CHAPTER 2 INTERFERENCE ISSUES OF RFID IN MRI AND CT SCAN

Transcription

1 31 CHAPTER 2 INTERFERENCE ISSUES OF RFID IN MRI AND CT SCAN This chapter explored the detailed experimental procedure adopted in this investigation for evaluating certain MRI issues with different types of RFID tags for patient identification at 0.3 T MRI and CT scans. 2.1 SCOPE AND OBJECTIVE In developed countries, it is believed that healthcare is one of the largest as well as fastest growing service industries dealing with patient s life and any shortfall in its quality of service has the potential to develop irrevocable and incurable loss to patients. Medical errors, categorized as slips or lapses, are not only detrimental to a patient s life, but also incur further healthcare expenses. The medical error may occur at any point or at any stage in hospital care process. Some of them are, incorrect patient identification, misinterpretation of hand written prescription, mislabeling of patient s bio samples, selection of wrong site for surgery, incorrect administration of the drug and improper transfusion of blood. Incorrect patient identification is the most important issue to be addressed, and utmost care must be taken to ensure correct patient identification during the treatment process in hospitals. Patient safety can be enhanced by adopting suitable IT tools in hospitals to minimize the occurrences of medical errors. Although several auto identification technologies are available, RFID is termed as the emerging technology of the last decade and it has

2 32 enhanced the quality of process in retail marketing, supply chain management, inventory control, and logistics. With the help of built in memory, each RFID tag can carry a limited amount of information about the person or object to be tagged. RFID does not require line of sight for communication and it can read multiple numbers of tags simultaneously even in adverse light conditions. By exploiting these characteristics, RFID can be employed in healthcare for numerous applications ranging from tracking and locating valuable assets, patient identification, patient tracking, medication tracking, monitoring of doctors and staff. With the introduction of RFID in healthcare, it is certain to improve the quality of medical services rendered, decrease in the cost of healthcare, improve in consistency and reliability in patient care as well as increase in nursing efficiency by reducing the burden of the nursing staff. In order to guarantee patient identification throughout the hospital stay, RFID tags must be attached to the wrists of the patients and these need not be removed till a patient leaves the hospital. During a hospital stay, the patients with RFID tags mounted on their wrists may need to undergo an MRI examination or CT scan. During these procedures, the RFID tags are exposed to harsh EM radiations from MRI, as well as high intensity X-rays of CT scan. However, patients carrying certain types of electronically activated implants, which are actually magnetically activated devices or electrical devices representing larger metallic masses and higher magnetic susceptibilities are unacceptable for MR imaging procedure. Such devices /implants in the presence of strong magnetic field interactions as well as the heating related to MR imaging process, not only change the functionalities of the devices itself, but may also produce MR imaging artifacts. Consequently, the presence of those devices/implants on patient affects the overall diagnostic use of MRI as well as the patients safety. Therefore, pre examination check by an MRI operator for all the patients should be made mandatory to ensure

3 33 that no patients carry contraindicated implants or devices to the MR imaging environment. If the device does not pose risks to patient during MRI examination, it could be labeled as MR safe and if it is able to continue its operation after an MRI examination, then it is termed as MR compatible. Various combinations of factors, including device size, orientation of the device, magnetic susceptibility of the device, and the pulse sequences performed determine the size of an artifact in MR images. Depending on its severity, the artifacts are classified as no artifact, mild artifact, moderate artifact, and strong artifact. As far as CT scan is concerned, although it could not harm the patient safety, it has the capacity to destroy the semiconductor structure of the RFID tags with its powerful X-rays. The objective of this proposed work is to evaluate the functional aspects and imaging artifacts for passive HF, MHz RFID tags, which are most suitable for medical applications. Also, the aim of this work is to assess the functional aspects and imaging artifacts for passive UHF, 900 MHz RFID tags, which are most suitable for medical applications demanding longer reading range, and higher data rates. Finally, the proposed work designed an experiment for evaluating the passive UHF RFID reader s performance kept outside the MRI copper shield area considering three important factors: angle of the antenna, tag s position and condition of the MRI scanner. 2.2 MATERIALS RFID System An RFID system contains three important components: reader (interrogator), transponder (tag), and a host computer. With the help of the software in the host computer, a reader can encode (write) or decode (read) the data in the transponders through a wireless medium. A schematic diagram

