Radiomics

from Wikipedia, the free encyclopedia

Radiomics describes a sub-area of ​​medical image processing and basic radiological research that deals with the analysis of quantitative image features in large medical image databases. The term is a portmanteau of " Radiology " and " Genomics ", based on the underlying idea that, on the basis of radiological image data, statistical statements can be made about tissue properties, diagnoses and disease processes. B. would use the genome (or proteome ).

The procedure consists primarily in the automatic extraction of a large number (for example a few hundred) of quantitative features, their visualization and statistical analysis (in particular the correlation test with clinical endpoints). In addition to statistics, machine learning is used for this.

Individual evidence

  1. ^ Robert J. Gillies, Paul E. Kinahan, Hedvig Hricak: Radiomics: Images Are More than Pictures, They Are Data . In: Radiology . tape 278 , no. 2 , November 18, 2015, p. 563-577 , doi : 10.1148 / radiol.2015151169 , PMID 26579733 , PMC 4734157 (free full text).
  2. ^ Philippe Lambin, Emmanuel Rios-Velazquez, Ralph Leijenaar, Sara Carvalho, Ruud GPM van Stiphout: Radiomics: extracting more information from medical images using advanced feature analysis . In: European Journal of Cancer (Oxford, England: 1990) . tape 48 , no. 4 , March 1, 2012, p. 441-446 , doi : 10.1016 / j.ejca.2011.11.036 , PMID 22257792 , PMC 4533986 (free full text).
  3. ↑ Chintan Parmar, Patrick Grossmann, Johan Bussink, Philippe Lambin, Hugo JWL Aerts: Machine Learning methods for Quantitative Radiomic Biomarkers . In: Scientific Reports . tape 5 , August 17, 2015, doi : 10.1038 / srep13087 , PMID 26278466 , PMC 4538374 (free full text).

literature

R. Kikinis, G. Krombach, S. Schönberg, H.-G. Stavginski: White Paper: More Quality in Personalized Medicine with Radiomics , Bremen, July 5, 2016