<> "The repository administrator has not yet configured an RDF license."^^ . <> . . "Clinical Decision Support for Multiple Myeloma: Computer-Interpretable Guidelines and the openEHR Specification"^^ . "Clinical decision support systems provide healthcare professionals with knowledge at the point-of care to support medical decisions. Systems medicine has emerged as a discipline that integrates and analyzes data to deliver decision support for complex diseases, facing the challenge of heterogeneous clinical and omics data integration. For the disease of multiple myeloma, a gene-expression profiling report known as the GEP-R report and a case based-reasoning approach had been proposed as decision support components in a systems medicine research project.\r\n\r\nThe GEP-R report resulted from a series of analytical steps in a bioinformatics pipeline. Analysis of genomic data in bioinformatics pipelines support diagnosis and treatment decisions, yet the availability of new tools and software versions can lead to unintended side effects that affect stability and validation of results. To address this first research problem, strategies and software tools to document and archive pipelines in a reproducible way were assessed, using the GEP-R report as a test case. The formalization of guidelines from a narrative into a computer-interpretable format is known as guideline-based clinical decision support, in which knowledge is matched to patient data to provide recommendations. Within the multiple myeloma research project, the delivery of evidence-based knowledge found in clinical practice guidelines had not been explored. To address this second problem, a guideline-based application was proposed and termed as the G-CDS approach. In addition, the lack of semantic interoperability is a challenge observed in healthcare systems, which results in isolated systems unable to share and reuse data. One of the proposed solutions has been the openEHR specification, which enables the definition of health data as reusable clinical data models. It was unknown if openEHR could support an interoperable integration of three clinical decision support components proposed for multiple myeloma (GEP-R report, case-based reasoning and guideline-based application). To address this third problem, the feasibility of openEHR to represent data regarding GEP-R report results, case-based reasoning attributes and modelling of clinical guidelines was explored. Additionally, an integration combining rule-based and case-based reasoning components was investigated.\r\n\r\nSeven tasks were defined to address these three research problems: 1. reproducibility approaches for bioinformatics pipelines, 2. knowledge synthesis, 3. knowledge modelling, 4. semi-structured knowledge representation, 5. knowledge formalization, 6. knowledge localization and 7. integration of CDS approaches. Tasks 2 – 6 follow three phases for clinical decision support development and implementation. They were mainly implemented for the design of the new G-CDS application, however knowledge synthesis and modelling were also performed for the other two components.\r\n\r\nFollowing a qualitative evaluation of technical and organizational challenges for bioinformatics pipelines, snapshots and pipeline documentation were two strategies identified to improve stability, reproducibility and validation. A hybrid approach termed as the KNIME-Docker implementation, which combined the advantages of two open-source software packages, was proposed and implemented for the GEP-R pipeline to generate the GEP-R report. Afterwards, openEHR archetypes were modelled to represent 32 data points for a gene-expression profiling report, 67 data points for a guideline-based decision support application and 42 data points for a case-based reasoning algorithm. As a result, a set of 31 openEHR archetypes were defined, consisting of 13 newly authored and 10 specialized archetypes, of which 80% were shared by at least two methods and 10% by all three. This demonstrated their appropriateness as resources to enable a complementary decision support approach.\r\n\r\nFor the proposed G-CDS application, synthesized knowledge was extracted from clinical practice guidelines and formalized as computer-interpretable guideline models. Knowledge sources were the multiple myeloma guidelines published by the European Society of Medical Oncology, and a report by the Multiple Myeloma Working Group referring to the use of geriatric assessments in elderly patients to define a frailty profile and support treatment decisions. Both sources were translated into a semi- structured representation as clinical algorithms maps for decision scenarios regarding diagnosis, staging, frailty and treatment settings. Eight guideline models were authored applying the Guideline Definition Language, which is a formal language to express decision support logic as rules in a machine-readable format using openEHR archetypes. The G-CDS required 56 input data points to provide a full summary of recommendations for four decision scenarios. To validate rule logic for all guideline models, test cases were performed, resulting in 100% of execution according to expected outcomes. Lastly, features and requirements were defined to design a user interface.\r\n\r\nTo support decisions at different stages of multiple myeloma patient care, outputs provided for diagnosis, risk stratification and treatment were examined. The G-CDS application and GEP-R report support diagnosis and risk stratification, whereas the G-CDS, GEP-R report and case-based reasoning all provide relevant information to support treatment decisions. An architecture concept was described, to enable sharing of data through a data repository of openEHR archetypes. Although each component is intended to execute independently, the proposed integration facilitates to utilize output data from rule- based components as input patient attributes, in order to contribute to a case base used in a case-based reasoning algorithm.\r\n\r\nThe digitization in healthcare is bringing new opportunities for medical informatics research and innovation in clinical settings. The proposed complementary approach of clinical decision support would maintain a knowledge base for rule-based components, while using each new patient case to gradually increment the case base in a case-based reasoning component. Thus, general evidence-based knowledge as well as specific knowledge would be delivered, outweighing the disadvantages of each component. This approach resembles the way human decisions are made for complex and real-world situations, and may refine decision support outputs, accelerate the adoption of evidence-based knowledge, improve the quality of decisions and support a holistic view of diseases and patients."^^ . "2024" . . . . . . . "Blanca"^^ . "Flores Marroquin"^^ . "Blanca Flores Marroquin"^^ . . . . . . "Clinical Decision Support for Multiple Myeloma: Computer-Interpretable Guidelines and the openEHR Specification (PDF)"^^ . . . "Flores_Marroquin_Blanca_16_08_1986_Dissertation.pdf"^^ . . . "Clinical Decision Support for Multiple Myeloma: Computer-Interpretable Guidelines and the openEHR Specification (Other)"^^ . . . . . . "indexcodes.txt"^^ . . "HTML Summary of #34913 \n\nClinical Decision Support for Multiple Myeloma: Computer-Interpretable Guidelines and the openEHR Specification\n\n" . "text/html" . . . "600 Technik, Medizin, angewandte Wissenschaften"@de . "600 Technology (Applied sciences)"@en . . . "610 Medizin"@de . "610 Medical sciences Medicine"@en . .