4 34 of an RFID system is given in Figure 2.1. Each tag holds a Unique Identification number (UID) of several bytes length depending on the manufacturer. The tags can be broadly classified as passive and active. Passive tags are activated by EM signals emitted from the reader antenna and scattered back to the reader. In contrast, active tags are activated by its internal battery. Passive tags have several advantages, including longer life and are economical for bulk applications. Single usage and cheaper cost involve in the implementation of passive tags for many applications. Since the lesser cost of the system and long term usage, the two significant factors governing the healthcare application, the passive RFID system becomes the suitable choice for healthcare logistics process. However, they lag in several aspects including lower transmission range (distance between reader and tag) and prone to interference from other sources. Active tags have stable transmission, wider coverage and higher detection rate when compared to its counterpart, but they have high costs and limited life span. RFID systems can operate in four different frequency bands such as LF ( KHz), HF (13.56 MHz), UHF ( MHz) and microwave frequency (2.4 GHz). Passive systems operate in LF, HF and UHF range while most of the active RFID systems operate at UHF and microwave frequencies. Figure 2.1 Schematic diagram of RFID system

5 35 The passive HF RFID system used in this proposed work is a Milfare CR 500 standard, MHz, compatible with heterogeneous standards such as ISO 144A, ISO 144B, and ISO supplied by Switzer Instruments, Chennai. The antennas of all of the RFID tags are made up of copper or aluminium, and they do not contain weak ferromagnetic materials; a prerequisite for the implant or medical device to undergo MRI testing. Table 2.1 and Table 2.2 present the technical specifications of the RFID reader and tags used in this proposed work. Table 2.1 Technical specifications of passive HF RFID reader Parameter Operating frequency Multitag read capability Communication standard Immunity to noise and interference Read/Write distance Operating temperature Transmission speed Output power Specifications MHz No USB Yes Up to 60 mm -20 o C to 50 o C bps 0.75 mw Table 2.2 Technical specifications of passive HF RFID tag Parameter Tag type Operating frequency UID Memory Specifications Wristband and ID card type MHz 4 Bytes 1 KB

6 36 The passive UHF RFID system used in this study is an EPC C1G2 standard, 900 MHz, compatible with ISO C. The antennas of all the RFID tags used in this study are made up of aluminium or copper, and they do not contain weak ferromagnetic materials; a precondition for medical device or implant to undergo MRI testing. Table 2.3 and Table 2.4 presents the technical features of the passive UHF RFID reader and tags used in this study. Table 2.3 Technical specifications of passive UHF RFID reader Parameter Specifications Operating frequency 900 MHz Multitag read capability Yes Communication standard RS 232 Immunity to noise and interference Yes Read/Write distance Up to 10 m Operating temperature -20 o C to 50 o C Transmission speed bps Output power 30 dbm Table 2.4 Technical specifications of passive UHF RFID tag Parameter Tag type Operating frequency UID Memory Specifications Wristband and ID card type 900 MHz 8 Bytes 4 KB In this proposed work, an active RFID system working at 2.4 GHz is used is supplied by the Spectrum group private limited, Mumbai. The technical

7 37 details of both the reader and the tags are summarized in Tables 2.5 and 2.6, respectively. The antenna of the tag is made up of copper material, which is considered as a weak ferromagnetic, and the batteries are made up of nickel. The reader is of desktop type and can be fixed at any point. A schematic diagram of an active RFID system is given in Figure 2.2. Figure 2.2 Schematic diagram of an active RFID system Table 2.5 Technical specifications of active RFID reader Parameter Specifications Operating frequency GHz Maximum simultaneous tag detection 200 tags Communication standard RS 232, Adaptive ethernet interface Immunity to noise and interference Yes Reading range 0-60 m Operating temperature -20 o C to 60 o C Dimension 190 mm X 120 mm X 40mm Output power +15 dbm

