This study was performed to evaluate patient safety as well as the reliability and data integrity of passive RFID devices under clinical conditions with MRI and CT scan. Passive RFID devices are likely to be used for applications with RFID wristbands for patient identification. Such patient identification is central to many other RFID-related processes in hospitals.
For the first time, we were able to show that the RFID tags used in this study sustained no damage after being exposed to typical everyday conditions during CT or MR examination. Reading and saving data was unaltered after clinical MRI or CT scans. Moreover, patient safety was not impaired by wearing an RFID wristband during MRI or CT examinations.
Within the last few years, the use of autoidentification technologies has rapidly entered the hospital environment . A number of RFID applications have been implemented in hospitals so far, mostly for logistic purposes such as material tracking or inventory management, according to the original applications of the technology [2, 21–23]. Only recently, more complex applications have been implemented in hospitals, for example in patient-care management processes such as blood transfusion or the prevention of wrong side surgery [1, 3, 13, 24–27]. RFID technology seems ideal for complex environments like hospitals since the technology itself is to be applied for more gainful and differentiated applications. In this context, it was shown that autoidentification systems, ideally implemented as RFID applications, can contribute to improving efficiency and patient safety [28–30].
To ensure the proper functioning of passive RFID devices, it must be shown that no memory alterations occur and thus the ability to operate after exposure to electromagnetic fields is crucial. For the use of RFID tags in MRI systems, it is a precondition that they must not contain any ferromagnetic material. Our tags contained etched aluminum and copper, respectively, which we consider to be the important factor in finding no device movements in our tests.
All transponders had been scanned and re-written successfully each time. The large version of the tags has a 50% longer range to be read and can be used for instance as personal badges, for equipment or documents. They were tested due to their larger antenna area and consequently greater potential for energy absorption than smaller-sized RFID tags. This means that two parameters were altered in this test. The size of the antenna as well as the metal (copper in the small version and aluminum in the large version) of the RFID tags. However, no RFID tag failed in the tests.
Ferromagnetic metals are used in low-frequency RFID tags (<148.5 kHz), which are not likely to be used for the purposes described here. We used RFID tags working at 13.56 MHz because these tags are considered the most appropriate for use in a hospital environment. Today, 1.5 T MR systems operating with a frequency of 64 MHz are standard; however, low-field systems with 0.5 T or even 0.3 T operate at a frequency near that used in the RFID system tested (13.56 MHz). For these low-field systems, interference is very likely but was not tested in this study! The results presented are limited to 1.5 T and 3 T MR scanners.
The quantitative measurement showed only very small artifacts of 2 - 4 mm in size. These are minor artifacts compared to the artifacts caused by implants and do not lead to impairment of the image quality. However, we recommend not placing an RFID tag directly on the region of interest in MRI examinations (e.g. skin lesion, malignant melanoma).
According to the ASTM standard, a device is considered as MR safe if it causes no known hazards to patients in all MR environments. Since the RFID tag contains conducting materials, RFID may only be MR conditional, meaning be safe under certain conditions for MR imaging during the scan. Significant increasing of temperature and unexpected strong movements are potential risk factors for patients with an implant during MRI examination. As for the measured heating of the RFID device during MRI examination, a maximum rise in temperature of about 4°C on the skin surface is not harmful, in the normal operating mode of the MR scanner. In clinical examinations, the rise in temperature would even be reduced by the cooling effect due to sweating, air convection, blood circulation and perfusion .
Furthermore, our results regarding device movement showed that such concerns are irrelevant with less than 1 N/kg for MRI examinations in MRI scanners with field strength of up to 3 T in patients wearing an RFID wristband. No torque effects could be seen.
Although the types of RFID tags tested in our study are those most likely to be used in hospitals, we are aware of the fact that there is a huge market of different types of RFID tags. These other types of RFID tags need to be tested in different MR conditions and also under other clinical conditions. Given that we used passive RFID tags, no statement can be made about active RFID tags under the clinical conditions described. However, the utilization of the much cheaper passive RFID tags in hospitals is much more likely than the more expensive active tags.