Linked Data for Improved Vaccine Information Systems
4 likes3,475 views
The document discusses the Vacseen project, which aims to improve vaccine information systems through linked data technology. It addresses challenges in vaccine storage, tracking, and supply chain visibility, highlighting the benefits of using barcodes and mobile applications for data management. Future work will focus on developing advanced linked data applications to enhance vaccine handling operations and logistics.
More Related Content
Similar to Linked Data for Improved Vaccine Information Systems (20)
Linked Data for Improved Vaccine Information Systems
- 1. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Getting VacSeen-ated: Linked Data for Improved Vaccine Information Systems Monika Solanki https://w3id.org/people/msolanki @nimonika University of Oxford Joint work with Partha S Bhattacharjee, Rahul Bhattacharyya, Isaac Ehrenberg, Sanjay Sarma Auto ID Labs, MIT
- 2. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation The expansion of vaccine programs globally... [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 3. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation ...has accentuated challenges for governments... Most developing countries are not equipped to efficiently store, track, and transport increased vaccine volumes [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 4. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation ...and manufacturers Lack of visibility and control over supply chain Limited ability for demand forecasting Exposure to counterfeiting and diversion of products Limited ability to cope with regulatory developments [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 5. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation ...and manufacturers Lack of visibility and control over supply chain Limited ability for demand forecasting Exposure to counterfeiting and diversion of products Limited ability to cope with regulatory developments [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 6. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation Several approaches are under development to address the vaccine access challenge [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 7. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation Observations Despite improved information flow, issues about interoperability and at-scale last mile tracking continue to persist. Several studies have reported the effectiveness of using barcodes in tracking vaccine consumption. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 8. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Motivation Our research focuses on the Information Technology dimension [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 9. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge Contributions Demonstrate the utility of VacSeen in authenticating barcode scans and generating attendant rich contextual information. Geolocate and classify the scans based on whether or not they were recorded by authorized personnel using authenticated devices. Exploit multiple data sources: mobile applications, barcode databases, healthcare provider databases and LOD cloud. Two (mobile/Android) applications: generate biomedical factsheets about the vaccines, identifying nearest airport to a selected scan location. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 10. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Process flow Mobile-based barcode scans on the field are retrospectively validated using a data lake formed from logistical and health information systems [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 11. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Knowledge representation EEM*: The EPCIS Event Model Vaccine transaction events as abstractions for traceability. Focuses on a tight conformance with the EPCIS 1.1* standard and Simplicity. EEM has been mapped* to PROV-O*. *http://purl.org/eem# *http://www.w3.org/ns/prov# *http://www.gs1.org/epcis *http://fispace.aston.ac.uk/ontologies/eem_prov.html M. Solanki et al. EPCIS event-based traceability in pharmaceutical supply chains via automated generation of linked pedigrees. ISWC 2014 [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 12. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Knowledge representation EEM modules Temporal time of occurrence time of record timezone offset BusinessAction Disposition Business step Transaction Product EPC EPCClass Aggregation Identifiers Input/Output Quantity List Spatial ReadPoint Location Business Location Source Destination EPCIS Event Object Event AggregationEvent TransactionEvent TransformationEvent QuantityEvent* * Deprecated in EPCIS 1.1 Structure of EEM Sensors EPC Readers GoodRelations, Collections DUL Semantic Sensor Network Ontology wgs84, vCard DUL CBVVocab PROV-O EPCISException Hierarchy of EPCISException [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 13. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Knowledge representation VacSeen1* The VacSeen1 ontology uses and extends EEM to model barcode scan, product, and personnel data Extended with existing, well known ontologies to fully address the needs of the project. Light-weight ontology with just enough formalization to enable detailed querying. Minimal rules to avoid issues pertaining to inferencing. *http://www.violinet.org/vaccineontology/ [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 14. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Knowledge representation Important steps Generation of persistent URIs Using ontology classes in RDB-RDF translation Next steps Making URIs dereferenceable Using classes from Vaccine Ontology and Vaccine Adverse Event Ontology for new applications [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 15. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: RDB→RDF using D2RQ While R2RML is the standard for RDB→RDF translation, D2RQ is used because of its similarity with RDF, extensive support, and compatibility with the former. Logically closer to RDF than RDB compared to R2RML resulting in relatively less complex queries. Supports standards such as R2RML. Extensively used and well documented. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 16. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: RDB→RDF using D2RQ Important steps Customization of mapping file using ontology terms. Enforcement of integrity constraints on data by specifying integrity constraints in the mapping file. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 17. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Data storage Triplestore: GraphDB-Lite over Sesame Graphical Interface for implementation by less technical users. Freely available. DL and rule-based reasoning support. Multiple options for storage, access, and querying. Datasets loaded and used: vaccine data, healthcare system data, and barcode scan data. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 18. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Typical SPARQL queries Latitude, Longitude, operatorID, deviceID of scan events where the GTIN scanned exists in the sc_data database, the operator ID exists in the scan_operator table, and the device ID exists in the scan_device table Query using EEM & VacSeen1 SELECT DISTINCT ?EPCISEvent ?GTIN ?operatorID ?deviceID ?latitude ?longitude WHERE { ?EPCISEvent a eem:EPCISEvent. ?EPCISEvent vacseen1:ScanID ?scanID. ?EPCISEvent eem:EPCNumber ?GTIN. ?s vacseen1:sc_data_GTIN ?GTIN. ?EPCISEvent vacseen1:scan_event_operatorID ?operatorID. ?scanOperator vacseen1:scan_operator_OperatorID ?operatorID. ?EPCISEvent vacseen1:scan_event_deviceID ?deviceID. ?scanDevice vacseen1:scan_device_DeviceID ?deviceID. ?EPCISEvent vacseen1:LatitudeOfBarcodeScanEvent ?latitude. FILTER(?latitude > ’0.0’). ?EPCISEvent vacseen1:LongitudeOfBarcodeScanEvent ?longitude. } [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 19. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Typical SPARQL queries Data about Rotavirus vaccine from DBpedia INSERT{?s ?p ?o} WHERE { SERVICE <http://dbpedia.org/sparql> { { :Rotavirus_vaccine ?p ?o. BIND(:Rotavirus_vaccine AS ?s) } UNION { ?s ?p :Rotavirus_vaccine. BIND(:Rotavirus_vaccine AS ?o) } } } [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 20. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Data storage While it is easy to get started with Sesame and do routine operations, issues can arise when using plug-ins Challenges Issues with “jsonp callbacks” because of incompatibility of GraphDB-Lite with recent releases of Sesame. Absence of access to SPARQL lifting because of aforementioned incompatibility Absent GraphDB customer-support for non-paying users Resolution Sesame mailing lists Replacement of jar files as a temporary measure Permanent solution through replacement of jsonp callback with CORS headers for cross-domain access [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 21. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Exploiting the LOD cloud Two applications: Creating biomedical factsheets about vaccines using Dbpedia. Identifying nearest airports to every scan location for logistical planning. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 22. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Scan Authentication Provides additional context for each barcode scan to get a deeper understanding about vaccine handling operations. [email protected], @nimonika Linked Data for Improved Vaccine Information Systems
- 23. https://w3id.org/vacseen SWAT4LS, 8th December 2015, Cambridge VacSeen: Next steps Future work will focus on developing more advanced Linked Data-based applications that leverage large scale and secure data [email protected], @nimonika Linked Data for Improved Vaccine Information Systems