Linear Correlation and Regression


When conducting a radiologic research study on a sample of patients or healthy volunteers, a database to collect the data needs to be built. These data may be of various types (patient history, clinics, radiology, histopathology, etc.). In practice, we obtain the information thought to be useful to the research study for each enrolled individual. Frequently, we are interested in understanding if there are relationships between the data, i.e. in testing the existence of associations between the variables.


Regression Line Radical Prostatectomy Image Noise Association Strength Gleason Grade 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Altman DG (1991) Practical statistics for medical research. London: Chapman & HallGoogle Scholar
  2. Kuno S, Katsuta S, Inouye T et al (1998) Relationship between MR relaxation time and muscle fiber composition. Radiology 169:567–568Google Scholar
  3. Li AH, Chu YT, Yang LH (2007) More coronary artery stenosis, more cerebral artery stenosis? A simultaneous angiographic study discloses their strong correlation. Heart Vessels 22:297–302PubMedCrossRefGoogle Scholar
  4. Smith SR, Williams CE, Davies JM, Edwards RHT (1989) Bone marrow disorders: characterization with quantitative MR imaging. Radiology 172:805–810PubMedGoogle Scholar
  5. Soliani L (2007) Statistica applicata alla ricerca e alle professioni scientifiche. Manuale di statistica univariata e bivariata. Parma: Uninova-Gruppo Pegaso; 21:1–6Google Scholar
  6. van Laarhoven HWM, de Geus-Oei LF, Wiering B et al (2005) Gadopentetate dimeglu-mine and FDG uptake in liver metastases of colorectal carcinoma as determined with MR imaging and PET. Radiology 237:181–188PubMedCrossRefGoogle Scholar
  7. Wang L, Mazaheri Y, Zhang J et al (2008) Assessment of biological aggressiveness of prostate cancer: correlation of MR signal intensity with Gleason grade after radical prostatectomy. Radiology 246:168–176PubMedCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Italia 2009

Personalised recommendations