Phenoflow: A Microservice Architecture for Portable Workflow-based Phenotype Definitions

Abstract

<jats:p>Clinical phenotyping is an effective way to identify patients with particular characteristics within a population. In order to enhance the portability of a phenotype, it is often defined abstractly, with users expected to realise the phenotype computationally before executing it against a local dataset. However, complex definitions, which also provide little information about how best to implement a phenotype in practice, mean that this process is often not easy. To address this issue, we propose a new multi-layer model for a phenotype definition, which is realised as a workflow, and can be combined with different implementation units in order to produce a computable form. A novel authoring architecture, Phenoflow, supports the generation of these structured definitions. To illustrate the utility of our approach, we re-author a diabetes phenotype definition, and then compare its portability to the original definition, in the context of a population of 26,406 patients at Northwestern University.</jats:p>

Desiderata for the development of next-generation electronic health record phenotype libraries

“…In addition, the authors examined existing libraries from within their own communities—including the Phenotype Knowledge Base (PheKB) [ 12 ], CALIBER [ 13 ], Phenoflow [ 14 ], and the Concept Library [ 15 ]—to identify instances of functionality currently supporting the phenotype definition lifecycle. Common functionality provided by these libraries—which has been shown to result in reproducible, portable, and valid phenotype definitions, and thus represent best practice—was also extracted and summarized.…”

“…Like QDM/CQL, this syntax also makes provision for temporal elements (e.g., associating patient observations to an elapsed time period) but looks more holistically at the cohort relating to the phenotype being defined, through, for example, the use of specified inclusion and exclusion criteria. As a final example, Phenoflow’s workflow-based model relies on a categorized set of steps to express phenotype definitions, with the same benefits [ 14 ]. An example of a phenotype realized in a higher level modelling language (CQL) is also given in Fig.…”

Using Computable Phenotypes in Point-of-Care Clinical Trial Recruitment

“…To evaluate our approach, a novel AOMd phenotype was developed to represent the EC from the REST study. For this, we utilised the Phenoflow authoring platform, which allows the logic of a phenotype to be captured as a multi-layer, high-level representation, which can then be connected to different implementation units, and is executable across different platforms [6]. The ability for phenotypes to be executed across different platforms is essential in the Transform system, where components are executed within a variety of different clinical environments (e.g.…”

Why does human phenomics matter today?

“…In the United Kingdom, the Health Data Research UK network is building the National Phenomics Resource (https://www.hdruk. org/projects/national-phenomics-resource/) based on technologies developed in CALIBER, Sail Databank6 and Phenoflow architecture 7. 1 | THE STATE OF RESEARCH IN PHENOMICS: WHAT THIS SPECIAL ISSUE TELLS USThe separation of design and execution is one of the main researchefforts in phenotype research.…”

“…Initial efforts in building standardised phenotype repositories, such as the Phenotype Knowledge Base (PheKB), 2 UK's CALIBER, 3 Million Veterans Program (MVP), 4 and All of Us consortium 5 have attracted thousands of users within their research programmes. In the United Kingdom, the Health Data Research UK network is building the National Phenomics Resource (https://www.hdruk.org/projects/national-phenomics-resource/) based on technologies developed in CALIBER, Sail Databank 6 and Phenoflow architecture 7 …”