Radiomics—a portmanteau word that blends radiology and genomics—refers to the analysis of quantitative image features in large medical databases. It enables statistics-based conclusions to be drawn on tissue characteristics, disease progression and diagnoses using radiological image data.
VA4Radiomics derives information from radiological image data, and then relates this knowledge back to the corresponding patient data. This allows the generation of patient cohorts and the visualization of individual patient attributes. These, in turn, serve medical professionals as a basis for comparison with regard to diagnosis, treatment and outcomes.
A further advantage is that physicians can, theoretically, include patients they have never seen in person—for example, where the condition in question is extremely rare. Patients can be selected not just by age or gender, but by any attribute extracted from the image data.
The aim of visual analytics methods of this kind is to help medical professionals present clinical, radiological and pathological data in a way that generates actionable insights. As Prof. Jörn Kohlhammer, Head of the Information Visualization and Visual Analytics Department at Fraunhofer IGD, explains, “Going forward, the goal is to predict which treatment method will achieve the best outcomes for the individual patient. We are currently trialing our technology with clinical partners in Germany, with the intention of helping medical practitioners to learn more from clinical data.”
Jörn Kohlhammer and his team will be presenting their most recent research findings from the VA4Radiomics project at the joint Fraunhofer booth at Medica 2017 in Düsseldorf, Germany, from Nov. 13 to 16 (Hall 10, Stand G05).