Dr Nikos Dikaios awarded ISMRM

Monday 15 May 2017

Dr Nikos Dikaios from the Centre for Vision, Speech and Signal Processing (CVSSP) has just been awarded with the ISMRM (International Society for Magnetic Resonance in Medicine) Summa Cum Laude Merit Award for his work on ‘Deep Learning to Improve Prostate Cancer Diagnosis’.



Multiparametric MRI (mp-MRI) can localise tumour within the prostate, guide biopsy, and assess disease burden. Nevertheless, mp-MRI itself remains imperfect. Almost 40 per cent of mp-MRI studies are reported as indeterminate for significant cancer. An indeterminate mp-MRI confers no patient benefit, such patients require either repeat interval mp-MRI and/or subsequent biopsy. There remains a clear unmet need to improve diagnostic imaging over and above standard mp-MRI protocols. Previous work has developed zone specific logistic regression (LR) models for the determination of significant cancer (any cancer-core-length (CCL) with Gleason>3+3 or any grade with CCL≥4 mm) in the peripheral (PZ) and the transition zone (TZ) based on quantitative mp-MRI parameters following MR imaging. 

This work proposes a state-of-art deep learning method to detect prostate cancer both in the PZ and the TZ. The proposed model is trained on a cohort of patients imaged at a 3T scanner, and validated on independent cohort of patients imaged at a 1.5T scanner. The performance of the model was compared with LR, Support Vector Machines and traditional Neural Networks. Deep learning outperformed the other models in terms of diagnostic performance, and exhibited high accuracy in areas scored as indeterminate by the radiologist.

Page Owner: rm0045
Page Created: Monday 15 May 2017 10:37:19 by rm0045
Last Modified: Tuesday 16 May 2017 17:32:50 by rm0045
Expiry Date: Wednesday 15 August 2018 10:32:22
Assembly date: Thu Oct 19 09:02:55 BST 2017
Content ID: 170551
Revision: 3
Community: 1379

Rhythmyx folder: //Sites/surrey.ac.uk/CVSSP/news/stories/cvssp_and_ee/2017
Content type: rx:News