Patent Application
20190243845
Present invention embodiments relate to data integration for a plurality of data systems, and more specifically, to hierarchical association of entity records from different data systems, for example, aggregating from individual source system entity matching to galaxy, or integrated network, level entity matching.
Healthcare networks have very complicated organization structures. An organization typically comprises multiple source systems (e.g., a source of electronic medical records including electronic health records (EHR), records from a claims system, lab feed, various data sources implementing the HL7 standard, patient satisfaction survey, etc.). Clinically integrated networks (CIN) or galaxies (e.g., a group of organizations) are collections of individual healthcare systems with data sharing agreements. Data governance restrictions may exist within organizations or galaxies (e.g., not all data can be shared with all providers). Accordingly, examining and associating records of the different healthcare systems with common entities may be complex, burdensome, and processing intensive (with respect to processing resources and processing time). This is typically exacerbated in the case of data governance restrictions.
According to one embodiment of the present invention, a system links data objects for the same or common entities across source systems and includes at least one processor. The system compares data objects within each of a plurality of source systems to identify data objects associated with corresponding common entities for entity resolution. The identified data objects for each common entity within each of the plurality of source systems are linked to form a group of data objects for each common entity. The groups of data objects for each of the common entities are compared across the plurality of source systems to identify groups of data objects associated with common entities. The identified groups of data objects for common entities are linked across the plurality of source systems to form a set of data objects for each corresponding common entity. Embodiments of the present invention further include a method and computer program product for linking data objects for common entities across source systems in substantially the same manner described above.
Generally, like reference numerals in the various figures are utilized to designate like components.
An organization may comprise multiple source systems (e.g., a source of electronic medical records including electronic health records (EHR), records from a claims system, lab feed, various data sources implementing the HL7 standard, patient satisfaction survey, etc.), while clinically integrated networks (CIN) or galaxies (e.g., a group of organizations) are collections of individual healthcare systems with data sharing agreements. These agreements may define restrictions on how the data may be used, and such restrictions must be complied with according to an entity's data governance policies and procedures. Present invention embodiments pertain to hierarchical patient or other entity matching (e.g., aggregating from entity matching at a source system level to galaxy level entity matching). For example, intra-organization assignments (or entity matches within an organization) are derived from intra-source system assignments (or entity matches within individual source systems), while intra-galaxy assignments (or entity matches within a galaxy) are derived from the intra-organization assignments (or entity matches within individual organizations). Further, inter-galaxy (e.g., universe of galaxies) assignments (or entity matches within a universe) are derived from the intra-galaxy assignments (or entity matches within individual galaxies). The hierarchical entity matching may be applied in a similar manner to any quantity of levels in a hierarchy.
Each additional level in a hierarchy (e.g., from a source system to a galaxy) provides a better representation of a person or other entity, which leads to better data integration performance. For example, when five source persons are matched at a single organization, the information from all five persons is now available for matching with persons outside of the organization in the same galaxy. By way of example, each source system may provide varying information for persons or other entities (e.g., entity characteristics, clinical information, healthcare information, etc.). This information is aggregated as additional persons or entities are matched, thereby providing additional information for matching at the next hierarchical level. In other words, person assignments are treated as a hierarchy from a single source system, to an organization (collection of source systems), and ultimately to a galaxy or clinically integrated network (CIN) (e.g., collection of organizations).
Present invention embodiments provide several advantages. For example, a present invention embodiment supports complex data governance issues of an organization or galaxy by choosing whether or not to respect or enforce assignments in the component source systems/organizations.
Further, computer processing performance improvements are attained by considering resolved patients or entities from the previous level of the matching hierarchy. For example, matching metrics lower the number of incorrectly matched entities, and decrease the number of incorrectly unmatched entities. Moreover, hierarchical matching of a present invention embodiment reduces processing time of a processor by respecting the hierarchy. The entity matching is preferably implemented in a distributed computing environment as described below, and is highly scalable to hundreds of millions of patients or other entities. By way of example, 150 million source persons may be processed into 95 million resolved persons in a short time interval (e.g., a couple of hours, etc.).
An example computing environment for use with present invention embodiments is illustrated in
Healthcare network 105 includes a health data gateway 110 coupled to end-user systems 118 and one or more clinical/operational data sources 125 providing various medical information (e.g., electronic health records (EHR), records from a claims system, lab feed, various data sources implementing the HL7 standard, patient satisfaction survey, etc.) stored according to a source data model.
Data center 115 includes an application server cluster 140, a gateway controller 145, a staging grid 150, and a factory grid 160. Health data gateway 110 of healthcare network 105 is configured to acquire data from data sources 125 and transmit the acquired data to gateway controller 145 of data center 115. The gateway controller receives the incoming data from the communications network and processes that data to staging grid 150. The staging and factory grids each include a cluster of computer systems to store data and perform parallel processing. By way of example, the staging and factory grids each employ a HADOOP cluster with a HADOOP distributed file system (HDFS).
