Shorter scanning times, improved image quality: With the new parallel imaging method CAIPIRINHA, Mark Griswold, Peter Jakob and their colleagues have taken the technology of magnetic resonance imaging (MRI) a huge step forwards. Siemens Healthineers uses the technique under a license agreement in many MRI scanner models. In the category ‘”Open Innovation”, both researchers were honored by Siemens AG with the Inventor of the Year Award 2018. The prize has been awarded annually since 1995 to outstanding researchers and developers who make significant contributions to the company’s success.
Collaborative research is of immense importance to Siemens Healthineers. Substantive progress in the reduction of MR scanning times and in image quality not only requires ingenious ideas but also persistence and patience. Very often it is the result of work by a research group. As in the case of the CAIPIRINHA technique: The acronym stands for Controlled Aliasing in Parallel Imaging Results in Higher Acceleration. To accelerate scanning, the images are composed of separate slices superimposed on top of each other and captured simultaneously. In this way, CAIPIRINHA intelligently refrains from using a certain amount of data, which is subsequently computed with the help of algorithms. “It’s a combined data acquisition and reconstruction method that allows the images that were taken at the same time to be separated again, which greatly improves their quality”, explains Peter Jakob, professor of experimental physics at the University of Würzburg, an internationally renowned talent hotbed for MRI researchers. In early 1996, Peter Jakob began two years of research work at the Beth Israel Deaconess Medical Center in Boston, where he met the young high-frequency engineer, Mark Griswold, who was just beginning to specialize in MRI. “From then on, we developed a wonderful collaborative work relationship with each other”, says Jakob.
After his return, he convinced Griswold to follow him to Würzburg to do his doctorate at the chair of experimental physics there. As part of his doctoral thesis, Griswold invented one of the earliest parallel imaging techniques, the GRAPPA method (GeneRalized Autocalibrating Partial Parallel Acquisition), a predecessor of the CAIPIRINHA method. This also marked the beginning of their collaboration with Siemens Healthineers. “With GRAPPA, Siemens was able to offer the very first commercial parallel imaging method”, says Jakob. It is integrated into the TIM technology.
Siemens Healthineers and the experts from the University of Würzburg were in regular contact then, as now, to exchange research ideas with each other. “We typically have an idea in which Siemens is interested, and then carry out the research work with the help of doctoral students”, says Jakob explaining the procedure. So too with the CAIPIRINHA method.
Acceleration of scanning made possible for the first time
In so-called parallel imaging, several different receivers simultaneously record the high-frequency signals being received by the patient. As the individual location of each of the receivers is known, this additional information makes it possible to likewise decode the location of the signal sources. It also allows data undersampling and, as a result, accelerated imaging sequences. This shortens the time the examination takes, making it much more pleasant for the patient. One great advantage of this system is that the patients no longer need to hold their breath for such a long time during the scanning process to minimize movements that can lead to image distortions.
Innovative reconstruction method substitutes missing data
CAIPIRINHA takes acceleration one step further. The idea of the two researchers Jakob and Griswold was to deliberately omit data points during data acquisition to speed things up even more. Usually, this shows up in the recorded image as image anomalies. The breakthrough ingenuity of CAIPIRINHA is its ability to combine the omission of data points so specifically with interspersed phase shifts that the missing information can subsequently be reconstructed by a mathematical model – thereby eliminating image distortions. “CAIPIRINHA can be used in any kind of MR measurement, which is the beauty of this method”, notes Jakob.
Today, Jakob and Griswold no longer carry out research work with each other. “But we are still very close friends, and always happy when we see each other at the expert conferences”, says Jakob. Both of them are still important partners for Siemens Healthineers. “The CAIPIRINHA method is continuously being improved”, explains Jakob. He has been the Chairman of the Research Center for Magnetic Resonance Bavaria (MRB) since 2005, a spin-off of the Chair of Experimental Physics at the University of Würzburg. Griswold today researches and teaches at the Case Western Reserve University in Ohio, USA, where, as part of a research project with Siemens Healthineers, he heads a research group for Magnetic Resonance Fingerprinting1 (MRF), an innovative method for qualitative tissue analysis. In this capacity he also experiments with quantum computer technology in cooperation with Microsoft.