suspects a diagnosis of Shprintzen-Goldberg syndrome due to the presented phenotype. However, the patient tests negative for a known causative mutation in the SKI gene. The query includes the following set of patient characteristics: Demographics: Age, Gender Phenotype: Presenting symptoms suggestive of Sphrintzen-Goldberg syndrome Genotype: Negative for mutations in the SKI gene.

[0050]

[0052] Since the genetic knowledge storehouse (database) was established prior to this scenario, no database setup is required. In this case, the fully private query is performed with the following restrictions: 1. The clinician client signs into their account using their public key certificate. 2. The client communicates with the server over a secure channel to obtain OPRF-evaluated tags for the attribute-value pairs in the query. These tags are generated only for query terms that have a valid digital signature from a third party. The server keeps a record of the total number of tags issued to a client, enforcing a cap according to the client's access rights as specified in the client's public key certificate. 3. After receiving the tag, the client sends it to the proxy. 4. The proxy retrieves the encrypted coefficients of the polynomial whose root is an index to records that satisfy the query condition, and sends these coefficients to the client. 5. The client communicates with the server to perform oblivious decryption of the encrypted coefficients. 6. The client solves these coefficients to get the record index. 7. The client retrieves the encrypted record using the index of the above record and communicates with the server to perform oblivious decryption and get the final unencrypted record.

[0051]

[0053] Using the unencrypted records, other cases sharing many phenotypic characteristics with the patient are found in which a diagnosis of Kosaki overgrowth syndrome with a confirmed causative mutation in the PDGFRB gene has been made. Based on this information, the clinician orders genetic testing for a mutation in PDGFRB. It is discovered that the patient does indeed have a new mutation in PDGFRB, and treatment decisions are then made based on this new diagnosis.

[0052]

[0054] In the second example, the optimization process of a clinical pathway is illustrated. A well-known cancer center that licenses clinical pathways has negotiated with several customers (other medical facilities that pay for such licenses) to participate in a special licensing model that reduces the direct license costs for use of the clinical pathway in exchange for participation in a shared clinical knowledge base for fellow facilities. The cancer center would like to use the clinical knowledge base to evaluate how a particular clinical pathway is performing and identify areas for improvement, while keeping its proprietary clinical pathways confidential. Since the cancer center (in this case the client) has already registered, it has a set of tags that correspond to pre-determined acceptable query terms based on each clinical pathway. Using a combination of query tags that represents some or all of the pathways of interest, the cancer center can evaluate the records in the knowledge base for aggregated outcome data. Using this data, the cancer center concludes that this pathway does not provide measurable benefit to patients with aggressive disease receiving a particular systemic therapy, and decides to modify the pathway to reflect this new knowledge.

[0053]

[0055] In a third example, a well-known person has been diagnosed with major depression and severe insomnia, and is taking currently available prescription medications, but his symptoms are worsening. He wants to find clinical trials that may improve his condition, but is worried that unsecured online searches may reveal his mental health issues to the media. He learns of a clinical trials database that uses a restricted, completely private query technique and decides to try it. He obtains both an anonymized digital certificate and an electronic clinical report from the hospital that contains information about his diagnosis and symptoms ("Diagnosis = "Major Depression", "Symptoms = [Insomnia]"), and personal background ("Gender = [Male]", "Ethnicity = [White]", "Age = [20-30]", etc.)). Each of the key items in the report comes with the hospital's digital signature, and each can be used as a query term. He creates an account to access the clinical trials database using his anonymized digital certificate, and then submits the relevant query terms, along with their digital signatures, to the database server. None of his sensitive personal data is exposed to the database server or network throughout the query process. He finds a match in a clinical trial that uses an investigational digital therapeutic to provide cognitive behavioral therapy for insomnia and a sleep restriction algorithm to reduce the severity of insomnia and symptoms of depression. He enrolls in the trial, and after several months of treatment, his symptoms improve significantly.

[0054]

[0056] Figure 4 illustrates a block diagram of exemplary hardware 400 for implementing a client, server, or proxy as described above. As illustrated, device 400 includes a processor 420, memory 430, a user interface 440, a network interface 450, and storage 460 interconnected via one or more system buses 410. It will be appreciated that Figure 4 constitutes an abstraction in some respects, and the actual arrangement of components of device 400 may be more complex than depicted.

[0055]

[0057] Processor 420 may be any hardware device capable of executing instructions or otherwise processing data stored in memory 430 or storage 460. Thus, the processor may include a microprocessor, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or other similar device.

[0056]

[0058] Memory 430 may include various memories, such as, for example, an L1, L2, or L3 cache or system memory, etc. Thus, memory 430 may include static random access memory (SRAM), dynamic RAM (DRAM), flash memory, read only memory (ROM), or other similar memory devices.

