A Domain-specific Modelling Language for Clinical Pathways in the Realm of Multi-perspective Hospital Modelling

Goal name : string description : string MedicalGoal i dueDate : date PatientspecificMedicalGoal dueDate : date i 0..* 0..* associatedTo 1..1 1..* defines 1..* 0..* assignedTo InformationObject ... 0..* 0..* associatedTo 1..* 1..1 assignedTo 1..1 1..1 supports Figure 2. Meta model excerpt: Additional concepts associated to the concept MedicalProcess Completed and finallyNotExecuted allow to state at runtime, i.e., during the pathway-execution within a HIS, that the check/task has been completed or finally not executed in order to allow deriving the execution/completion status of the checklist. In case it is not executed within an instantiated MedicalProcess (M0 level), a JustificationStatement needs to be provided (see constraints C01 and C02 respectively). Furthermore, a Checklist may have to be extended with any number of PatientSpecificChecklistElements on the instance level, e.g., because of additional patient-specific characteristics. Here, the same applies regarding the evaluation of the checklist’s complete execution. Allocation of medical resource types (R9) is possible due to the integration of an extension of MEMO ResML providing the concept MedicalResource (Heß, 2014, pp. 91-93; Heß, Burwitz, and Schlieter, 2014, pp. 743-745), which for now is further specialised into the concepts MedicalProduct, Drug, and Aid. As the DSML4CPs is in fact extending the already existing MEMO OrgML with domain-specific concepts, it is also fully integrated with all other DSMLs being part of the MEMO method. This allows for reusing existing concepts from other DSMLs as well (e.g., from MetricM (Strecker et al., 2012)) and, whenever necessary, to design further domain-specific extensions to address additional requirements. 4.3 Language Specification: Concrete Syntax Figure 3 shows the application of the DSML4CPs based on the example of the CP Diagnostics and therapeutic strategy for Soft Tissue Sarcoma (Schütte et al., 2007). This case has been modelled together with the West German Sarcoma Centre to evaluate the developed DSML using MEMO’s modelling tool MEMOCenterNG (Gulden and Frank, 2010) with a prototypical extension allowing to create models applying the DSML4CPs. Soft Tissue Sarcoma (STS) can be considered as a rare malignant tumour disease (1.5 % of all malignant tumour diseases) with an incidence of 6 per 100,000 per year (Goldblum, Folpe, and Weiss, 2014, p. 1). The excerpt shows the situation starting with the tumour staging2 resulting 2 For detailed information on the staging of STS see Goldblum, Folpe, and Weiss (2014, pp. 3-9) and for more details on modelling the corresponding decision see Heß, Schlieter, and Täger (2012, pp. 285-286). Twenty-Third European Conference on Information Systems, Münster, Germany, 2015 11 Heß et al. / A DSML for Clinical Pathways in the Realm of Multi-Perspective Hospital Modelling A ss es s re sp on se of tu m ou r( s) an d re se ct ab ili ty