8 38 Table 2.6 Technical specifications of active RFID tag Parameter Tag type Operating frequency UID Output power Battery life Reading distance Specifications Wristband and ID card type GHz 4 Bytes -6 dbm 3 Year 0-80 m MRI System and CT Scan MRI is considered to be superior in several aspects when compared to several other imaging techniques available including ultrasound, X-rays, and CT scan. It is a non-invasive procedure, that does not require any contrasting agents, and can provide three-dimensional images with good tissue contrasting. The MRI scanner used in this proposed work is an active shielded 0.3 T system (AIRIS Vento III, HITACHI, Japan) with an open bore system. The MRI system used in this work is shown in Figure 2.3. The attractive feature of this MRI scanner is its ability to perform numerous imaging sequences that are exclusively associated with high-field scanners. The CT scanner used in this work was made by General Electric, USA and it is shown in Figure 2.4.

9 39 Figure T MRI scanner Figure 2.4 CT scanner

10 EXPERIMENTAL TECHNIQUE FOR PASSIVE HF AND UHF RFID Evaluation of Tag s Functionality Before conducting the test the following assumptions have been made. Several studies have already shown that certain implants (ossicular implants, Cerebrospinal Fluid (CSF) shunt valves and retinal prosthesis) or devices representing larger metallic masses and higher magnetic susceptibilities (nickel, titanium and small magnets) do not cause hazards (no displacement, negligible heating effects and no functional alterations) that may affect patients in an MRI environment of 1.5 T and 3.0 T. The only concern is artifacts, which compromise the quality of MR images, especially when the size of an artifact is comparable to the size of the device. The size of the artifacts can be reduced by optimizing the MR imaging parameters such as change in pulse sequence from Gradient-Echo (GRE) to Fast Spin-Echo (FSE) pulse sequence. The RFID tags (both HF and UHF) used in this study have smaller metallic mass with less magnetic susceptibility and they create a smaller diameter conducting loops compared to the implants used in earlier studies. Based on the above facts, it was assumed that the MRI related issues, including MRI heating and magnetic field interactions would be negligible for these RFID tags. At the same time, HF RFID tags and MRI scanner used in this study were working at close frequencies, consequently, tiny electronics of RFID tags, when placed in MRI environment, would absorb a higher amount of energy due to resonance. However, this might cause the tags to get damaged as well as make changes in the functional aspects. Similar to the earlier studies, these tags under MRI environment might also create artifacts in MR images. Similarly, UHF RFID system used in this study has better read

11 41 range and higher data rates as required for healthcare applications, the proposed work experimented the possibility of using this RFID system under 0.3 T MRI environment. Although patient safety is guaranteed with these tags, changes in functional aspects and image artifacts can impact the diagnostic use of MRI. This prompted the proposed work to undertake the current study, which evaluated two issues related to MR imaging risks including functional alterations and imaging artifacts. Thirty samples (including wristbands and card tags) of both HF and UHF RFID tags (shown in Figure 2.5) were comprehensively evaluated to determine their functional alterations. Each tag was verified for its functionality before being exposed to the MRI examination. All of them shown normal functioning. All the tags were mounted on non-conductive plastic containers in three different orientations (sagittal, axial and coronal), which is similar to the one expected during regular patient examination. The RFID reader was excluded from the evaluation process. The entire examination was carried out with one human volunteer. Informed consent was obtained from the volunteer as well as the local ethics committee of the hospital. Similar tests with a human volunteer are also reported in several studies. In order to substantiate whether harmful events are associated with the use of passive RFID tags in MRI environment, a sample test was conducted with the tags placed on the cylindrical phantom (shown in Figure 2.6). With the use of transmit-receive body Radio Frequency (RF) coil, a certain number of MRI sequences was performed. After ascertaining that no such adverse events occurred, the plastic box containing RFID tags was positioned at the center of the eight-channel wrist coil to ensure maximum exposure to all the tags. With the consented volunteer lying on the patient

