Domain Specific Modelling for Clinical Research Jim Davies, Jeremy Gibbons, Adam Milward, David Milward, Seyyed Shah , Monika Solanki, James Welch Department of Computer Science The University of Oxford October 2015
Background The Model Catalogue Future Work Outline ◮ Clinical Data for Research ◮ Model Catalogue, Case studies ◮ Future work, Conclusions Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 1/21
Background The Model Catalogue Future Work Clinical Research Data: motivation for DSLs ◮ What motivates our work: ◮ Interoperability ◮ Meta-analysis ◮ Standards ◮ Portability of datasets ◮ Capturing meaning Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 2/21
Background The Model Catalogue Future Work Clinical Data : Producers and Consumers ◮ Not just clinicians and patients ◮ Data Managers ◮ Programmers ◮ Researchers ◮ Pharmaceutical companies ◮ NHS/Department for Health divisions ◮ Auditing Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 3/21
Background The Model Catalogue Future Work Clinical Data for Research : meta analyses ‘Analyse findings from a collection of previous studies, as a whole, to find patterns or results not always obvious from the individual studies.’ 1. Formulate hypotheses 2. Identify candidate studies 3. Set incorporation criteria (quality, data availability, reproducibility etc) ◮ Determine interoperability of variables 4. Select dependent variables or summary measures tone included 5. Selection of a statistical model 6. Carry out analyses, determine correlation, avoid pub bias Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 4/21
Background The Model Catalogue Future Work Hypothetical data experiment “histological type of tumour” is an common data point collected in breast cancer studies. ◮ determine whether two data sets using this data point are compatible ◮ share the data point from one organisation to the next Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 5/21
Background The Model Catalogue Future Work Meta data structural interoperability Two clinical trials collect the data point “histological type of tumour” ◮ Study one: in-situ ductal only | tubular/cribriform | ductal grade unknown | mixed ◮ Study two: invasive ductal or no specific type | tubular | mucinous | invasive cribriform Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 6/21
Background The Model Catalogue Future Work Meta data pathway: context of data collection E1 Line 1 GP Referral Unit Referral A & E Other MDT Line 2 Line 3 Step 1 Step 1 Step 2 Step 2 E3 E3 Initial Contact Initial Contact Initial Contact Initial Contact E5 1 E4 E4 Step 46 Step 3 Could be in 9 9 E8 USS/CT, USS/CT, Step 30 Step 30 0 Biopsy,Cytology, primary or Biopsy,Cytology, E4 E3 Off-treatment Off-treatment E7 6 secondary care 8 CA125 if required CA125 if required Follow up/ Follow up/ 8 Step 23 Step 23 E3 Mainten. Therapy Mainten. Therapy 9 MDT Review MDT Review (Response) (Response) E2 E2 No Relapse Discharg Step 33 Step 47 Step 47 Step 4 Step 4 e End Test Results Test Results Test Results Test Results E5 More tests required or E11 0 review in x months E11 Step 31 Step 32 Step 32 8 No E3 Recurrence of Hospital Hospital Treatment 3 E5 Disease? Admission Admission Step 5 0 Step 48 Step 48 Step 6 No Suspect Death E5 E7 Step 24 Confirm Diagnosis Confirm Diagnosis End 2 E11 E11 E11 E11 Cancer? 7 2 3 4 6 and Site ID and Site ID Clinic appointment @ GONC Yes Step 34 No End Step 7 Step 7 E6 E6 Step 8 Step 8 Yes Step 9 Step 9 Step 10 MDT Review MDT Review Confirm MDT Confirm MDT Treatment for E12 Treatment Treatment Discuss with E3 (Pre-Treatment (Pre-Treatment Diagnosis Diagnosis Relapse 0 recommendations recommendations patient 7 Diagnosis) Diagnosis) E7 E7 and Site ID and Site ID + Line E1 E1 E8 E8 Step 25 Step 25 0 0 IDS IDS Step 36 Step 36 E7 Refer to other Refer to other 5 Step 35 Step 35 MDT including MDT including Step 11 E10 Palliative Palliative Pre-Treatment Pre-Treatment Primary or Curative Yes Primary 9 tests tests Neoadjuvant Respons (Scan,CA125) (Scan,CA125) Step 22 Step 22 e CT CT Neoadjuvan Response Response E5 E5 after #6/Completion after #6/Completion E8 t 4 4 Chemo? Chemo? 