An ontological approach for the exploitation of clinical data Ariane - - PowerPoint PPT Presentation
An ontological approach for the exploitation of clinical data Ariane Assl Kama 1 ,Rmy Choquet 1 , Giovanni Mels 2 , Christel Daniel 1,3 , Jean Charlet 1,3 , Marie-Christine Jaulent 1 1 INSERM, UMR_S 872, Eq. 20, Universit Pierre et Marie
1 Assises GDR I3 – Strasbourg 01/07/2010
1INSERM, UMR_S 872, Eq. 20, Université Pierre et Marie Curie, Paris, France; 2AGFA HealthCare NV, Moutstraat 100, 9000 Gent, Belgium; 3APHP, Assistance Publique des Hôpitaux de Paris, Paris, France;
2 Assises GDR I3 – Strasbourg 01/07/2010
Disclaimer: this presentation reflects solely the views of the authors and no guarantee or warranty is given that it is fit for any particular purpose. The European Commission, Directorate General Information Society and Media, Brussels, is not liable for any use that may be made of the information contained therein.
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 2
q Aggregate data stored in European hospitals q Build a unified system q Infectious disease control q Antimicrobial resistances
3 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 3
ETL Process
Extract – Transform – Load Diagnostic or Therapeutic decisions Clinical or Epidemiological research
Clinical Databases
Consolidate, archive Large data streams
Analysis Space
Decision making Data analysis Data mining
4 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 4
ID Bacterium Antibiotic Result Value 1
Amoxicilline Resistant 25 2 Escherichia coli Cefoperazone Sensitive 28 3 Pseudomonas aeruginosa Ofloxacine Intermediate 10 ID Bactérie Antibiotique R V 10 Escherichia.Coli Cefpirome R 25 20 Hafnia alvei Quinolones S 28 30 Pseudomonas aeruginosa Ofloxacine I 10
Data Sources
What are the results of susceptibility testing of E. Coli resistant to B-lacatam?
Lack of knowledge : Amoxicillin is a B-lactam
Fields “bacterium” & “bactérie” refer to the same “bacteria” concept
Using an ontology to enrich clinical data exploitation
To exploit domain knowledge throughout ontology, as a user oriented view, to query clinical data, building a SPARQL Endpoint.
5 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 5
Ressources
D2R Server / Joseki
SPARQL Endpoint
(b) SPARQL Endpoint building approach (a) Traditionnal multidimentionnal datawarehouse
6 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 6
7 Assises GDR I3 – Strasbourg 01/07/2010
Data Concepts
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 7
8 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 8
PREFIX biotop : <http://purl.org/biotop/1.0/biotop.owl#> PREFIX dco: <http://www.debugit.eu/ontology/1.0/dco.owl#> PREFIX inserm: <http://debugit1.spim.jussieu.fr/resource/biotop.owl#> SELECT DISTINCT * WHERE { GRAPH <http://debugit.eu/inserm-map.n3> { ?antibiotic1 a dco:BetalactamAntibiotic. ?bacteria a biotop:SpeciesEcherichiaColiRegion. } GRAPH <http://debugit1.spim.jussieu.fr/resource> { ?susceptibility a inserm:ResultAntibiogram; inserm:antibiogram_ID ?antibiogram; inserm:antibiotic_tested ?antibiotic1; inserm:antibiotic_RESULT ?r1. ?antibiogram a inserm:antibiogram; inserm:bacteria_analyzed ?bacteria. } }
Mapping file between data and ontology We get the corresponding URIs of the B-Lactam antibiotics and E. coli bacteria From the mapping file, specifications data from the database are recovered
9 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 9
10 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 10
20 01 20 02 20 03 20 04 20 05 20 06 2007 HEGP (Internal report) 73 85 90 91 91 91 89% DebugIT (Sparql Endpoint) 73 86 89 91 92 91 91% Total 12 44 12 44 28 53 27 80 28 38 28 50 2727 ddo:Sensitive 91 1 20 74 25 53 25 25 26 07 25 96 2479
Graphic rate of E. Coli sensitivity to Cefixim at HEGP hospital over a period of 6 years. Rate sensitivity of E. coli to Cefixim at HEGP over a period
11 Assises GDR I3 – Strasbourg 01/07/2010
q
Ontologies are used to enrich the data analysis process
q
We run fewer queries using ontology and retrieve all data
q “is-a” hierarchy was used in this study q All existing semantic relations could be used (e.g. equivalence)
q
Mapping manually done
q
Link an instance to a concept
q
Build a data definition ontology from the database information model
q
Define mapping rules between the domain ontology (DCO) an the data definition
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 11
12 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 12
13 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 13
14 Assises GDR I3 – Strasbourg 01/07/2010
MEDINFO Copenhague 22th of August 2013 Presented by Ariane Assélé Kama Slide 14