12 42 table and the plastic box containing RFID tags positioned at the center of the wrist coil, an MRI was performed using four different pulse sequences representing typical techniques used for clinical MRI examination, running consecutively, for approximately four minutes per pulse sequence. The MRI sequences performed on each tag were listed in Table 2.7. After MR imaging sequences, a CT examination was also performed using a standard procedure for abdominal examination in the presence of the RFID tags with a CT technique of 100 mas at 120 kv. The exposure time for CT examination was 20 seconds. A functional check was carried out on all RFID tags outside the MRI and CT scan area after they were scanned by MRI and CT scan. The correct function of the RFID tag was characterized by three factors in three steps after exposure to the CT and MRI examination: (i) UID of each RFID tag must be readable without any error, ii) read and write operations should be carried out in all memory blocks of each tag without programmable error, and (iii) error free operation of the tag was assumed when the content that already stored in memory of each RFID tag was retrieved. Figure 2.5 Group of passive RFID tags tested

13 43 Figure 2.6 MRI phantom Table 2.7 MRI sequences performed on each passive HF and UHF tags at 0.3 T MRI for evaluating tag s functionality Pulse Séquence T1- SE T2 - SE T1- FSE T2- FSE Image Conditions TR (ms) TE (ms) Flip angle 90 o 90 o 90 o 90 o Field of view (cm) Section thickness (mm) Imaging plane Sagittal Axial Sagittal Coronal Assessment of MRI Artifacts MRI artifacts were evaluated at 0.3 T MRI system by acquiring MR images of the volunteer with and without an RFID tag attached to the wrist and placing it on wrist coil. The MRI examination was performed using 0.3 T system with the following imaging sequences listed in Table 2.8.

14 44 Table 2.8 MRI sequences performed on each passive HF and UHF tags at 0.3 T MRI for image artifacts Pulse Séquence Image Conditions T1- FSE T1 - SE TR (ms) TE (ms) Flip angle 90 o 90 o Field of view (cm) Section thickness (mm) 6 6 Imaging plane Sagittal Coronal For qualitative analysis of artifacts, the MR images acquired with RFID tags on the wrist were compared with the MR images acquired without an RFID tags Performance Assessment of Passive UHF RFID Reader Prior to conducting the experiment for testing passive UHF RFID reader s performance outside the MRI scan area, it was assumed that the EM field existed outside the MRI room was uniform and symmetrical similar to the study by Cheng & Chai Noise generated from lights, laptops and nearby metal stand were considered to be negligible. The steps followed in the experimental procedure are as follows. 1. RFID reader placed in front of the MRI machine s copper shield, at a distance of 1 m from the copper shield. The distance between the copper shield and the patient entry point is 4.3 m. With the RFID tag on the wrist, the patient s entry into MRI scan was allowed in only one direction, i.e. opposite to the position of the reader. There were no other possible

15 45 paths available for patient entry. This was in contrast to the earlier study carried out, wherein they considered four possible directions of patient entry. 2. The reading distance between tag and antenna was evaluated by varying the angle of the antenna. The antenna was positioned at three angles, 0, 45, and 90 as shown in the Figures Two types of tags (wristbands and card tags) were used. Two types of MRI machine conditions were considered; on and off. The experiment was conducted with total of 30 tags. 3. The maximum read distance for RFID tags in all angles and directions for both occasions were measured, while MRI machine was on as well as when the machine was off. Figure 2.7 Passive UHF RFID reader positioned at 0 degree angle

16 46 Figure 2.8 Passive UHF RFID reader positioned at 45 degree angle Figure 2.9 Passive UHF RFID reader positioned at 90 degree angle

