Invented by Kai-How FARH, Donavan CHENG, John SHON, Jorg HAKENBERG, Eugene BOLOTIN, James Casey GEANEY, Hong Gao, Pam CHENG, Inderjit Singh, Daniel ROCHE, Milan Karangutkar, Illumina Inc

The market for federated systems and methods for medical data sharing is rapidly growing, driven by the increasing need for seamless and secure exchange of healthcare information. In today’s digital age, healthcare providers, researchers, and patients alike require access to comprehensive and up-to-date medical data to make informed decisions and improve patient outcomes. Federated systems offer a promising solution by allowing disparate healthcare organizations to share data while maintaining control over their own information. Federated systems enable the aggregation of data from various sources, such as electronic health records (EHRs), medical devices, and wearable technologies. These systems use standardized protocols and technologies to ensure interoperability and facilitate the secure exchange of data. By connecting different healthcare entities, federated systems break down data silos and enable a holistic view of a patient’s medical history, regardless of where the data is stored. One of the key advantages of federated systems is the ability to preserve data ownership and control. Healthcare organizations are often hesitant to share data due to concerns about patient privacy, data security, and regulatory compliance. Federated systems address these concerns by allowing organizations to maintain control over their data while still participating in a larger network. This decentralized approach ensures that sensitive patient information remains protected and only accessible to authorized individuals. The market for federated systems and methods for medical data sharing is driven by several factors. Firstly, the increasing adoption of EHRs and other digital health technologies has led to a massive accumulation of healthcare data. However, this data is often fragmented across different systems, making it difficult to access and analyze. Federated systems provide a solution by enabling data integration and sharing, leading to more comprehensive and accurate insights. Secondly, the growing emphasis on value-based care and population health management requires a collaborative approach to healthcare delivery. Federated systems allow healthcare providers to collaborate and share data across different organizations, leading to more coordinated and efficient care. This is particularly important in cases where patients receive care from multiple providers or when conducting research across different institutions. Furthermore, the rise of precision medicine and personalized healthcare has increased the demand for comprehensive patient data. Federated systems enable the aggregation of diverse datasets, including genomic information, clinical data, and lifestyle data, to support personalized treatment plans and research initiatives. The market for federated systems and methods for medical data sharing is highly competitive, with several key players offering innovative solutions. These solutions range from cloud-based platforms to blockchain-based systems, each with its own unique features and advantages. As the market continues to evolve, interoperability and data security will remain critical factors for success. In conclusion, the market for federated systems and methods for medical data sharing is experiencing significant growth due to the increasing need for seamless and secure exchange of healthcare information. These systems enable healthcare organizations to share data while maintaining control over their own information, addressing concerns about patient privacy and data security. With the rise of digital health technologies and the emphasis on collaborative healthcare delivery, federated systems are poised to play a crucial role in improving patient outcomes and advancing medical research.

The Illumina Inc invention works as follows

Systems, computer implemented methods, and nontransitory computer readable media are provided to share medical data. The disclosed systems can be configured to create first workgroups with a single knowledgebase. This first knowledgebase can be federated to a common knowledgebase and a second knowledgebase in a separate workgroup. One or more of the three knowledgebases, namely, first knowledgebase (common knowledgebase), and second knowledgebase (second knowledgebase) may be configured to contain data items containing associations, signs and evidence. Signs may include measurements and contexts and associations may describe relationships between measurements and contexts. These associations may be supported by evidence. The disclosed systems can be configured to accept a request from an user in the first group, retrieve matching items of data, and then optionally output at least some matching items to the user. The request can include at least one first measurement and/or a first association.

Background for Federated Systems and Methods for Medical Data Sharing

Medical providers may not be familiar with rare medical conditions. Individual genetics laboratories, for example, may encounter rare genetic variations so rarely that they do not recognize the impact these variants can have on human health. The disclosed systems allow for collaborative sharing of patient data without exposing personally identifiable information. The disclosed platform allows medical providers to share their expertise and content providers to provide updated or new information about medical conditions.

The disclosed systems and method may allow for collaborative sharing of medical information. At least one processor and at least non-transitory memories can be included in a computer system that shares data. At least one nontransitory storage can be used to store instructions which, when executed by at least one processor can cause the system perform various operations, including creating a federated network comprising a shared knowledgebase and a knowledgebase for a first group, as well as a knowledgebase for a secondary workgroup. The federated data system can have a data structure configured to store signs, associations and evidences for variant data. Operation can include storing the data items that comprise the signs, associations or evidence in the federated systems. The category of signs stores biomarker measurements, contexts for phenotypes or diseases, and the category of associations stores an association between a biomarker measure and a context. And the category of evidence contains information to support the associations. Operation can include connecting a common knowledgebase with the first knowledgebase and second knowledgebase via a federated level, where each knowledgebase that is connected participates in the federated systems. On receiving a request from an individual of a first or second group to stop participating in the federated systems, the operations can include disconnecting one or both groups. Receiving a request, which includes at least one association or measurement, from a user can be included in the operations. The operations include a real-time federated query through the federated layer for the request received from the user, through the knowledgebases participating in the system. The operations include retrieving matching items from the federated searches of at least one of common knowledgebases, first knowledgebases, and second knowledgebases; and presenting to the user some or all of those matching items.