Staging grid 150 inspects and publishes the data to factory grid 160 in accordance with a data model employed by the factory grid. Factory grid 160 includes various engines to perform desired analytics on the data based on queries received from end-user systems 118 and other end-user systems 155 accessing data center 115 over network 120. The queries are handled in conjunction with application server cluster 140 to produce desired results.
Referring to
Gateway controller 145 receives the incoming data from communications network 120 and processes that data to staging grid 150 employing data models of the source systems. Staging grid 150 includes a data inspection module 252, a data publishing module 254, and a publish auditing module 256 to inspect, publish, and audit the data to factory grid 160 in accordance with the data model employed by the factory grid.
Factory grid 160 includes a data curation module 262, a patient matching module 264, an indexing module 266, and various calculation/analytic engines 268. Data curation module 262 performs data curation operations including mapping codes, data cleansing, and standardization, while patient matching module 264 performs patient matching operations to determine records associated with the same patient. Indexing module 266 performs indexing operations including combining records based on patient matching, mappings, and application of risk models. The calculation/analytic engines perform the desired analytics based on queries received from end-users from an interaction layer 230 enabling application server cluster 140 to provide various applications for processing and accessing the data (e.g., analytic applications, SQL access, etc.). The staging and factory grids form an aggregation and engines layer 220 to process the acquired data, while the queries are handled by factory grid 160 in conjunction with application server cluster 140 to produce desired results for the interaction layer.
The various applications of application server cluster 140 may be provided in a cloud environment. It is understood in advance that although this disclosure includes a detailed description on cloud computing, implementation of the teachings recited herein are not limited to a cloud computing environment. Rather, embodiments of the present invention are capable of being implemented in conjunction with any other type of computing environment now known or later developed.
Cloud computing is a model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service. This cloud model may include at least five characteristics, at least three service models, and at least four deployment models.
Characteristics are as follows:
Service Models are as follows:
Deployment Models are as follows:
A cloud computing environment is service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability. At the heart of cloud computing is an infrastructure comprising a network of interconnected nodes. Referring now to
Referring now to
Hardware and software layer 460 includes hardware and software components. Examples of hardware components include: mainframes 461; RISC (Reduced Instruction Set Computer) architecture based servers 462; servers 463; blade servers 464; storage devices 465; and networks and networking components 466. In some embodiments, software components include network application server software 467 and database software 468.
Virtualization layer 470 provides an abstraction layer from which the following examples of virtual entities may be provided: virtual servers 471; virtual storage 472; virtual networks 473, including virtual private networks; virtual applications and operating systems 474; and virtual clients 475.
In one example embodiment, management layer 480 may provide some or all of the functions for data center 115 described herein. Resource provisioning 481 provides dynamic procurement of computing resources and other resources that are utilized to perform tasks within the cloud computing environment. Metering and Pricing 482 provide cost tracking as resources are utilized within the cloud computing environment, and billing or invoicing for consumption of these resources. In one example, these resources may comprise application software licenses. Security 486 provides identity verification for cloud consumers and tasks, as well as protection for data and other resources. User portal 483 provides access to the cloud computing environment for consumers and system administrators. Service level management 484 provides cloud computing resource allocation and management such that required service levels are met. Service Level Agreement (SLA) planning and fulfillment 485 provide pre-arrangement for, and procurement of, cloud computing resources for which a future requirement is anticipated in accordance with an SLA.
Workloads layer 490 provides examples of functionality for which the cloud computing environment may be utilized. Examples of workloads and functions which may be provided from this layer include: mapping and navigation 491; software development and lifecycle management 492; virtual classroom education delivery 493; data analytics processing 494; transaction processing 495; aggregation and engines layer 220 (
Referring now to
In computing node 510, there is a computer system 512 which is operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and/or configurations that may be suitable for use with computer system 512 include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems or devices, and the like.
Computer system 512 may be described in the general context of computer system executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. Computer system 512 may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
As shown in
Bus 518 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnects (PCI) bus.
Computer system 512 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by computer system 512, and it includes both volatile and non-volatile media, removable and non-removable media.
System memory 528 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) 530 and/or cache memory 532. Computer system 512 may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only, storage system 534 can be provided for reading from and writing to a nonremovable, non-volatile magnetic media (not shown and typically called a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected to bus 518 by one or more data media interfaces. As will be further depicted and described below, memory 528 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
Program/utility 540, having a set (at least one) of program modules 542, may be stored in memory 528 by way of example, and not limitation, as well as an operating system, one or more application programs, other program modules, and program data. Each of the operating system, one or more application programs, other program modules, and program data or some combination thereof, may include an implementation of a networking environment. Program modules 542 generally carry out the functions and/or methodologies of embodiments of the invention as described herein.