1 Step 26 Step 26 Step 37 Step 37 Get Pre-Treatment Get Pre-Treatment Step 12 Step 12 Get Surgery Get Surgery Pre-Treatment Pre-Treatment E5 E6 E6 E6 E6 E7 E7 2 Samples Samples Samples Samples tests tests 9 0 3 5 7 1 4 E11,E14,E1 x6 (Scan,CA125) (Scan,CA125) E11,E14,E1 Adjuvant 6,E17,E19, 6,E17,E19, cycles E24,E26,E2 E24,E26,E2 E10 E10 E10 1 7,E29,E31, 7,E29,E31, E58,E61,E62,E6 E83,E88,E90,E92, E41,E43,E4 4, E93,E96,E98,E100,E10 4 6 8 E5 E41,E43,E4 Step 38 Step 38 E9 E9 5,E46 E66,E68,E69,E7 3,E105,E107,E108,E110 E3 5 5,E46 0, Pre-Treatment Pre-Treatment 4 E72,E73,E76 Assessment Assessment Step 13 Step 13 Step 18 Step 18 Step 15 Step 15 Emergency Pre-Treatment Pre-Treatment Step 14 Step 14 Step 16 Step 16 On-Treatment On-Treatment Step 21 Step 27 Step 27 Get Pre-Cycle Get Pre-Cycle Neoadjuvant or Assessment @ Assessment @ Pre-Cycle tests Pre-Cycle tests Chemo Cycle Chemo Cycle Clinic Response Clinic Response Pathology Pathology Surgery Samples Samples Adjuvant? Yes Gynmo/HEC Gynmo/HEC Assessment Assessment diagnosis diagnosis Adjuvant Step 39 Step 39 E5 E8 Clinic appointment Clinic appointment 7 2 (GONC) OPA (GONC) OPA E1 E1 2 2 Step 17 Yes E3 No Complete 8 Cycle? Step 28 Step 28 Step 45 Step 45 E2 MDT Review MDT Review x3 Neoadjuvan 1 1 End of Treatment End of Treatment Step 40 Step 40 cycles t (Definitive Diag/ (Definitive Diag/ @ GONC @ GONC E2 Primary Surgery Primary Surgery Staging) Staging) E1 E1 E1 E1 E2 E2 2 2 Step 41 3 3 5 5 0 0 Step 20 Step 20 x3 additional No E4 Chemo CT Response after #3 at CT Response after #3 at 7 cycles Response Required? MDT? MDT? E2 E2 E3 E3 E1 E2 Step 44 Step 44 5 8 0 0 8 3 Step 29 Step 29 MDT Review MDT Review Step 43 Step 43 Step 42 Step 42 Completion Chemo x2 additional End of Treatment End of Treatment (Pathology (Pathology Pathology Pathology Get Surgery Get Surgery No cycles @ Gynmo @ Gynmo Diagnosis/ Diagnosis/ Diagnosis Diagnosis Samples Samples E4 E4 E4 E4 0 Staging) Staging) 2 4 4 Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 7/21
Background The Model Catalogue Future Work Expected benefits for Clinical Data researchers ◮ Compatible/comparable datasets ◮ Combine datasets ◮ Interlinking separate metadata sets ◮ Traceability/provenance links ◮ Data/metadata sharing across departments ◮ ‘Grok’: Comprehension/communication Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 8/21
Background The Model Catalogue Future Work ISO11179: Metadata Registries ◮ Design for metadata registries ◮ Registration, versioning, publishing, user roles ◮ Generic, structure-less ◮ Assumes a single conceptual model Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 9/21
Background The Model Catalogue Future Work Model Catalogue: Data Modelling Language * * Model Contents 0..1 * newVersionOf * refines 1 * * Class Element Type EnumValue * * contains 1 * Reference Primitive Enumeration ◮ ISO11179-like - registration versioning, publishing support ◮ GRAILS MVC implementation ◮ Concept domain, value domain tagging ◮ ‘Discourse’ plugin for collaborative development Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 10/21
Background The Model Catalogue Future Work Model Catalogue Modelling instanceOf Form Model Data Model Form Model Implementation Metalanguage Language Language instanceOf Language instanceOf instanceOf instanceOf Data Form Form Model Model Implementation refines implements instanceOf instanceOf Form Data Model Form Model Implementation Language Language Language instanceOf instanceOf instanceOf Data Form Form Model Model Implementation refines implements ◮ Common meta-language vs. model refinement ◮ Parameterised transforms vs derived/composed(?) models ◮ Languages and tools limit implementation Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 11/21
Background The Model Catalogue Future Work Case study I: Overview ◮ Mapping of metadata from clinical information systems Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 12/21
Background The Model Catalogue Future Work Case study I: Reuse ◮ Common measurement units, automatic reuse ◮ Factorising models Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 13/21
Background The Model Catalogue Future Work Case study I: Reporting ◮ Producing reports from data models Jim Davies et. al. Domain Specific Modelling for Clinical Research DSM2015 14/21
Recommend
More recommend
