Abstract #4379

Use of Texture Analysis to Predict Prostate Artery Embolization Outcomes with MR imaging

Susanna E Kallioinen¹, Terence A Jones^1,2, James Harding², Manpreet Dhillon², Sachin Modi³, Nigel Hacking³, Drew Maclean³, and Charles E Hutchinson^1,2

¹Warwick Medical School, University of Warwick, Coventry, United Kingdom, ²Radiology Department, University Hospitals Coventry & Warwickshire NHS Trust, Coventry, United Kingdom, ³Radiology Department, University Hospital Southampton NHS Foundation Trust, Southampton, United Kingdom

Pre-prostate artery embolization (PAE) magnetic resonance images (MRI) from patients with benign prostatic hypertrophy (BPH) were segmented and analysed using two different texture analysis software, qMaZda and TexRad. Percentage reduction in prostate volume and percentage reduction in the International Prostate Symptom Score (IPSS) were used as MRI based outcome measures to build models to be able to predict outcomes from PAE. MRI texture analysis using qMaZda with a linear regression model is able to somewhat predict the percentage prostate volume reduction three months after PAE but not the percentage reduction in IPSS.

This abstract and the presentation materials are available to members only; a login is required.

Join Here