Computer system 512 may also communicate with one or more external devices 514 such as a keyboard, a pointing device, a display 524, etc.; one or more devices that enable a user to interact with computer system 512; and/or any devices (e.g., network card, modem, etc.) that enable computer system 512 to communicate with one or more other computing devices. Such communication can occur via Input/Output (I/O) interfaces 522. Still yet, computer system 512 can communicate with one or more networks such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet) via network adapter 520. As depicted, network adapter 520 communicates with the other components of computer system 512 via bus 518. It should be understood that although not shown, other hardware and/or software components could be used in conjunction with computer system 512. Examples, include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
A manner of associating or linking records of a common patient or other entity across different source data systems in a hierarchical fashion (e.g., via the factory grid and patient matching module 264) is illustrated in
Patient matching is initially performed for each source system 630 to match patient records within that source system and determine the records associated with the same or common patient. The patient matching process may employ various matching algorithms to determine for each patient of a source system 630 a group of records of the source system associated with that patient.
The patient matching process includes a data pre-processing stage, a blocking stage, a matching stage, a compaction stage, and a splitting stage. By way of example, each of these stages may be performed as a respective MapReduce job within the distributed computing environment of factory grid 160 (
Referring to
The resulting clean and standardized data is evaluated to arrange records of potentially matching patients into blocks at step 710. Each block contains records of patients that are sufficiently similar in order to compare records of those patients to each other in a pairwise fashion for associating the records with a common patient. Since the amount of patient records in the source systems is voluminous, comparing all pairs of patient records is impractical. For example, the amount of comparisons is proportional to the square of the number of records (e.g., N2 comparisons are performed for N patient records). Thus, one trillion comparisons are performed for one million patient records. This task becomes burdensome and time consuming even within a distributed computing environment.
Reducing the number of comparisons is necessary to enhance computer processing performance, reduce processing time, and reduce certain types of errors. This is accomplished by organizing the collection of patient records into subsets or blocks. Each block contains patient records of patients that are sufficiently similar, where the records of those patients within the blocks are compared to reduce the overall number of comparisons.
The blocks are formed by comparing one or more fields of patient records to desired criteria. Exact matches on particular fields define blocks of patient records for pairwise comparisons of those records within each block (e.g., comparisons for social security number (SSN), medical record number (MRN), enterprise master patient index (EMPI), telephone number, phonetic name matching, etc.). This enables the pairwise comparisons of patient records to be limited to the patient records within the blocks (which are likely to be associated with the same or common patient), thereby reducing the overall number of patient record comparisons.
Various blocking schemes (e.g., criteria or combinations of record fields) may be utilized for the record comparisons to form the blocks of patient records. Thus, a single patient may be assigned to a plurality of different blocks based on the blocking schemes utilized for those blocks (e.g., combination of fields utilized for comparison of the patient records). The blocking schemes may include any quantity of any fields of the patient records for the record comparisons to form the blocks.
Once the blocks of patient records are formed, the patient records within each block are analyzed on an individual block basis to determine patient records associated with the same or common patient at step 715. The patient records within a block are compared to each other in a pairwise fashion to determine a likelihood score for the pair of records. The comparisons are limited to the patient records within the individual blocks in order to reduce the overall number of record comparisons as described above. When the likelihood score exceeds a similarity threshold, the patient records are linked and associated with the same patient.
In particular, the likelihood score is computed for a pair of patient records in a block to indicate the likelihood that the two patient records are associated with the same or common patient. The determination of the likelihood score for a pair of patient records is based on evidence from various features or record fields (e.g., first name, middle name/initial, last name, gender, birth year, birth month/day, social security number (SSN), medical record number (MRN), enterprise master patient index (EMPI), address, postal code, telephone number, etc.). Each feature is associated with different matching levels for a comparison. By way of example:
Each matching level for a feature comparison has a corresponding associated weight. The associated weights for the matching levels indicate a likelihood the patient records are associated with the same or common patient based on the level of matching of the corresponding feature or record field. The weights for the matches may be added to produce the likelihood score. When the likelihood score is greater than the similarity threshold, the patient records are linked and associated with the same patient. If the likelihood score is less than a difference threshold, the patient records are split or disassociated with one another (since the patient records reside in the same block). By way of example, higher valued weights (and hence, a higher likelihood score) may indicate a greater likelihood of the patient records being associated with the same or common patient. However, the magnitude of the weight values (and likelihood score) may be associated with any desired degree of likelihood of association of the patient records with a common patient (e.g., greater likelihood, less likelihood, etc.). In addition, the similarity and difference thresholds may be set to any desired values to control the sensitivity or degree of matching needed to associate or link patient records with a common patient.