17 Results After an MRI and CT scanning, all of the 30 tags tested exhibited proper function without any physical damage. It could be able to read the UID of each tag with no indication of loss of memory or loss of data. In addition, read and write operations in all memory blocks of each tag remained functional. Previously saved content could be retrieved completely without alteration. No artifacts or signal loss were found in MR images of the volunteer wearing the RFID tags on the wrist, positioned near to the skin surface. Also the quality of the images was not impaired due to the presence of RFID tags on wrist of the volunteer. The MR images of the volunteer with RFID tags (both HF and UHF) attached to the wrist are shown in Figures Artifacts were not found even in CT images, and the presence of RFID tags did not influence the quality of images captured. The CT images of the volunteer with HF and UHF RFID tags are shown in Figures 2.14 and Figure 2.10 MR image of the volunteer with passive HF RFID tag attached to the wrist (Imaging sequence: T1-SE; Imaging plane: Coronal)

18 48 Figure 2.11 MR image of the volunteer with passive HF RFID tag attached to the wrist (Imaging sequence: T1- FSE; Imaging plane:sagittal) Figure 2.12 MR image of the volunteer with passive UHF RFID tag attached to the wrist (Imaging sequence: T1-SE; Imaging plane:coronal)

19 49 Figure 2.13 MR image of the volunteer with passive UHF RFID tag attached to the wrist (Imaging sequence: T2- FSE; Imaging plane: Sagittal) Figure 2.14 CT image of wrist of the volunteer with passive HF RFID tag

20 50 Figure 2.15 CT image of wrist of the volunteer with passive UHF RFID tag During the period when the MRI machine was off, the maximum read distance for ID card type tags was more than 3.3 m with the antenna positioned at 0, 45 and 3 m at 90, while the maximum detection range of the wristband tags was 3 m at 0, 45 and 1 m at 90. On the other hand, during the period when the MRI machine was on, the maximum read distance for ID card type tags was 3 m with the antenna positioned at 0, 45 and 2.3 m at 90, whereas the maximum detection range of the wristband tags was 0.5 m at 0, 45 and about 20 cm at 90. The reading distance results are shown in Figures 2.16 and It was evident from these results that MRI machine might cause interference on RFID readability outside the copper shield. This is in contrast to the results obtained in an earlier study where the authors found no interference on RFID by MRI machine. In addition, it was also found that detection of the tag depended on the angle of the antenna and tag s position, which was in agreement with the study by Cheng & Chai (2012).

21 51 Figure 2.16 Reading range of passive UHF RFID reader at different angles during MRI is turned off Figure 2.17 Reading range of passive RFID reader at different angles during MRI is turned on

22 Discussion This proposed work was performed to assess the device safety as well as the image artifact issues pertained to the use of passive, HF, MHz and UHF, 900 MHz RFID tags in 0.3 T MRI system and CT scan at specific clinical conditions. The proposed work successfully demonstrated that the RFID tags used in this study did not experience physical damage after being subjected to several MR imaging sequences and CT scanning. Importantly, these tags also showed the capability to retain their full functionality even after being exposed to harsh EM environments coupled with MRI system and harmful X-rays of CT scan. After an MRI and CT scanning, RFID tags neither lose any data, nor has any alteration in its memory. This was also evident from earlier studies involving characterization of RFID tags operating in different frequency bands in 1.5 T or 3.0 T MRI systems. Patients wearing these RFID tags on their wrists may not need to remove while undergoing MRI and CT examination. Although this study did not specifically focus on the issues such as device heating or device movement, the volunteer neither felt heat on the surface of the skin where these RFID tags were worn nor was any device movement noticed. Interestingly, the absence of artifacts was quite amazing and encouraging. This could be due to the fact that the antennas of the RFID tags were made of either copper or aluminum having lesser magnetic susceptibility compared to titanium or nickel. Also the MRI scanner involved in this study was of low field, 0.3 T. The artifacts have been reported to appear in MR images involving higher field systems, including 1.5 T and 3.0 T in several studies. Also the size of the artifact depends on several factors; however, in most cases, they do not hamper the diagnostic quality. This study did not even find any artifacts in the CT images when the volunteer underwent scanning while wearing these RFID tags on the wrist.

