Quantitative imaging techniques are defined by the Radiologic Society of North America’s Quantitative Imaging Biomarkers Alliance (RSNA QIBA) as ”… the extraction of quantifiable features from medical images for the assessment of normal, or the severity, degree of change, or status of a disease, injury, or chronic condition relative to normal.” The use of these image-defined characteristics based on anatomic or physiological changes, regardless of the imaging modality used (i.e., CT, MRI, US or Molecular Medicine), establishes a link to patient decision-making (e.g. predictive diagnoses) and facilitates the standardized collection of new imaging information/data for analysis to determine patient safety, efficacy and outcome.
Different imaging modalities have different challenges in terms of generating robust and reliable data. Imaging standards, such as anthropomorphic calibration phantoms are often used in large multicenter trials to ensure accurate and reproducible data, especially when there are method-related uncertainties (e.g., complex lesion shapes). Multicenter imaging studies utilize a broad range of imaging equipment providers and the use of standards would allow pooling of the results across the scanner types. The study team or imaging experts will need to review the different imaging methods used to understand any potential differences or biases in terms of how the software and imaging equipment impacts the measurement of the quantitative endpoint. This includes slice thickness, resolution, smoothing, reconstruction, contrast, ROIs, greyscale, automated versus manual contouring, etc. For example, tissue characteristics analyzed quantitatively to provide a computer-generated output (e.g., tissue analysis, score, dimensions, etc.).
The advantages of using quantitative imaging in clinical research studies include the generation of mineable translational data sets and a reduction in variance across clinical site assessments. Within the context of performing independent blinded reads, quantitative imaging may reduce reader process time, time-based expenses, and quality. Additionally, their use has the potential to improve statistical power related to reducing reader errors and intra- and inter-reader variability.
Regulatory bodies have accepted the use of quantitative imaging to support primary and secondary endpoints for different and varied indications based on validated demonstration of how precisely they measure the disease process. The process of novel imaging endpoint and the validation process of new methods compared to an accepted clinical standard of care was the subject of a past blog.