Patient matching compares patient records within a block to each other in a pairwise fashion, and links records together based on the result of the comparison. Thus, various pairs of patient records may be linked to one another based on the comparisons within the blocks. For example, a patient record within a plurality of blocks may be linked to multiple records from among those plurality of different blocks. The linked patient records are processed to transform pairs of linked patient records into groups of patient records 640 (
The linked pairs of patient records may be represented graphically (
The connected components process is iterative and passes information pertaining to connectedness throughout the graphical representation. By way of example, a graphical representation 800 of linked patient records is illustrated in
The links or connections between vertices or patient records are indicated by a table 820a (
Table 820a indicates the initial connections, assigned group identifiers, and activated status for graphical representation 800. For example: vertex 1 is connected to vertices 2, 3, and has a group identifier set to the minimum vertex identifier of 1; vertex 2 is connected to vertices 1, 4, and has a group identifier set to the minimum vertex identifier of 1; vertex 3 is connected to vertex 1, and has a group identifier set to the minimum vertex identifier of 1; vertex 4 is connected to vertex 2, and has a group identifier set to the minimum vertex identifier of 2; vertex 5 is connected to vertex 6, and has a group identifier set to the minimum vertex identifier of 5; vertex 6 is connected to vertices 5, 7, and has a group identifier set to the minimum vertex identifier of 5; vertex 7 is connected to vertices 6, 8, and has a group identifier set to the minimum vertex identifier of 6; vertex 8 is connected to vertices 7, 9, and has a group identifier set to the minimum vertex identifier of 7; and vertex 9 is connected to vertex 8, and has a group identifier set to the minimum vertex identifier of 8. In addition, the activated status for each vertex is initially set to True.
Each activated vertex (e.g., a vertex with an activated status of True) passes the assigned group identifier of that vertex to corresponding adjacent vertices. The assigned group identifier for each of the adjacent vertices is updated to be the minimum of the current assigned group identifier for that adjacent vertex and the group identifiers passed to that adjacent vertex. By way of example with respect to graphical representation 800: the assigned group identifier for vertex 4 is updated from group identifier 2 to group identifier 1 (e.g., minimum of vertex group identifier 2 and passed group identifier 1 (from vertex 2)); the assigned group identifier for vertex 7 is updated from group identifier 6 to group identifier 5 (e.g., minimum of vertex group identifier 6 and passed group identifiers 5 (from vertex 6) and 7 (from vertex 8)); the assigned group identifier for vertex 8 is updated from group identifier 7 to group identifier 6 (e.g., minimum of vertex group identifier 7 and passed group identifiers 6 (from vertex 7) and 8 (from vertex 9)); and the assigned group identifier for vertex 9 is updated from group identifier 8 to group identifier 7 (e.g., minimum of vertex group identifier 8 and passed group identifier 7 (from vertex 8)). In addition, vertices with changed group assignments have an activated status set to True, while remaining vertices have an activated status set to False. These changes (or iteration of the process) are reflected in connection table 820b (
Each activated vertex (e.g., a vertex with an activated status of True) from table 820b passes the assigned group identifier of that vertex to corresponding adjacent vertices. The assigned group identifier for each of the adjacent vertices is updated to be the minimum of the current assigned group identifier for that adjacent vertex and the group identifiers passed to that adjacent vertex. By way of example with respect to graphical representation 800 and table 820b: the assigned group identifier for vertex 8 is updated from group identifier 6 to group identifier 5 (e.g., minimum of vertex group identifier 6 and passed group identifiers 5 (from vertex 7) and 7 (from vertex 9)); and the assigned group identifier for vertex 9 is updated from group identifier 7 to group identifier 6 (e.g., minimum of vertex group identifier 7 and passed group identifier 6 (from vertex 8)). In addition, vertices with changed group assignments have an activated status set to True, while remaining vertices have an activated status set to False. These changes (or iteration of the process) are reflected in connection table 820c (
Each activated vertex (e.g., a vertex with an activated status of True) from table 820c passes the assigned group identifier of that vertex to corresponding adjacent vertices. The assigned group identifier for each of the adjacent vertices is updated to be the minimum of the current assigned group identifier for that adjacent vertex and the group identifiers passed to that adjacent vertex. By way of example with respect to graphical representation 800 and table 820c: the assigned group identifier for vertex 9 is updated from group identifier 6 to group identifier 5 (e.g., minimum of vertex group identifier 6 and passed group identifier 5 (from vertex 8)). In addition, vertices with changed group assignments have an activated status set to True, while remaining vertices have an activated status set to False. These changes (or iteration of the process) are reflected in connection table 820d (
Each activated vertex (e.g., a vertex with an activated status of True) from table 820d passes the assigned group identifier of that vertex to corresponding adjacent vertices. Since no further changes to assigned group identifiers occurs, all vertices are inactive (e.g., activated status set to False) and the process has converged. The final groups are reflected in table 820e (
The patient records within each resulting group are subsequently examined at step 723 to split that group into a plurality of groups as illustrated in
Initially, all patient records within a group are compared in a pairwise fashion at step 905 to produce a likelihood score for the pair as described above. For example, weights are associated for matching levels of features, and indicate a likelihood the patient records are associated with the same or common patient. The weights for matches between the pair of patient records may be added to produce the likelihood score based on the level of matching of the corresponding features or record fields. The resulting likelihood scores for the patient records are compared to a splitting threshold to identify pairs of patient records with a likelihood score below the splitting threshold at step 910.
