Search

Funder: North East London Cancer Alliance, Funder: KCL/UCL Comprehensive Cancer Imaging Centre, Funder: UCLH Biomedical Research Centre, Objectives: National guidelines recommend prostate multiparametric (mp) MRI in men with suspected prostate cancer before biopsy. In this study, we explore prostate mpMRI protocols across 14 London hospitals and determine whether standardisation improves diagnostic quality. Methods: An MRI physicist facilitated mpMRI set-up across several regional hospitals, working together with experienced uroradiologists who judged diagnostic quality. Radiologists from the 14 hospitals participated in the assessment and optimisation of prostate mpMRI image quality, assessed according to both PiRADSv2 recommendations and on the ability to “rule in” and/or “rule out” prostate cancer. Image quality and sequence parameters of representative mpMRI scans were evaluated across 23 MR scanners. Optimisation visits were performed to improve image quality, and 2 radiologists scored the image quality pre- and post-optimisation. Results: 20/23 mpMRI protocols, consisting of 111 sequences, were optimised by modifying their sequence parameters. Pre-optimisation, only 15% of T2W images were non-diagnostic, whereas 40% of ADC maps, 50% of high b-value DWI and 41% of DCE-MRI were considered non-diagnostic. Post-optimisation, the scores were increased with 80% of ADC maps, 74% of high b-value DWI and 88% of DCE-MRI to be partially or fully diagnostic. T2W sequences were not optimised, due to their higher baseline quality scores. Conclusions: Targeted intervention at a regional level can improve the diagnostic quality of prostate mpMRI protocols, with implications for improving prostate cancer detection rates and targeted biopsies.

Any novel medical imaging modality that differs from previous protocols e.g. in the number of imaging channels, introduces a new domain that is heterogeneous from previous ones. This common medical imaging scenario is rarely considered in the domain adaptation literature, which handles shifts across domains of the same dimensionality. In our work we rely on stochastic generative modeling to translate across two heterogeneous domains at pixel space and introduce two new loss functions that promote semantic consistency. Firstly, we introduce a semantic cycle-consistency loss in the source domain to ensure that the translation preserves the semantics. Secondly, we introduce a pseudo-labelling loss, where we translate target data to source, label them by a source-domain network, and use the generated pseudo-labels to supervise the target-domain network. Our results show that this allows us to extract systematically better representations for the target domain. In particular, we address the challenge of enhancing performance on VERDICT-MRI, an advanced diffusion-weighted imaging technique, by exploiting labeled mp-MRI data. When compared to several unsupervised domain adaptation approaches, our approach yields substantial improvements, that consistently carry over to the semi-supervised and supervised learning settings.

The need for training data can impede the adoption of novel imaging modalities for learning-based medical image analysis. Domain adaptation methods partially mitigate this problem by translating training data from a related source domain to a novel target domain, but typically assume that a one-to-one translation is possible. Our work addresses the challenge of adapting to a more informative target domain where multiple target samples can emerge from a single source sample. In particular we consider translating from mp-MRI to VERDICT, a richer MRI modality involving an optimized acquisition protocol for cancer characterization. We explicitly account for the inherent uncertainty of this mapping and exploit it to generate multiple outputs conditioned on a single input. Our results show that this allows us to extract systematically better image representations for the target domain, when used in tandem with both simple, CycleGAN-based baselines, as well as more powerful approaches that integrate discriminative segmentation losses and/or residual adapters. When compared to its deterministic counterparts, our approach yields substantial improvements across a broad range of dataset sizes, increasingly strong baselines, and evaluation measures.

The VERDICT (Vascular, Extracellular and Restricted Diffusion for Cytometry in Tumours) technique estimates non-invasively cancer microstructure features. The clinical application of VERDICT for prostate cancer requires constraining some of the model’s parameter. This work uses the Accelerated Microstructure Imaging via Convex Optimization (AMICO) formulation for VERDICT (VERDICT-AMICO), to investigate parameter estimation for prostate tissue in an attempt to minimize the parameter constraints. We examine various dictionaries for VERDICT-AMICO enabling different levels of flexibility on the choice of parameter values. Depending on the stability of the fitting this procedure leads to the selection of a dictionary (or dictionaries) with the fewest number of model parameter constraints. Results show that with the current VERDICT imaging acquisition, the model can have an extra free parameter to fit, the extracellular diffusivity. In conclusion, the AMICO adaptation for VERDICT allowed testing of different values for the previously fixed model parameters, and helped relax assumptions of fixed extracellular diffusivity that the model currently uses for prostate cancer characterisation.

Currently, non-invasive imaging techniques such as magnetic resonance imaging (MRI) are emerging as powerful diagnostic tools for prostate cancer (PCa) characterization. This paper focuses on automated PCa classification on VERDICT (Vascular, Extracellular and Restricted Diffusion for Cytometry in Tumors) diffusion weighted (DW)-MRI, which is a non-invasive microstructural imaging technique that comprises a rich imaging protocol and a tissue computational model to map in vivo histological indices. The contribution of the paper is two fold. Firstly, we investigate the potential of automated, model-free PCa classification on raw VERDICT DW-MRI. Secondly, we attempt to adapt and evaluate novel fully convolutional neural networks (FCNNs) for PCa characterization. We present two neural network architectures that adapt U-Net and ResNet-18 to the PCa classification problem. We train the networks end-to-end on DW-MRI data and evaluate the diagnostic performance employing a 10-fold cross validation approach using data acquired from 103 patients. ResNet-18 outperforms U-Net with an average AUC of \(86.7\%\). Our results show promise for the utilization of raw VERDICT DW-MRI data and FCNNs for automating the PCa diagnostic pathway.

© Springer International Publishing AG 2017. MR image acquisition of moving organs remains challenging despite the advances in ultra-fast 2D MRI sequences. Post-acquisition techniques have been proposed to increase spatial resolution a posteriori by combining acquired orthogonal stacks into a single, high-resolution (HR) volume. Current super-resolution techniques classically rely on a two-step procedure. The volumetric reconstruction step leverages a physical slice acquisition model. However, the motion correction step typically neglects the point spread function (PSF) information. In this paper, we propose a PSF-aware slice-to-volume registration approach and, for the first time, demonstrate the potential benefit of Super-Resolution for upper abdominal imaging. Our novel reconstruction pipeline takes advantage of different MR acquisitions clinically used in routine MR cholangio-pancreatography studies to guide the registration. On evaluation of clinically relevant image information, our approach outperforms state-of-the-art reconstruction toolkits in terms of visual clarity and preservation of raw data information. Overall, we achieve promising results towards replacing currently required CT scans. ispartof: pages:3-13 ispartof: Lecture Notes in Computer Science vol:10129 pages:3-13 ispartof: Workshop on Reconstruction and Analysis of Moving Body Organs (RAMBO) / Workshop on Whole-Heart and Great Vessel Segmentation from 3D Cardiovascular MRI in Congenital Heart Disease (HVSMR) location:Athens: GREECE date:17 Oct 2016 status: published