23 53 With regard to device functionality and image artifact, it was found that passive HF MHz and UHF 900 MHz RFID tags are safe to use at 0.3 T MRI system and CT scan. The UHF reader s performance was degraded during the period when MRI machine was on. Additionally, poor readability was observed at an angle 90. It was also found that the type of tags used also influences the performance; ID card type cards were detectable at greater distance than the wrist band tags. There are some limitations in this study. The whole study was performed for a single frequency in both 0.3 T MRI system and CT scan by a single radiologist. Although passive RFID tags are available in several frequency bands, this study restricted the use of passive HF MHz and passive UHF 900 MHz RFID tags supplied by Switzer Instruments, Chennai and Spectrum private limited, Mumbai, India. Extensive research is required to explore the possibility of using RFID tags at other frequency ranges under 0.3 T MRI and CT scan. Although there are few limitations in this study, it ascertained the advantages of using passive HF MHz and UHF 900 MHz RFID tags on patient safely in 0.3 T MRI system and CT scan results in no artifacts in images as well as no changes in the functionality of the tag itself. 2.4 EXPERIMENTAL PROCEDURE FOR ACTIVE 2.5 GHz RFID Evaluation of Tag s Functionality Ten samples of RFID tags (five wristbands and five card tags as shown in Figure 2.18) were comprehensively evaluated to determine their functional alterations. Each tag was verified for its functionality before being exposed to the MRI examination. All the tags were mounted on non-

24 54 conductive plastic sheet (dimension 24 cm 13 cm 10 cm) in three different orientations (sagittal, axial and coronal), which is similar to the one expected during regular patient examination. The RFID reader was excluded from the evaluation process. The entire examination was carried out using cylindrical MRI phantom containing nickel solution. The plastic sheet containing RFID tags was affixed on the phantom by using glue. During the initial setup, it was found that wristband tags, which are bulkier than card tags, were attracted towards the magnet when the phantom was moved closer to the magnet. No such effects occurred with the card tags, even if they are moved closer to the magnet. Thus, wristband tags were excluded from this test and continued with card tags only. With the use of transmit-receive body RF coil, a certain number of MRI sequences listed in Table 2.9, which are representative of MRI sequences performed during regular patient examination, were performed on each RFID tag successively for an average duration of 5 minutes per sequence. After thorough exposure to the radiation, the tags were taken outside the MRI room. After MRI scanning, the CT images of the right wrist of the volunteer with and without the RFID tags (both wristbands and card tags) attached were captured using CT technique of 80 mas at 120 kv for the duration of 60 seconds. Figure 2.18 Group of active RFID tags

25 Assessment of MRI Artifacts In order to evaluate the imaging artifacts qualitatively, the imaging of the right wrist of the human volunteer was performed in two different settings. Informed consent was obtained from the volunteer as well as a local ethics committee of the hospital. In the first setting, the image of right wrist was captured when no tags were attached to any part of the body. In the second setting, the image of the right wrist was captured when the tag was attached to it. For each case, the MRI sequences listed in Table 2.9 were performed to capture the images. For qualitative analysis, the image captured with an RFID tag on the right wrist was compared with images captured without RFID tags. Based on the level of severity, the artifacts were categorized into no artifact, mild artifact, moderate artifact and strong artifact. Table 2.9 MRI sequences performed on each active RFID tag at 0.3 T MRI for evaluating tag s functionality Pulse Séquence T1-FSE T1-GE T2-FSE T2-FSE T1-SE T2-SE Image Conditions TR (ms) TE (ms) Flip angle 90 o 90 o 90 o 90 o 90 o 90 o Field of view (cm) Section thickness (mm) Imaging plane Sagittal Coronal Coronal Axial Sagittal Axial