A nontransitory computer-readable media can store instructions which, when executed by an at least one processor in a system, causes a node to perform operations including: creating a Federated System comprising a Common Knowledgebase, a First Workgroup having a First Knowledgebase, and Second knowledgebases of a Second Workgroup. The federated data structure can be configured to store different categories of signs, associations and evidences. Data items containing signs, associations or evidence can be stored in the federated systems 117. The category for signs stores biomarker measurements, contexts for phenotypes or diseases, and the category of associations stores an association between a biomarker measure and context. Operation can include connecting a common knowledgebase with the first knowledgebase and second knowledgebase via a federated level, where each knowledgebase that is connected participates in the federated systems. On receiving a request from an individual of a first or second group to stop participating in the federated systems, the operations can include disconnecting one or both groups. Receiving a request, which includes at least one association or measurement, from the user can be part of the operations. The operations include a real-time federated query, through the federated layer for the received user request, through the knowledgebases participating in the system. The operations include retrieving matching items that satisfy the request for the federated system from at least one knowledgebase (first knowledgebase), common knowledgebase and second knowledgebase. At least some of these matching items can be output to the user.

A computer-implemented data sharing method can include creating a federated network that includes a common knowledgebase. A first workgroup has a knowledgebase. And a second knowledgebase for a workgroup. The federated data structure can be configured to store signs, associations and evidences for variant data. The federated systems can include multiple processors and multiple storage devices. The method may include storing the data items on the storage devices in the federated systems. Data items may include signs, associations or evidence. The category of signs can include biomarker measurements, contexts for phenotypes or diseases, drugs and drug interactions, while the category of associations stores an association between a biomarker measure-ment and a context. The category of evidence contains information to support the associations. The method can involve connecting the common knowledgebase with the first knowledgebase and second knowledgebase via a federated level, where each knowledgebase that is connected participates in the federated systems. On receiving a request from an individual of a first or a second workgroup, the method can include disconnecting one or both of these workgroups from each other. The method may include receiving a user request, which includes at least one association or measurement. The method can include a real-time federated query, through the federated layer for the user’s request, through the knowledgebases participating in the system. This federated query is performed without utilizing personal information about patients in the first and second knowledgebase. The method can include retrieving data matching items from the federated searches of at least one common knowledgebase and at least one first knowledgebase and second knowledgebase. At least some of these matching data items are then output to the user.

It is understood that the general description above and the detailed description below are only exemplary and explanatory and do not restrict the disclosed embodiments as claimed.

The accompanying drawings illustrate the examples of the disclosed embodiments. The same reference numbers are used to refer to similar or identical parts wherever possible.

The disclosed systems and method may implement a platform that allows users to share data in multiple domains, without revealing individual patient information. Users involved in genome analysis and genetic diagnosis can interact with disclosed systems and method to share evidence regarding variant pathogenicity. In certain embodiments, the platform may include a diagnostic laboratory. In a knowledgebase that is associated with a domain, information about genetic variants classified as benign or pathogenic can be securely and privately stored. Users associated with the lab can also query the network when they encounter an unfamiliar genetic variation. The disclosed systems and method may be configured for tracking aggregate statistics about genetic variants and observed symptoms, providing statistical certainty once a genetic variation has been observed enough times with a phenotype.

The clinical and research applications of next-generation sequence technology for disease diagnosis and treatment are in their early stages. This technology was first implemented in targeted panels that analyzed subsets cancer-relevant genes and/or highly actionable ones for mutations. This approach is widely used, as it offers high redundancy in sequence coverage at low cost for a small number of clinically useful sites.

However many more potentially clinically-actionable mutations could exist in both known disease-related gene (such as cancer genes), and in other genes that have not yet been identified as disease genes. The efficiency of next generation sequencing has improved, making it possible to use whole genome sequencing (WGS), as well as omic tests such as RNA sequencing. However, there are still uncertainties about the amount of additional information that can be derived from these tests. WGS data or whole-transcriptomes are not only expensive, but also require expertise and time to interpret all genetic and somatic mutations. Due to a lack of highly-trained personnel who can interpret genomes, it takes genetic testing laboratories up to four months to deliver results. Concerns also exist about the protection of sensitive personal data of patients from various workgroups, when it comes to collaborating between workgroups.

