Fraunhofer Institute for Computer Graphics Research IGD
Psoriatic arthritis (PsA) is a complex chronic inflammatory disease. Diagnosis and treatment require the consideration of a wide range of information, from clinical findings and laboratory results to medical imaging data, comorbidities, and each individual’s experience of the disease. However, the influence of sex and gender on disease progression and treatment outcomes is not yet always sufficiently visible.
In the PsA-vINGEST project, the project partners are developing interactive digital technologies that combine different types of health data and present sex- and gender-specific relationships in an understandable way. Two user-oriented dashboards are intended to support healthcare professionals in analysis and treatment planning, while also giving patients a better understanding of their individual disease progression and treatment journey.
Gender-Sensitive Precision Medicine for Psoriatic Arthritis
People with psoriatic arthritis can differ considerably in terms of their symptoms, comorbidities, burdens, and responses to treatment. To enable more personalized care, it is therefore important to systematically investigate possible sex- and gender-specific patterns in the available data.
To achieve this, PsA-vINGEST combines clinical parameters, laboratory analyses, medication data, disease-specific scores, and medical imaging data with so-called patient-reported outcomes (PROs). These include patients’ own assessments of pain, fatigue, quality of life, physical functioning, and psychosocial burden.
The technological requirements are being developed jointly with healthcare professionals and patients. The aim is to create a solution that combines scientific findings with the needs of future users and can be meaningfully integrated into everyday clinical practice.
Two Dashboards for Different Requirements
The project is developing two web-based dashboard versions. The dashboard for healthcare professionals combines clinical data, imaging information, and analysis results. Interactive visualizations are intended to enable users to compare patient cohorts, track disease progression, and identify sex- and gender-specific similarities or differences.
The patient dashboard focuses on particularly simple and, wherever possible, mobile use. It is intended to present personal health data and treatment progress in an understandable way, while also supporting communication between patients and healthcare professionals.
Both applications are based on a shared technical foundation but are specifically tailored to the respective user group in terms of the scope of information, interaction options, and presentation.
The Role of Fraunhofer IGD: Making Health Data Understandable and Usable
Fraunhofer IGD is responsible for the technical and scientific development of the interactive visualizations and the two dashboards. It contributes its many years of expertise in medical image processing, AI-based data analysis, interactive cohort analysis, and the development of medical dashboard solutions.
Fraunhofer IGD’s work covers the entire data processing pipeline. First, the heterogeneous health data are structured, curated, and prepared for automated analysis. This involves accounting for different data formats, time-based developments, missing values, and a range of medical data sources.
The researchers then extract image-based biomarkers from routinely acquired medical imaging data. Among the methods used are radiomics techniques, which describe anatomical structures or pathological changes using measurable features. AI-based and statistical methods help identify particularly relevant characteristics and link them to clinical information.
Interactive Cohort Analysis for Data-Driven Decision-Making
One of Fraunhofer IGD’s key areas of focus is interactive cohort analysis. It enables healthcare professionals to compare selected patient groups and investigate relationships between disease characteristics, treatment, and treatment response.
Automated analytical methods and artificial intelligence can identify patterns in complex datasets. Interactive visualizations make these results understandable to users and allow them to interpret the findings from a medical perspective. In this way, hypotheses concerning sex- and gender-specific disease progression and treatment outcomes can be developed and tested on the basis of data.