The group of patient records may be represented graphically, where vertices represent the patient records and each edge between a pair of vertices or patient records is associated with a corresponding likelihood score for that pair. Initially, the graphical representation includes edges between each pair of patient records in the group. The edge between patient records associated with the lowest likelihood score is removed from the graphical representation at step 915. The graphical representation is examined to determine whether the records of each of the identified pairs of patient records (with a likelihood score below the splitting threshold) remain connected after removal of the edge. This may be accomplished by applying a conventional or other shortest path algorithm to determine a shortest path between the patient records of the identified pairs. When records of at least one of the identified pairs of patient records remain connected within the graphical representation as determined at step 920 (e.g., a shortest path can be determined between the patient records of at least one identified pair), the above process is repeated from step 915 by removing the edge between patient records associated with the next lowest likelihood score. The process continues until no connections exist in the graphical representation between the records of each of the identified pairs of patient records.
Thus, the splitting process iteratively breaks the weakest edges (e.g., associated with the lowest likelihood scores) until there are no connections between the identified pairs of patient records with likelihood scores below the splitting threshold. In other words, there are no edges in the graphical representation of the group with likelihood scores below the splitting threshold. The resulting groups after the removal of edges are provided as the groups of patient records 640 associated with common patients.
By way of example and referring to
Once the resulting groups 640 are formed, the presence of remaining levels for the hierarchy is determined at step 725. When additional levels of the hierarchy remain, the resulting groups 640 are set as the input to the patient matching process for the next level at step 730. In particular, the resulting groups 640 for each source system 630 within the same organization 620 are processed to perform patient matching among those groups in substantially the same manner described above for
In particular, patient groups 640 of each source system 630 are processed to perform patient matching to determine sets of records 650 associated with the same patients. Patient groups 640 of a source system 630 are retrieved and pre-processed at step 705. This includes cleaning and standardizing data within the patient records of the retrieved patient groups. For example, the data may be analyzed to identify: invalid social security numbers or other identifiers; invalid addresses and telephone numbers; and/or default values (e.g., a Birth date of 1900-01-01 (year/month/day), etc.) as described above. Further, the data of the retrieved groups of patient records is standardized to enable accurate comparisons of the data for patient matching (e.g., providing social security numbers in a desired format (e.g., with or without hyphens, providing data in corresponding fields (e.g., middle names in a first name or other field), etc.) as described above. The cleaning and standardizing may be customized for each level based on the comparisons (e.g., individual patient records, groups of patient records, etc.), and may even be bypassed in certain instances since the data should have already been cleansed and standardized at the prior level.
The resulting clean and standardized data is evaluated to assign groups of potentially matching patients to blocks at step 710. Each block contains groups of patient records 640 that are sufficiently similar in order to compare groups to each other in a pairwise fashion for associating the groups with a common patient.
Each block contains groups 640 (of patient records) that are sufficiently similar, where the groups within the blocks are compared. The blocks are formed by comparing one or more fields of patient records within the groups 640 to certain criteria. Exact matches on particular fields define smaller blocks of groups 640 for pairwise comparisons of those groups. This enables the pairwise comparisons to be limited to the groups 640 within the blocks (which are likely to be associated with the same or common patient), thereby reducing the overall number of patient record comparisons.
Various blocking schemes (e.g., criteria or combinations of record fields) may be utilized for the group comparisons to form the blocks of groups 640. Thus, a single group may be assigned to a plurality of different blocks based on the blocking schemes utilized for those blocks (e.g., combination of fields utilized for comparison of the patient records). The blocking schemes may include any quantity of any fields of the patient records within the groups for the comparisons to form the blocks. Further, since a group may contain a plurality of patient records from different sources, additional information (relative to a prior hierarchical level) is available to use as criteria for the blocking schemes. For example, patient records within a group may include variations within the same record field (e.g., a first name field among the patient records within a group may include a nickname, an abbreviated name or a formal name). A blocking scheme may include any quantity of the variants to form the block (e.g., a block may be formed based on groups with a first name field having a certain nickname, etc.).
Once the blocks of groups 640 are formed, the groups within each block are analyzed on an individual block basis to determine sets of groups (or patient records) associated with the same or common patient at step 715. The groups 640 within a block are compared to each other in a pairwise fashion to determine a likelihood score for the pair of groups. The comparisons are limited to the groups within the individual blocks in order to reduce the overall number of record comparisons as described above. When the likelihood score exceeds a similarity threshold, the patient groups are linked and associated with the same patient.