26 56 Table 2.10 MRI sequences performed on active RFID tag at 0.3 T MRI for image artifacts Pulse Séquence Image Conditions T1-FSE T1-GE TR (ms) TE (ms) Flip angle 90 o 90 o Field of view (cm) Section thickness (mm) 5 5 Imaging plane Coronal Coronal Assessment of RFID Reader Performance It was assumed that a uniform and symmetrical field exists outside the MRI scan room as done by Cheng & Chai (2012). The RFID reader was positioned one meter away from the copper shield separating the MRI scanner area and the patient waiting room. With the help of a tripod stand and a protractor, the RFID reader can be fixed at any desirable angle. The objectives of this particular test were to find out whether the following factors have any impact on the performance of the RFID reader stationed outside the MRI scan room: i) on/off of the MRI machine; ii) angle of the RFID reader; and iii) the type of RFID tags. The test was conducted with the following steps. When the MRI machine was turned on, the patients with RFID tags attached on the wrist were requested to enter the waiting room, where the RFID reader was located. The distance between the RFID reader and the patient entry point was about 4.3 m. There was only one possible direction of entry to the waiting room, and it was opposite to the position of the RFID reader. Initially, the reader was kept at a 0 angle (vertical position as shown in Figure 2.19). When the patient with the RFID tags entered the room, the

27 57 maximum distance at which the RFID tag was detected was noted down, and the corresponding number of tag responses sent to the reader. This step was repeated for several times to confirm the reproducibility of the results. A similar set of procedures was followed when the reader was positioned at 45 and 90 angles (shown in Figures 2.20 and 2.21). When the MRI machine was turned off, the same procedures as mentioned above were repeated, and the results were noted down. Figure 2.19 Active RFID reader positioned at 0 o angle Figure 2.20 Active RFID reader positioned at 45 o angle

28 58 Figure 2.21 Active RFID reader positioned at 90 o angle Experimental Results After the MRI and CT examination, it was found that none of the card tags sustained any physical damage nor did they exhibit any malfunction. The reader was able to read the UID of each tag and perform read/write operations on each tag. The results also indicate that the contents stored in the memory of each tag can be retained. Overall, the results indicate that the functionality of the tags remains unchanged. It was found that severe artifacts were present in the MRI images of the volunteer with RFID tag attached to the right wrist when the area of imaging is close to the RFID tags. A visible signal loss or void was present in the images, and no useful information could be gathered from those images. The MR images of the volunteer with and without the RFID tags attached to right wrist are shown in Figures After CT scan, the functionality of both wristbands and card tags remain unaffected. But, the quality of CT images was affected due to the presence of both types of RFID tags when the area of imaging was close to RFID tags. However, if the area of imaging was far from the location of the RFID tags, the quality of images remained intact. The CT image of the volunteer with

29 59 RFID tag attached to the right wrist is shown in Figure The reader s performance was not affected during on/off of the MRI machine, types of tags used as well as the variation of angle of the reader. The reader s performance was constant throughout the test. Figure 2.22 MR image of wrist of the volunteer with active RFID tag (Imaging sequence: T1-FSE; Imaging plane- Coronal) Figure 2.23 MR Image of wrist of the volunteer with active RFID tag (Imaging sequence: T1-GE; Imaging plane - Coronal)

30 60 Figure 2.24 MR image of wrist of the volunteer without active RFID tag (Imaging sequence: T1-GE; Imaging plane -Coronal) Figure 2.25 CT image of wrist of the volunteer with active RFID tag (CT imaging technique: 80 mas at 120 kv)