Embodiments” of the invention could address bottlenecks when reporting results of medical tests. The disclosed methods and systems may also allow for collaborative analysis of medical information, such as the collaborative annotation of genetic variations. The systems and methods disclosed may allow laboratories to work with external interpreters in order to accelerate the reporting of test results. Content providers may be able to use the disclosed systems and method to push medical information to users. Researchers, pharmaceutical companies and other providers can access the system disclosed to locate patients who have particular medical profiles for clinical trials or studies.

As described in the following, medical data can be stripped of any personally identifiable information before being imported into the system or when it is retrieved to respond to a query. In this way, the disclosed systems and methods are not able to reveal any personally identifiable information. The disclosed systems and method may instead reveal general population-level data such as aggregate statistics. The disclosed systems and method may prevent privacy breaches by not disclosing all medical data of a patient. By filtering meta data corresponding to particular queries, computational savings and bandwidth efficiency can also be achieved.

The disclosed embodiments can include a data sharing system.” This system can include at least a processor and a non-transitory storage. At least one nontransitory memory can store instructions. The instructions can cause the system’s operations to be performed when they are executed by at least one processor. These operations could include creating a workgroup with a knowledgebase. The first knowledgebase can be federated to a common knowledgebase, and a secondary knowledgebase from a separate workgroup. One or more of the three knowledgebases can store data containing associations, signs and evidence. Receiving a request, containing at least a single first association or a single first measurement, from an user of the first workgroup may be part of the operations. The operations can also include retrieving data items matching from one or more of the common knowledgebase (first knowledgebase), second workgroup knowledgebase and first knowledgebase. “The operations can also include displaying to the user some of the matching data items that have been retrieved.

In some aspects, it is possible that the first and second workgroups are associated with different entities. The second workgroup can be hosted by a different node than the one hosting the first group in certain aspects. The first data sharing regulations can control the provision of data from the node and the second data sharing regulations can control the provision of the data from the node.

In some aspects, operations may also include receiving data items from a user and storing those data items in the knowledgebase. The operations can also include receiving data items from the common knowledgebase, and storing them in the first knowledgebase.

In some cases, operations could also include receiving a second request from the workgroup. This request could include at least one second measurement or a second association. Operation may include determining data items that match in the first knowledgebase and supplying the determined data to the second group. These data items can include personally identifiable information. “The operations could also include removing personally identifiable information prior to providing the determined data item to the second group.

In some aspects, it is possible that the first knowledgebase contains versions of the matching items. The matching data items retrieved may meet a version criteria. The version criterion may be included in the request. The version criteria may be satisfied by versions of matching data items that were created after a certain date. The retrieved matching items can satisfy various quality criteria in different aspects.

In certain aspects, matching data items can be retrieved from a second knowledgebase. The matching data items may also be stored in the first knowledgebase. The associations can include at least one variant, exon gene, copy number and pathway association. Some aspects of the associations include Mendelian, prognostic and predictive, pharmacokinetic and prevalence associations, as well as classification. The data items can also include curation information in some cases.

The disclosed embodiments can include a nontransitory computer readable medium that stores instructions. The instructions can be executed by atleast one processor in a system and cause a node to perform certain operations. These operations can include creating a workgroup that is associated with a single entity. The first workgroup could have a knowledgebase. The first knowledgebase can be federated to a common knowledgebase as well as a second knowledgebase from a separate workgroup. The second workgroup can be linked to a second entity that is hosted on a different node from the first. In at least one of the knowledgebases, including the common knowledgebase and the second knowledgebases, versioned data elements containing associations, signs and evidences may be stored. In the first knowledgebase versioned items may be stored. These versioned items of data may have been received by a member of the first workgroup. One of these versioned items can be pushed out from the knowledgebase. The second knowledgebase may share one of the versioned items. Operation may also include receiving a user request, which includes a version criteria and at least one association or measurement. The operations can also include retrieving matching items that satisfy the criterion in at least one knowledgebase (first, common, or second) and comparing them. The operations can also include displaying to the user some of the matching data items that have been retrieved.

In some aspects, the first data sharing regulations can control the provision of data from the first node and the second data sharing regulations can control the provision of the data from the second node.

In some cases, operations could also include receiving a second request from the workgroup. Other requests may include at least a second association or a secondary measurement. Operation may include matching data items from the first knowledgebase, and transferring the determined data to the second group. The determined data items may include personally identifiable information in various aspects. “The operations may also include removing personally identifiable information prior to providing the determined data item to the second group.

FIG. FIG. 1 shows an example schematic of a system 100 that is used to share genomics data. In certain embodiments, the system 100 may include nodes (e.g. node 101 and 102), a user device 105 and an interface layer 104. It could also contain federated databases layer 103 and ontology services layers 106 and 107. The nodes can be configured to store data related to medical care, including genomics, drug efficacy, and phenotype. In certain aspects, an user (e.g. user 105A), may interact with system 100 components, such as the user device 105 to retrieve, modify, or provide this medical data.

Click here to view the patent on Google Patents.