Oleg S. Pianykh, Presenter: Nothing to Disclose

Processing DPP perfusion sequence with any perfusion algorithm/software of choice, and comparing the results to the expected DPP patterns provides a robust and straightforward way to control the quality of perfusion analysis, software, and protocols. DPPs make processing errors clearly visible, and suggest approaches to correct them.

Despite the increasingly broad use of perfusion applications, we still have no generally accessible means for their verification: common sense of perfusion maps and “bona fides” of perfusion software vendors remain the only grounds for acceptance. This makes perfusion applications one of a very few clinical tools considerably lacking practical objective hands-on validation.  To solve this problem, we introduce digital perfusion phantoms (DPP) - simulated DICOM images sequences, specifically designed to have known perfusion maps with simple visual patterns.

We developed a 45-image DPP withFB, VB and TM maps consisting of simple linear patterns. Processed our DPP with popular perfusion packages produced the actual observed maps, often different from the expected DPP patterns. This comparison provided an instantaneous and objective visualization of perfusion algorithms’ deficiencies, as well as their sensitivity to noise and small details. DPP, used with different perfusion software and protocols, can clearly visualize processing errors and discrepancies.

Let S be a temporal perfusion image sequence S= {I0, I1,⋯, In,⋯IN}. We define digital perfusion phantom (DPP) as a computer-simulated image sequence S, where each temporal pixel value In(x,y)= In(x,y;tn) is generated to have predefined visual perfusion map patterns. That is, given blood flow FB, blood volume VB and mean transit time TM maps, we build DPP image sequence S to produce these maps. While any perfusion algorithm works a numerical solver for perfusion convolution equation, DPP becomes a known solution to it, which can be used to verify the algorithm correctness.To see DPPs in action, we tested them with several popular clinical perfusion packages, using brain perfusion option.

Pianykh, O, Can We Validate Perfusion Analysis? Introducing Digital Perfusion Phantoms.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12036895.html