In the absence of stable variables, quantitative imaging data continues to prove unreliable in supporting clinical decisions. This was the conclusion of Adam E. Flanders, MD, of Thomas Jefferson University Hospital in Philadelphia who voiced his concerns at a June 8 session at the annual meeting of the Society for Imaging Informatics in Medicine (SIIM). Source: HealthImaging
While quantitative imaging data is being heralded as the next “big thing” in medical imaging, clinicians’ true understanding may not be keeping pace with advancing technologies, he cautioned.
“… We are giving these tools to clinicians and they are generating their own quantitative results without understanding the data behind it and then making a clinical decision support,” Flanders said.
Questions of Variability
As with any emerging technology, the potential for good may seem unlimited. However, when dealing with the intricacies of the human system, uncertainties abound. Parameters must be set. Variables must be accounted for. Both radiologists and clinicians need to agree on acceptable thresholds.
Dr. Flanders cited a number of instances when assessing normal tissue versus disease was off the mark. For example, different parameters can affect quantitative medical imaging values. Complicated measurements make it difficult to produce reliable results.
Flanders stated that the standard deviation of quantitative measures in CT brain perfusion can reach or exceed 50 percent, casting doubt about the validity of relying on these measures.
He cited a second example, explaining that widely used physiologic MRI tests depend heavily on pre-processing, filtering, pre-defined threshholding and non-standardized algorithms.
Questions of Validity
Luciano M.S. Prevedello, MD, of Brigham & Women’s Hospital in Boston took a more optimistic view. He cited a host of clinical applications: carotid artery stenosis, calcium scoring, coronary artery stenosis, lung nodule volumetry, liver and tumor volumetry, brain perfusion, CT colonoscopy and emphysema quantification, where quantitative imaging could fill a gap by extracting quantifiable features from medical images.
Despite clinicians’ demands for more data, Flanders questioned the validity of the applications. “Vendors have made it easy to create quantitative data from imaging datasets. How reliable are these? Will all vendors give us the same results on the same dataset?”
And so, the debate continues. I encourage you to join it, via a comment below.