In particular, the likelihood score is computed for a pair of groups 640 in a block to indicate the likelihood that the two groups of patient records are associated with the same or common patient. The determination of the likelihood score for a pair of groups is based on evidence from various features or record fields (e.g., first name, middle name/initial, last name, gender, birth year, birth month/day, social security number (SSN), medical record number (MRN), enterprise master patient index (EMPI), address, postal code, telephone number, etc.) as described above.
Each feature has different matching levels, where each matching level for a feature has a corresponding associated weight as described above. The associated weights for the matching levels indicate a likelihood the groups 640 are associated with the same or common patient based on the level of matching of the corresponding feature or record field between the groups. Since a group may contain a plurality of patient records, the record fields may have varying values for the same field as described above. In this case, the comparisons may be expanded to include the variants (or different record field values) from the groups. The maximum weight from a comparison of the feature or record field between records of each of the groups (e.g., in pairwise comparisons of records between the groups) is determined and used as the weight for that feature. For example, a highest attained matching level (associated with the greatest weight value) may be utilized for a particular feature.
Since groups may contain patient records from different sources, the groups may contain additional information or evidence (relative to the prior hierarchical level) that may be utilized to match the groups. Thus, each additional level in the hierarchy (from source system to galaxy) provides a better representation of a patient that leads to better performance. For example, when five patients at a single source system 630 get matched together, the information available from the five patients is now available to for blocking and/or matching with patients in other source systems 630 in the same organization 620. Accordingly, the blocking and patient matching criteria may be configured for each hierarchical level to account for additional information.
The weights for the features may be added to produce the likelihood score for the pair of groups. When the likelihood score is greater than a similarity threshold, the groups are linked and associated with the same patient. If the likelihood score is less than a difference threshold, the groups are split or disassociated with one another (since they reside in the same block). By way of example, higher valued weights (and hence, a higher likelihood score) may indicate a greater likelihood of the groups being associated with the same or common patient. However, the magnitude of the matching level and weight values (and likelihood score) may be associated with any desired degree of likelihood of association of the groups with a common patient (e.g., greater likelihood, less likelihood, etc.). In addition, the similarity and difference thresholds may be set to any desired values to control the sensitivity or degree of matching needed to associate or link groups with a common patient.
The patient matching compares groups of patient records within a block to each other in a pairwise fashion, and links the groups together based on the result of the comparison. Thus, various pairs of groups 640 may be linked to one another based on the comparisons within the blocks. For example, a group within a plurality of blocks may be linked to multiple groups from among those plurality of different blocks. The linked groups are processed to transform pairs of linked groups into sets of groups 650 (
The linked pairs of patient groups may be represented graphically (e.g., similar to
The patient groups within each resulting set are subsequently examined to determine whether to split the set into a plurality of sets in substantially the same manner described above (
The splitting process iteratively breaks the weakest edges (e.g., associated with the lowest likelihood scores between the groups) until there are no connections between the identified pairs of patient groups with likelihood scores below the splitting threshold. In other words, there are no edges in the graphical representation of the set with likelihood scores below the splitting threshold. The resulting sets after the removal of edges are provided as the sets of patient records 650 associated with common patients as described above.
Once the resulting sets 650 are formed, the presence of remaining levels for the hierarchy is determined at step 725. At this point, each set of groups 650 represents records across (source systems 630 of) an organization 620 associated with a same corresponding patient. When additional levels of the hierarchy remain, the resulting sets 650 are set as the input to the patient matching process for the next level at step 730.
Subsequent levels are processed in substantially the same manner described above, where the resulting grouped patient records from the prior level are utilized as input to the succeeding level at step 730. The grouped patient records from the prior level are processed at the next level similar to the patient record (or group) as described above. In these cases, the vertices of the graphical representations for a succeeding level represent the grouped patient records from the prior level. Each additional level in the hierarchy (from source system to galaxy) provides a better representation of a person, which leads to better data integration and computer performance. For example, when a plurality of patients are matched at an organization, the information available from all the matched patients is now available to match them with patients outside of the organization in the same galaxy.
In addition, the source systems may store protected health (PHI) or other protected or confidential information. The matching process described above may control the amount of protected information exposed at every hierarchical level of matching for compliance with data governance or other restrictions. For example, the initial hierarchical levels (e.g., patient records or groups within the same organization) may use all data, but the protected information may be restricted for some of the higher hierarchical levels (e.g., galaxies or universe across different organizations). Thus, the hierarchical patient matching process enables control of protected information for data governance purposes and to comply with data governance or other policies. The amount of protected information may be specified in the configuration for a hierarchical level (e.g., as part of the blocking and/or matching criteria, additional criteria, etc.).
