Although MRI scanners provide 3D images of exceptional quality, the strong magnetic fields used to create these images have fluctuations that can introduce errors and disturbances in scans. Consequently, these machines need frequent calibration to maintain their image quality. Additionally, innovative scanning methods like spiral sequences that could reduce scanning time are not feasible due to the magnetic field's high levels of instability. In theory, the problem could be corrected by adding a sensor to read and map changes in the magnetic field. Doing so would allow errors to be corrected computationally. In practice, however, this has been difficult to implement due to interference from electrical-based sensors and the metals used in cables. To address this, researchers at the Danish Research Center for Magnetic Resonance (DRCMR) and the Niels Bohr Institute at the University of Copenhagen have developed a quantum optical magnetometer. This sensor measures high magnetic fields and is expected to increase the longevity of MRI scanners while improving their quality and lowering costs. A prototype of the sensor is currently operational at Hvidovre Hospital at DRCMR. The optical sensor is based on saturated absorption spectroscopy on the extreme angular-momentum states of the cesium D2 line. It has four small, separate field sensors that are distributed in the MRI scanner. One of these probes remains out of range of the magnetic field and acts as a control. The system sends laser light through fiber optic cables to the four sensors located in the scanner. The prototype of the optical sensor is operational at Hvidovre Hospital near Copenhagen, where it will be fine-tuned after data from tests is collected. Courtesy of the University of Copenhagen. Inside the sensors, light passes through a small glass container that holds the cesium gas. At a certain frequency, the gas absorbs the light, and a resonance is created in the cesium atoms. The electrons in the cesium atoms oscillate more while absorbing the light, and the light is then re-emitted as the electrons fall back into place. The light dims and the gas vapor gets brighter. If the cesium is exposed to a magnetic field, the frequency of the light will change according to the strength of the field. When fluctuations in the magnetic field occur, the optical sensor maps the location of the disturbance in the field and registers how the disturbance has affected the strength of the field. The resulting data is used to identify errors in the MRI scan. In the future, corrections to disturbed and faulty images could be made based on the data collected by the four sensors, ensuring the MRI image is accurate. “When the laser has just the right frequency while passing through the gas, there is a resonance between the waves of light and electrons in the cesium atoms,” researcher Hans Stærkind said. “But the frequency — or wavelength — at which this happens changes when the gas is exposed to a magnetic field." “In this way, we can measure the strength of the magnetic field by finding out what the right frequency is. This happens completely automatically and lightning fast by the receiving device.” The optical sensor offers continuous readout, a high sampling rate, and sensitivity and accuracy in the parts per million (ppm) range. All support electronics and optics for the system are fitted into a single, 19-inch rack to make it compact, mobile, and robust. The field sensors are fiber-coupled and made from nonmetallic components, allowing them to be easily and safely positioned inside a 7 tesla (7 T) MRI scanner. The MRI sensor, or magnetometer, uses laser light and gas to measure magnetic fields. Courtesy of the University of Copenhagen. The researchers demonstrated the capabilities of the optical sensor by measuring two different MRI sequences. To highlight the system’s potential applications in medical MRI, they showed how it can be used to reveal imperfections in the gradient coil system. Before building the optical sensor, the researchers made high-precision measurements of the coefficients for the cesium atom, which enabled magnetic fields to be inferred optically with ppm accuracy. “First, we demonstrated that it was theoretically possible, and now we have proven that it can be done in practice,” Stærkind said. “We now have a prototype that can basically make the measurements needed without disturbing the MRI scanner.” Stærkind, who is the primary architect of the optical sensor, said that the sensor needs to be fine-tuned, but has the potential to make MRI scans cheaper, better, and faster. “An MRI scanner can already produce incredible images if one takes their time,” he said. “But with the help of my sensor, it is imaginable to use the same amount of time to produce even better imagery — or spend less time and still get the same quality as today. A third scenario could be to build a cheaper scanner that, despite a few errors, could still deliver decent image quality with the help of my sensor.” So far, tests on the prototype sensor show that it is working as it should. The researchers plan to further develop the prototype to make its measurements even more accurate and enhance its ability to identify errors in scans. Although the optical sensor’s initial target market will be MRI research units, it is the hope of the team that large MRI manufacturers could adopt the new technology in the long term. Eventually, the sensor could be integrated directly into new MRI scanners. “Once the prototype has been refined in a 2.0 version and its qualities documented with plenty of data from actual scans here at the hospital, we will see where this goes,” Stærkind said. “It certainly has the potential to improve MRI scans in a unique way that can benefit doctors and, not least, patients.” The research was published in PRX Quantum (www.doi.org/10.1103/PRXQuantum.5.020320) and Physical Review X (www.doi.org/10.1103/PhysRevX.13.021036).
