Systems, methods, and devices for indexing and searching digital imaging and communications in medicine (DICOM) metadata
Techniques, described herein, enable enhanced indexing and searching of digital imaging and communications in medicine (DICOM) systems and records. Upon receiving DICOM objects in one or more DICOM studies via an interface to medical modalities, DICOM metadata can be extracted from the DICOM object headers of the DICOM objects. DICOM values of DICOM tags from the DICOM object headers can be indexed to enable a relational database or a non-structural queried language (non-SQL) database to be searchable for each value of the DICOM tags. Additional features of the techniques are described herein.
1 . A server device, comprising:
a memory; and
one or more processors configured, when executing instructions stored in the memory, to:
receive, via an interface communicatively coupled to one or more medical modalities, digital imaging and communications in medicine (DICOM) objects in a DICOM study;
extract DICOM metadata from DICOM object headers of the DICOM objects;
index DICOM values of DICOM tags from the DICOM object headers to enable a relational database or a non-structural queried language (non-SQL) database to be searchable for each value of the DICOM tags; and
perform a search based on the indexed DICOM values to obtain DICOM results of the DICOM objects across one or more DICOM studies based on a variable combination of the DICOM tags that establish an information object hierarchy.
2 . The server device of claim 1 , wherein the one or more processors are further configured to:
modify tags of the DICOM object headers to be indexable based on a text based search or an elastic search.
3 . The server device of claim 1 , wherein the one or more processors are further configured to:
generate nested mappings of the DICOM tags with the DICOM values to enable correlations to be associated among business keys and value representations of the DICOM tags.
4 . The server device of claim 1 , wherein the one or more processors are further configured to:
generate a JSON metadata document based on a DICOM service object pair (SOP) instance unique identifier (SOPInstanceUID) of the DICOM objects and a concatenation of the SOPInstanceUID and a tenant domain to index the DICOM values in a virtual network archive (VNA).
5 . The server device of claim 1 , wherein the search is based on a simplified query language associated with a RESTful web service endpoint that is converted into an Elastic search query.
6 . The server device of claim 1 , wherein the one or more processors are further configured to:
interpret a query string for a search query of the DICOM objects;
create an elastic search query based on the query string; and
aggregate obtained results of the elastic search query in plain old common language runtime (CLR) objects (POCO) for aggregation of the DICOM objects into the one or more DICOM studies and performing clinical actions derived from obtained results of the search query.
7 . The server device of claim 1 , wherein the one or more processors are further configured to:
receive the DICOM metadata to index from the DICOM object headers; and
perform a comprehensive indexing on batches of DICOM studies that satisfy a threshold number of DICOM studies;
create metadata from each DICOM object of the batches of DICOM studies; and
send the metadata to one or more previously defined search indices with a nested mapping schema.
8 . The server device of claim 1 , wherein the one or more processors are further configured to:
compute a Levenshtein distance between a standard DICOM dictionary tag and a query language for performing a search of indexed DICOM values.
9 . A method, performed by a server device, comprising:
obtaining, via an interface communicatively coupled to one or more modalities, digital imaging and communications in medicine (DICOM) objects from one or more DICOM studies;
extracting DICOM metadata from DICOM object headers of the DICOM objects;
indexing DICOM values of DICOM tags from the DICOM object headers to enable a relational database or a non-structural queried language (non-SQL) database to be searchable for each value of the DICOM tags; and
performing a search based on the indexed DICOM values to obtain DICOM results of the DICOM objects across one or more DICOM studies based on a variable combination of the DICOM tags that establish an information object hierarchy.
10 . The method of claim 9 , wherein at least a portion of the DICOM objects comprises a header portion and an image portion, the method further comprising: generating JavaScript object notation (JSON) document or an extensible markup language (XML) document from the DICOM metadata.
11 . The method of claim 9 , further comprising:
generating nested mappings of the DICOM tags with the DICOM values to enable correlations to be associated among business keys and value representations of the DICOM tags, wherein each value of the DICOM tags of the DICOM object headers comprises a modification with identifying information.
12 . The method of claim 11 , wherein the DICOM tags comprise different value representations including one or more of: a person name, a unique identifier (UID), and short string, a long string, binary data or other tag type, and the business keys including one or more of: an issuer patient ID and patient ID/medical record number (MRN) at a patient level, a study instance UID at a study level, a series instance UID at a series level, or a DICOM service object pair (SOP) instance unique identifier (SOPInstanceUID) at an instance level, wherein the business keys enable a DICOM study to be assembled from the DICOM object headers.
13 . The method of claim 12 , further comprising:
retrieving an information object hierarchy across a plurality of DICOM studies based on a variable combination of different tags that are modified to be indexable based on the DICOM SOPInstanceUID of the DICOM objects and a concatenation of the DICOM SOPInstanceUID and a tenant domain.
14 . The method of claim 9 , further comprising:
searching the DICOM tags with the DICOM values based on nested mappings that enable correlations among business keys and value representations of the DICOM tags.
15 . A non-transitory computer-readable medium comprising:
instructions that when executed by one or more processors, cause the one or more processors to:
extract digital imaging and communications in medicine (DICOM) metadata from DICOM object headers of DICOM objects of a DICOM study of a medical modality;
index DICOM values of DICOM tags from the DICOM object headers to enable a relational database or a non-structural queried language (non-SQL) database to be searchable for each value of the DICOM tags; and
perform a search based on the indexed DICOM values to obtain DICOM results of the DICOM objects across one or more DICOM studies based on a variable combination of the DICOM tags that establish an information object hierarchy.
16 . The non-transitory computer-readable medium of claim 15 , wherein the one or more processors are further configured to:
retrieve the DICOM objects in response to a search query based on a Restful web service endpoint that interprets a search query string into an Elastic search query string.
17 . The non-transitory computer-readable medium of claim 15 , wherein the one or more processors are further configured to:
compute a Levenshtein distance between a standard DICOM dictionary tag and a query language for performing a search of indexed DICOM values.
18 . The non-transitory computer-readable medium of claim 15 , wherein the one or more processors are further configured to:
search the DICOM tags with the DICOM values based on nested mappings that enable correlations among business keys and value representations of the DICOM tags.
19 . The non-transitory computer-readable medium of claim 15 , wherein the one or more processors are further configured to:
retrieving an information object hierarchy across a plurality of DICOM studies based on a variable combination of different tags that are modified to be indexable based on a DICOM service object pair (SOP) instance unique identifier (SOPInstanceUID) of the DICOM objects and a concatenation of the SOPInstanceUID and a tenant domain.
20 . The non-transitory computer-readable medium of claim 15 , wherein the one or more processors are further configured to:
receiving the DICOM metadata to index from the DICOM object headers; and
perform a comprehensive indexing on batches of DICOM studies that satisfy a threshold number of DICOM studies.