[0057]

[0059] User interface 440 may include one or more devices for enabling communication with a user. For example, user interface 440 may include a display, a touch interface, a mouse, and / or a keyboard for receiving user commands. In some embodiments, user interface 440 may include a command line interface or a graphical user interface that may be presented to a remote terminal via network interface 450.

[0058]

[0060] Network interface 450 may include one or more devices for enabling communication with other hardware devices. For example, network interface 450 may include a network interface card (NIC) configured to communicate according to an Ethernet protocol or other communication protocol, including a wireless protocol. Additionally, network interface 450 may implement a TCP / IP stack for communication according to the TCP / IP protocol. Various alternative or additional hardware or configurations for network interface 450 will be apparent.

[0059]

[0061] The storage unit 460 may include one or more machine-readable storage media, such as a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium, an optical storage medium, a flash memory device, or a similar storage medium. In various embodiments, the storage unit 460 may store instructions for execution by the processor 420 or data that the processor 420 may process. For example, the storage unit 460 may store an underlying operating system 461 for controlling various basic operations of the hardware 400. The storage unit 460 may also store instructions 462 for performing the functions of a server, client, or proxy as previously described.

[0060]

[0062] It will be apparent that the various information described as being stored in storage 460 can also or instead be stored in memory 430. In this regard, memory 430 can be considered to constitute a "storage device" and storage 460 can be considered a "memory". Various other configurations will be apparent. Moreover, memory 430 and storage 460 together can be considered to be a "non-transitory machine-readable medium". As used herein, the term "non-transitory" is understood to include all forms of storage, including both volatile and non-volatile memory, but excluding transitory signals.

[0061]

[0063] Although host device 400 is shown as including one of each of the described components, various components may be duplicated in various embodiments. For example, processor 420 may include multiple microprocessors configured to independently perform the methods described herein or to perform steps or subroutines of the methods described herein, where the multiple processors cooperate to achieve the functions described herein. Furthermore, when device 400 is embodied in a cloud computing system, the various hardware components may reside in separate physical systems. For example, processor 420 may include a first processor in a first server and a second processor in a second server.

[0062]

[0064] The embodiments described herein solve the technical problem of improving security and patient privacy when queries are made on a patient database containing patient information. For example, when a patient database contains both genomic and phenomic data, it becomes much easier to identify a particular patient. Thus, the above-described embodiments only allow a user to query the database for permitted attribute-value pairs having specific values. As a result, clients are limited in what data they can access, which improves the security and privacy of the patient database.

[0063]

[0065] Any combination of specific software executed on a processor to implement an embodiment of the present invention constitutes a specific dedicated machine.

[0064]

[0066] As used herein, the term "non-transitory machine-readable storage medium" is understood to exclude transitory, propagating signals, but to include all forms of volatile and non-volatile memory.

[0065]

[0067] Although various exemplary embodiments have been described in detail with particular reference to certain illustrative aspects thereof, it should be understood that the invention is capable of other embodiments and its details can be modified in various obvious respects. As will be readily apparent to those skilled in the art, variations and modifications can be made while remaining within the spirit and scope of the invention. Accordingly, the foregoing disclosure, description and drawings are for illustrative purposes only and do not in any way limit the invention, which is defined solely by the claims.

Claims

1. 1. A method for securely accessing a patient database by a client, comprising: communicating with a server to convert the attribute-value pairs into tags; obtaining tags associated with the attribute-value pairs from the server over a secure communication channel, the tags being computed as a hash of each of the attribute-value pairs; transmitting a combination of tags defining the query terms in the combined query to the proxy over a secure communication channel; receiving from the proxy encrypted coefficients of a polynomial whose roots are indices to records satisfying the query term; communicating with the server to decrypt the encrypted coefficients; calculating the roots of the polynomial based on the decoded coefficients and discarding all redundant roots; obtaining from the proxy an encrypted patient database record associated with the calculated root; and communicating with the server to decrypt the encrypted record; The server only issues tags for attribute-value pairs that are individually digitally signed by a third party.

2. 2. The method of claim 1, wherein the tag is a hash value generated by transforming the attribute-value pairs through collaboration between the client and the server using a method that includes authorized oblivious pseudorandom function evaluation.

3. 2. The method of claim 1, wherein the tag is generated only if a criterion is met that the total number of tags issued to the client is within a predetermined limit according to access rights specified in the client's public key certificate.

4. The method of claim 1 , wherein the step of communicating with the server to decrypt the encrypted coefficients comprises oblivious decoding.

5. The method of claim 1 , wherein the step of communicating with the server to decrypt the encrypted record comprises oblivious decryption.

6. The method of claim 1, further comprising the step of secretly storing tags and query terms associated with the tags at the client for future queries.

