In Sweden the most frequently diagnosed cancer and the leading cause of cancer death in women is breast cancer (BCa) whilst in men it is prostate cancer (PCa). This picture holds more generally for the developed world except that PCa falls in behind lung cancer and colorectal cancer as the third leading cause of cancer death in men. The key strategy for improving morbidity and mortality figures is early detection together with therapies guided by accurate imaging studies.
UNDERLYING PROBLEM
The methods currently used to detect BCa and PCa—x-ray mammography (BCa), and both the digital rectal examination and the prostate-specific antigen blood test (PCa)—have known shortcomings in terms of sensitivity (ability to detect cancer) and specificity (certainty that what was detected is cancer). This has motivated the clinical use of magnetic resonance imaging (MRI) in recent years as a supplemental tool. In particular it is used for staging the cancer (i.e. assessing degree of progression), determining the most appropriate treatment, and for patient follow-up after cancer treatment.
The clinical utility of MRI, as it is presently used for BCa and PCa management, is likewise limited by poor to moderate sensitivity (PCa) and specificity. Recent research suggests that it is possible to make significant improvements in sensitivity and specificity by augmenting the MRI techniques currently used (conventional T1- and T2-weighted imaging and dynamic contrast-enhanced imaging) with newer methods (including diffusion-weighted and spectroscopic imaging). The problem with interpreting such studies is that they are invariably based on at least some qualitative assessment of the MRI data by an experienced reader. This not only brings into question the repeatability and reproducibility of the results but also whether better performance could be achieved with automated and objective assessment.
It is our hypothesis that the sensitivity and specificity of MRI for both BCa and PCa can be achieved by means of multi-technique MRI coupled with novel multi-parametric computer-assisted diagnosis (CAD) techniques.
CLINICAL BENEFITS WITH CONCEPT
The project aims at improving the detection and characterization of breast and prostate cancers using multimodal magnetic resonance imaging and quantitative image analysis.
From the point of view of the radiologist, the proposed concept—multi-technique MRI CAD—will both simplify the complex task of interpreting PCa and BCa MRI data and reduce/remove the subjectivity in performing this task. It will also aid in the management of the disease, from initial detection to treatment planning and follow-up, and reduce the incidence of unnecessary and costly surgical interventions.
From the point of view of the patient the proposed concept will mean not only improved outcomes in terms of mortality and morbidity but also a reduction in anxiety and unnecessary surgical interventions associated with false positive findings.
The outcomes of the project will be a set of multi-technique MRI protocols for BCa and PCa together with a set of MRI CAD tools for analysing data acquired using these protocols. The multi-technique MRI CAD tools may be implemented as either a stand-alone CAD system (similar to CADstream® or DynaCAD®) or as plug-ins to an extensible software platform such as OsiriX (www.osirix-viewer.com). The software will be able to process image data received from a PACS (picture archiving and communication system) or MRI scanner using the DICOM standard (medical.nema.org). The software will automatically process this data to correct for degradations (e.g. noise, bias field, motion); extract measurements/features that quantitatively characterise tissue morphology, microvasculature, microstructure, and metabolism; segment (delineate) suspicious tissue; classify this tissue; and provide the user with 2D and 3D visualisation and interaction tools.
Project Leader: Dr Andrew Mehnert, MedTech West
Co-researchers:
Prof Mikael Hellström, Sahlgrenska University hospital
Prof. Mikael Persson, Chalmers
Prof Ewert Bengtsson, Uppsala University
Prof Stuart Crozier, Univ of Queensland, Australia
Mr Dominic Kennedy, Greenslopes, Queensland, Australia
