- The essential role of post-processing steps in extracting information from dynamic contrast-enhanced data and dynamic susceptibility contrast data

- The key aspects dynamic data processing, including:

- Heuristic parameterisations (eg rate of enhancement, IAUC, etc)

- Motion correction

- Concentration estimation

- Arterial input function extraction

- Dealing with artefacts

- Parameter heterogeneity, summary statistics and tissue classification

- Application examples in a range of organs and disease settings

Basic research scientists and clinicians interested in knowing how to maximize and interpret the information that may be extracted from dynamic contrast-enhanced and dynamic susceptibility contrast methods.This lecture will cover the main stages in processing contrast-enhanced dynamic time series of images in order to allow optimum information extraction. Examples will be drawn from a range of organs and disease settings, including neurological, cancer, pulmonary and musculoskeletal applications. By the end of the lecture, the audience will be able to appreciate the importance of processing methods when working with contrast-enhanced imaging, their role in improving signal quality and improving both the precision and accuracy of quantitative outputs. They will be able to identify situations where such processing methods are helpful or essential and will understand some of the key methods available to them.To educate participants in the methods available to enhance and extract information from dynamic contrast-enhanced and dynamic susceptibility contrast data and the importance of the application of these methods.The use of contrast agents allows quantification of physiologically-relevant phenomena that reflect the status of tissue microvasculature. Sensitisation to the presence of contrast agent can be achieved using either T2*-weighted (dynamic susceptibility contrast, DSC) or T1-weighted (dynamic contrast-enhanced, DCE) methods, with a common theme of a time series of images being acquired in order to monitor the temporal dynamics of the signal changes induced by the presence of an intravenously-administered contrast agent within the tissue of interest. A suitably sampled time series then allows inference of physiological parameters via a process of tracer kinetic modelling or heuristic analysis. The aim of extracting physiological information from dynamic time series of contrast-enhanced images generally requires a number of data analysis steps. Some of these involve ‘cleaning’ the signal of unwanted artefact (for example motion effects), some involve the correction of the signal for bias (for example the correction of the influence of RF inhomogeneity), some involve extracting important adjunct information (for example an arterial input function) and some involve converting the signal into an appropriate scale for analysis (for example conversion from signal intensity to an estimate of contrast agent concentration. If these processing steps are suitably achieved, then it is possible to extract the primary parameters of interest. In addition to the outputs of tracer kinetic analysis, these include heuristic parameters, such as the time to peak of the dynamic signal or gradient of enhancement. These primary parameters may then be further processed to assess the characteristics of the tissue of interest as a whole (for example by histogram analysis).A selection of methods for correcting signal corruption due to motion and other artefacts will be presented. The justification for conversion from signal intensity to estimates of contrast agent concentration will be discussed. The importance of arterial input function estimation using direct measurement, reference tissue/blind estimation or population approximation will be presented. Methods for parameterizing contrast-enhanced time series will be introduced, with a focus on heuristic methods, such rate of enhancement, initial area under the curve (IAUC), time to peak; tracer kinetic modeling methods will be covered elsewhere in the course. The use of output parameter maps to understand tissue heterogeneity, the use summary statistics and methods for tissue classification will be touched upon.This lecture will cover each of the above main stages in processing dynamic time series of images in order to allow optimum information extraction. Examples will be drawn from a range of organs and disease settings, including neurological, cancer, pulmonary and musculoskeletal applications. By the end of the lecture, the audience will be able to appreciate the importance of processing methods when working with contrast-enhanced imaging, their role in improving signal quality and improving both the precision and accuracy of quantitative outputs.