In particular, the resulting sets 650 for each organization 620 within the same galaxy 610 are processed in substantially the same manner described above (
The resulting collections 660 for each galaxy 610 are processed in substantially the same manner described above to determine resulting groupings of records 670 associated with the same patients. The resulting groupings of records each include the records from the corresponding collections 660 of galaxies 610, and the blocking and patient matching may employ various blocking schemes and matching algorithms that utilize the information from all the records in the collections. Since the collections of records 660 typically include plural sets of patient records, the blocking schemes and matching algorithms may employ the techniques described above for groups of patient records (e.g., maximum level of matching for a feature, blocking scheme criteria, etc.) to process a collection of records 660 as a unit (e.g., for the blocking, matching, compacting, and splitting stages). In addition, the vertices of the graphical representations for determining resulting groupings 670 represent collections of records 660 from the prior level (e.g., for the compaction and splitting stages). At this point, each resulting grouping 670 represents records across clusters of series of source systems or galaxies 610 (of universe 600) associated with a same corresponding patient.
The process terminates when no further hierarchical levels remain for processing as determined at step 725. The process may be performed for any hierarchy including any quantity of levels.
Thus, patient matching assignments or groupings 670 for galaxies 610 are constructed or aggregated from single organization patient matching assignments or groupings 640, 650. Assignments from an organization 620 are respected within a galaxy 600 (e.g., groupings 640, 650, and 660 are processed as a unit and not modified at subsequent levels), and each organization 620 and galaxy 600 may be associated with a specific patient matching configuration (e.g., blocking or matching criteria; similarity, difference, and splitting thresholds; amount of confidential information permitted/restricted; indication of permitted/restricted information; etc.) which is useful for data governance. The patient matching process described above (
The patient matching processes described above (e.g., pre-process, blocking, matching, compaction, and splitting) are typically executed in batch at off-peak periods (e.g., a nightly basis, etc.). Further, these processes are executed in a distributed computing environment (e.g., factory grid 160) and can easily scale to hundreds of millions of patients or other entities. A data lineage (e.g., identifying the source system, organization, galaxy, etc.) is maintained throughout the patient matching process, where patient assignments (or groupings) can be tracked over time. Thus, source identifiers (e.g., indicating the source systems) can be retrieved and assigned patient identifiers that are stable over time. The patient matching processes are preferably performed in the off-peak processing, where downstream analytics are preferably performed on an intra-organization level rather than a source system level.
It will be appreciated that the embodiments described above and illustrated in the drawings represent only a few of the many ways of implementing embodiments for hierarchical association of entity records from different data systems.
The environment of the present invention embodiments may include any number of computer or other processing systems (e.g., client or end-user systems, server systems, etc.) and databases or other repositories arranged in any desired fashion, where the present invention embodiments may be applied to any desired type of computing environment (e.g., cloud computing, client-server, network computing, mainframe, stand-alone systems, etc.). The computer or other processing systems employed by the present invention embodiments may be implemented by any number of any personal or other type of computer or processing system (e.g., desktop, laptop, PDA, mobile devices, etc.), and may include any commercially available operating system and any combination of commercially available and custom software (e.g., browser software, communications software, server software, patient matching module, etc.). These systems may include any types of monitors and input devices (e.g., keyboard, mouse, voice recognition, etc.) to enter and/or view information.
It is to be understood that the software (e.g., patient matching module, etc.) of the present invention embodiments may be implemented in any desired computer language and could be developed by one of ordinary skill in the computer arts based on the functional descriptions contained in the specification and flow charts illustrated in the drawings. Further, any references herein of software performing various functions generally refer to computer systems or processors performing those functions under software control. The computer systems of the present invention embodiments may alternatively be implemented by any type of hardware and/or other processing circuitry.
The various functions of the computer or other processing systems may be distributed in any manner among any number of software and/or hardware modules or units, processing or computer systems and/or circuitry, where the computer or processing systems may be disposed locally or remotely of each other and communicate via any suitable communications medium (e.g., LAN, WAN, Intranet, Internet, hardwire, modem connection, wireless, etc.). For example, the functions of the present invention embodiments may be distributed in any manner among the various end-user/client and server systems, and/or any other intermediary processing devices. The software and/or algorithms described above and illustrated in the flow charts may be modified in any manner that accomplishes the functions described herein. In addition, the functions in the flow charts or description may be performed in any order that accomplishes a desired operation.
The software of the present invention embodiments (e.g., patient matching module, etc.) may be available on a non-transitory computer useable medium (e.g., magnetic or optical mediums, magneto-optic mediums, floppy diskettes, CD-ROM, DVD, memory devices, etc.) of a stationary or portable program product apparatus or device for use with stand-alone systems or systems connected by a network or other communications medium.