31 Discussion In this proposed work, the functional alteration and imaging artifact issues related to the use of active RFID tags attached to patients in 0.3 T MRI and CT scans was investigated. During initial experimental setup, wristband tags were excluded from the study due to their displacement towards the magnet when they were placed closer to the magnet. Wristband tags contained larger nickel batteries than the card tags and that could have been the reason for their displacement. It was found that the functionalities of card tags remain unchanged after exposure to MRI examination. Although functional alteration of active 2.4 GHz RFID tags was not reported under 1.5 T MRI scanners in earlier studies, such investigation is required for lower field systems because of the complexities involved in the design of MRI scanners. Severe artifacts were found in the captured MR images when the ROI was close to the RFID tags. Several factors, including the static magnetic field of the scanner, the orientation of the device, the size of the device, magnetic susceptibility of the material and imaging sequences performed, can influence the artifacts in the images. This result showed that the size of the card tags and the magnetic susceptibility of nickel battery resting inside the tags could extend the severity of the artifacts. However, the severity of the artifacts can be reduced by placing the RFID tags farther away from the ROI as well as by avoiding certain MRI sequences. This proposed work also proved that the on/off condition of the MRI machine, angle of the reader and the type of tags did not affect the performance of the RFID reader stationed outside the MRI room. Moreover, these results also indicated that the EMI between the MRI scanner and the RFID reader was not problematic. Furthermore, it was found that the RFID reader can be fixed at any point and

32 62 any angle, so that any tag can be used for user s convenience. As far as CT scan was concerned, the strong presence of nickel batteries inside the tags prevented them from getting damaged due to X-rays of CT scan. At the same juncture, the presence of those nickel batteries could be the reason for presence of strong artifacts in the CT images. There are a few limitations in this proposed work. This study involved only one MRI scanner that emitted a static magnetic field strength of 0.3 T as well as one CT scanner. Furthermore, only five tags were used for the exposure to the MRI field and ten tags for X-rays of CT scan. Even though active RFIDs are available in several frequencies, this study restricted the application of RFID tags working at 2.5 GHz. The experimental procedure was limited to certain types of imaging sequences for functional analysis and imaging artifacts assessment in MRI and CT environment. 2.5 COMPARISON BETWEEN DIFFERENT STUDIES A summary of different research works undertaken with respect to RFID in MRI is given in Table From the table, it was found that, the RFID tags are safe to use in MRI environment irrespective of magnetic field strength of scanners and type of RFID tags used except in one study. The only concern is the presence of artifacts in the images that has been captured in many similar studies.

33 63 Table 2.11 Summary of various research works undertaken Author RFID Lamberg (2006) Verichip TM human implant Steffen et al (2010) Cheng & Chai (2012) Titterington & Shellock (2013) Passive, MHz Passive, 915 MHz Passive, KHz and KHz Magnetic field of MRI scanner Is there any functional change in tag? Is there any artifact in MR images? 1.5 T Yes Yes 1.5 T and 3.0 T No 1.5 T No 1.5 T and 3.0 T No Yes Not evaluated Yes Fei et al (2014) Active, 2.5 GHz 1.5 T No Yes Proposed work Passive, MHz 0.3 T No No Proposed work Passive, 920 MHz 0.3 T No No Proposed work Active, 2.5 GHz (Card tags only) 0.3 T No Yes 2.6 SUMMARY Based on research findings of the proposed work, it is evident that both passive HF, MHz and UHF, 900 MHz RFID tags worn on the patient s wrist are safe in 0.3 T MRI system as well as in CT scan and also, these tags need not be removed during MRI and CT examination. Thus, it guarantees the patient identification throughout the entire procedure with no

34 64 harmful effects either on device functionality or patient safety or quality of MRI and CT images acquired. In addition, it was also demonstrated that the performance of the RFID reader was affected by the EMI from MRI machine as well as by the angle of the antenna. Based on the clinical results obtained, it was concluded that only card type active RFID tags operating at 2.5 GHz are suitable for patient identification purpose in 0.3 T MRI environment and both wristbands as well as card tags were suitable for the same purpose in CT environment. Careful consideration must be given in both MRI and CT environments if the area of imaging is closer to RFID tags. External factors did not affect the performance of active RFID reader stationed outside the MRI scan room.