7. 2. The method of claim 1, further comprising setting by the server an upper limit on the total number of tags allowed to a client according to access rights specified in the client's public key certificate, and maintaining in the server a record of the total number of tags issued to the client.

8. 2. The method of claim 1, further comprising the steps of creating and logging into a user account on a server using the client's authorized public key certificate and associated private key, and establishing a secure communications channel with the server.

9. The method of claim 1 , further comprising the steps of: secretly storing the tags and query terms associated with the tags for future queries; and establishing a secure communications channel with the proxy.

10. A method performed by a server for establishing secure access by a client to a patient database located at a proxy, comprising: receiving a query tag request including an authorized public key certificate of the client; verifying the client's digital signature using the client's public key; if verification of the digital signature is successful, establishing a secure communication channel with the client; sending a tag to the client, the tag being calculated as a hash of each of the attribute-value pairs; and storing said tag secretly in said client's account; the server issues only tags for attribute-value pairs that are individually digitally signed by a third party, the tags corresponding to the attribute-value pairs are sent to the proxy as an inverted index record, the inverted index record includes the tags and an encrypted polynomial function, the roots of the polynomial function are record indexes including the attribute-value pairs, and the tags are used to retrieve data from the proxy.

11. 11. The method of claim 10, wherein communicating with the client to convert attribute-value pairs into tags further comprises performing an oblivious pseudorandom function (OPRF) evaluation with the client.

12. The method of claim 10, further comprising tracking a total number of tags issued to the client.

13. The method of claim 10 , wherein verifying the digital signature of the client further comprises generating and sending a random message to the client.

14. 1. A client system for securely accessing a patient database using a proxy system and a server, comprising: A memory for storing data and computer instructions, and a processor, the processor comprising: communicating with the server to receive a tag computed as a hash of each of the attribute-value pairs; Retrieving the attribute-value pair tags from the server over a secure communications channel; sending, via a secure communications channel, to the proxy, a combination of tags that define the query terms in the combined query; receiving from the proxy encrypted coefficients of a polynomial whose roots are indices to records satisfying the query term; Decrypting the encrypted coefficients using a first protocol of the server; calculating the roots of the polynomial based on the decoded coefficients received from the server and discarding all redundant roots; Obtaining from the proxy an encrypted patient database record associated with the calculated root; and decrypting the encrypted record in a second protocol with the server; The server issues tags only for attribute-value pairs that are individually digitally signed by a third party to the client system.

15. 15. The client system of claim 14, wherein the tag is a hash value generated by transforming the attribute-value pairs through collaboration between a client and the server using a method including permitted oblivious pseudorandom function evaluations.

16. 15. The client system of claim 14, wherein the tag is generated only if a criterion is met that the total number of tags issued to the client is within a predetermined limit according to access rights specified in the client's public key certificate.

17. 15. The client system of claim 14, wherein the act of decrypting the encrypted coefficients in the first protocol with the server comprises oblivious decoding.

18. 15. The client system of claim 14, wherein the act of decrypting the encrypted record in a second protocol with the server comprises oblivious decryption.

19. 15. The client system of claim 14, wherein the processor is further configured to limit the number of tags that can be issued to a client, and to privately store tags and query terms associated with the tags on the client for future queries.

20. 20. The client system of claim 19, wherein the processor is further configured to set an upper limit on the total number of tags allowed for a client according to access rights specified in the client's public key certificate, and to keep a record on the server of the total number of tags issued to the client.

21. A server for establishing secure access by a client to a proxy patient database, comprising: A memory for storing data and computer instructions, and a processor, the processor comprising: receiving a query tag request including an authorized public key certificate of the client; verifying the client's digital signature using the client's public key; Upon successful verification of the digital signature, establishing a secure communication channel with the client; communicating with the client, returning attribute-value pairs to the client as a tag computed as a hash of each of the attribute-value pairs; and Secrecy storing the authorized tags in a client account; The server issues only tags for attribute-value pairs that are individually digitally signed by a third party, the tags corresponding to the attribute-value pairs are sent to the proxy as an inverted index record, the inverted index record includes the tags and an encrypted polynomial function, the roots of the polynomial function are record indexes including the attribute-value pairs, and the tags are used by the client to retrieve data from the proxy.

22. 22. The server of claim 21, wherein the act of communicating with the client to convert attribute-value pairs into tags further comprises an act of performing an oblivious pseudorandom function (OPRF) evaluation with the client.

23. 22. The server of claim 21, wherein the processor is further configured to track a total number of tags issued to the client.

24. 22. The server of claim 21, wherein the act of verifying the digital signature of the client further comprises an act of generating and sending a random message to the client.

Citation Information

Patent Citations

  • Method, computer program, and apparatus for initializing secure communication among and for exclusively pairing device

    JP2010158030A

  • Clinical research data and personal identification information management method, management device, program and information system

    JP2015022395A