The communication network may be implemented by any number of any type of communications network (e.g., LAN, WAN, Internet, Intranet, VPN, etc.). The computer or other processing systems of the present invention embodiments may include any conventional or other communications devices to communicate over the network via any conventional or other protocols. The computer or other processing systems may utilize any type of connection (e.g., wired, wireless, etc.) for access to the network. Local communication media may be implemented by any suitable communication media (e.g., local area network (LAN), hardwire, wireless link, Intranet, etc.).
The system may employ any number of any conventional or other databases, data stores or storage structures (e.g., files, databases, data structures, data or other repositories, etc.) to store information (e.g., blocking or matching criteria, thresholds, etc.). The database system may be implemented by any number of any conventional or other databases, data stores or storage structures (e.g., files, databases, data structures, data or other repositories, etc.) to store information. The database system may be included within or coupled to the server and/or client systems. The database systems and/or storage structures may be remote from or local to the computer or other processing systems, and may store any desired data.
The present invention embodiments may utilize data in any desired structure (e.g., records, data objects, data structures, etc.), and associate the data with any desired entity (e.g., person, corporate or other business entity, healthcare or other medical related entity, healthcare provider, etc.). Present invention embodiments may be applied to any hierarchical structure having any quantity of levels.
The blocking scheme may arrange any quantity or unit of records (e.g., individual records, groups of records, sets of groups, collections of sets, etc.) into blocks or sub-groups for comparisons based on any desired criteria (e.g., any type of matching of any portion of any quantity of record fields, based on record characteristics, etc.). Any quantity of records within a unit (e.g., group of records, sets of groups, collection of sets, etc.) may match the criteria in any fashion to qualify for a block. The similarity and difference thresholds may be set to any desired values.
The matching process may compare or match any quantity or unit of records (e.g., individual records, groups of records, sets of groups, collections of sets, etc.) based on any desired criteria (e.g., any quantity or combination of record fields or features, etc.). The likelihood score may include any quantity of any types of levels of matching (e.g., exact, partial, phonetic, typo, etc.). The weights may be assigned to the matching levels in any desired fashion, and include any values. The weights may be combined in any fashion to provide a likelihood score. The value of the likelihood score may be associated with any degree of similarity (e.g., a lesser likelihood score may indicate a greater chance for a match, a greater likelihood score may indicate a greater chance for a match, etc.).
Any quantity of records within a unit of records (e.g., group of records, sets of groups, collection of sets, etc.) may match to provide a match for the unit (e.g., any quantity of records for a group, any quantity of groups for a set, any quantity of sets for a collection, etc.). Further, the matching criteria may be expanded in any fashion to account for additional information at the hierarchical levels (e.g., increase variants or values provided for matching, expand matching types or levels, etc.). The matching level associated with any desired weight (e.g., greatest weight, lowest weight, etc.) may be selected for a feature of a plurality of records (e.g., group of records, set of groups, collection of sets, etc.).
The linked units (e.g., linked records, groups of records, sets of groups, collection of sets, etc.) may be represented graphically in any fashion (e.g., nodes and edges or arcs, etc.). The disjoint sets within the graphical representation of a unit (e.g., linked records, groups of records, sets of groups, collections of sets, etc.) may be identified in any fashion. The splitting may split a unit (e.g., group of records, set of groups, collection of sets, etc.) into any quantity of those units based on any desired criteria (e.g., high or low likelihood values, any suitable matching criteria, etc.). The splitting threshold may be set to any desired value.
Records may be associated with the resulting units (e.g., groups of records, sets of groups, collection of sets, etc.) based on any data (e.g., linking the actual records (e.g., pointer or other indicator, etc.), grouping record or other identifiers to associate the records with a unit, etc.). The process may include one or more from a group of the pre-processing, blocking, matching, compaction, and splitting stages, either individually or any combination.
The present invention embodiments may employ any number of any type of user interface (e.g., Graphical User Interface (GUI), command-line, prompt, etc.) for obtaining or providing information (e.g., queries, analytic results, etc.), where the interface may include any information arranged in any fashion. The interface may include any number of any types of input or actuation mechanisms (e.g., buttons, icons, fields, boxes, links, etc.) disposed at any locations to enter/display information and initiate desired actions via any suitable input devices (e.g., mouse, keyboard, etc.). The interface screens may include any suitable actuators (e.g., links, tabs, etc.) to navigate between the screens in any fashion.
The present invention embodiments are not limited to the specific tasks or algorithms described above, but may be utilized for associating data from various data systems with any type of common entity.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises”, “comprising”, “includes”, “including”, “has”, “have”, “having”, “with” and the like, when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
The descriptions of the various embodiments of the present invention have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein was chosen to best explain the principles of the embodiments, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
The present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.
The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the “C” programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
Aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14924858 | Oct 2015 | US |
| Child | 16387693 | US |
