Does wireless MRI have a future?

This question is almost impossible to answer. In the end, each participant in this session will have to decide for him/herself what he/she thinks is the most likely future of wireless RF coils. This summary and the presentation gives some arguments relevant to this future.
Before going into the arguments, the definition of wireless has to be made clear. Is a coil wireless when there are no electrical wires between the coil (array) and the rest of the system? In that case, a coil with optical fiber links for data and control between the part on the patient and the MR system would already qualify as wireless as soon as it does not need a wired connection to power the coil electronics. Apart from this power source, such coils have been produced for many years already, so we are only a small step away from a wireless coil. In the rest of the story, it is assumed, however, that, in order to qualify as truly wireless, the coil shall have no plug and socket interface with the rest of the system. Straps for mechanically fixing the coil to the patient or patient support do not count as wires. We also assume that electrical wires are allowed to exist within a multi-element coil array, otherwise, each individual receive element would have to have its own wireless power, data and control system and this would not be very practical.
A lot of feasibility work has already been done showing that wireless RF reception is feasible. It certainly works for single receive elements and small arrays and it appears feasible for extensive arrays. A number of technology steps have to be taken to enable wireless receivers. The coil electronics has to be kept in a stable operating state without a galvanic connection to a power source. Signals to control the safe operation of the coil and, most importantly, the immense data stream containing the measured RF signals have to be transmitted wirelessly. All this has been demonstrated to be possible.
However, even if there is no technological showstopper, making coils wireless has to have a purpose; otherwise, it would be foolish to spend a lot of money on developing this technology into a product. At first, one of the main reasons to start thinking about wireless coils was the fact that cables between the coils and the system constitute a safety risk. Pieces of cable can get resonant and if they are routed in close proximity to the skin of the patient, we risk RF burns. Many such accidents have been reported in the past. The problem can be mitigated by proper design but as there is an infinite number of ways the coils and their cables can be positioned relative to the RF transmit coil and the patient, there is always a risk that in a particular constellation the patient gets injured. However, a non-conducting (e.g. glass fiber) link between the coil array and the system completely eliminates this RF safety risk (provided the transfer of power is RF-safe). Therefore, RF safety can no longer be considered a compelling reason to develop wireless coils.
The next question after safety is: do wireless RF coils improve clinical outcome. Will images get better? It is unlikely that this will be the case. At best, we can expect to get the same signal to noise ratio when the signals received by the coils are transmitted without a cable or fiber between the coil and the system. If the signal is digitized on the coil (which is the most likely scenario), there will be no loss of signal content over the wireless link but phase errors may creep in if the wireless link causes some clock jitter. In addition, the wireless power transmission could introduce noise in the MR signal band and thus reduce SNR. A slight improvement in outcome could perhaps be expected if the patient is bothered less by coil cables and is more cooperative, but on the other hand, a coil which is loosely positioned on the patient can move during the scan and this can lead to artifacts. And, since it is to be expected that no signal/power cable and less mechanical strapping will go hand in hand, an improvement in image quality is not to be expected. A related question is: will the patients even notice that the coil placed on top of them does not have a cable and then prefer to be scanned on systems having this feature? Probably not if any other mechanical fixation is used.
Cable-related RF safety has already been discussed above. But the question about the future of wireless MRI has other safety related aspects. What can happen when wireless coils fail or are not handled the way they are supposed to be handled. It is likely that, following sound design procedures, we can achieve that RF transmit is inhibited as soon as the coil does not signal that the communication is working properly and that all elements are properly detuned. However, what happens if a tech uses a wireless coil of manufacturer A inside a wireless-prepared scanner of manufacturer B or in a scanner with a different field strength or in a non-wireless scanner? This scenario becomes particularly likely if coils evolve into wearable items and patient prepping is done remote from the systems. An awful amount of work has to be done, involving all major MRI manufacturers, to keep things safe if coils are no longer identified by a keyed plug, fitting only into a socket of the scanner for which they are intended.
A major factor in the decision whether or not wireless MRI is desirable is workflow. Placing the patient on the table and then getting everything set up for scanning constitutes a major fraction of a patient slot. It is, however, questionable if significant set-up time is saved if there is no longer a cable to be plugged in.
Yet another aspect of wireless MRI is the reliability of the system. The cables and plugs of coils are certainly among the parts of the coil most likely to fail. However, a wireless link, including wireless power transmission, adds quite a few extra components, which can all fail as well. Also misalignment of coils relative to the signal and power interfaces inside the magnet bore can lead to apparent malfunction which results in a service call. In summary, it is not at all certain that wireless MRI would lead to a longer mean time between service calls.
Assuming that in all preceding categories, image quality, safety, workflow and reliability, wireless MRI is demonstrated to be at least as good as or better than coils with a cable and a plug, and then the question is what price has to be paid for this. A wireless coil needs additional parts in the system (transmitters/receivers for data and control, a wireless power generator) and in each of the wireless coils. On the other hand, cables and plugs are not cheap. At least initially, even a wireless system will most likely still have sockets for wired coils, simply because many customers still have their favorite legacy or third-party specialty coils which they do not want to throw away. There are more cost-related aspects to wireless, but none of them will likely make this technology very cost-competitive.
All these above arguments are somehow subjective and not very scientific. In order to get proper answers to these questions about efficacy, safety, reliability and cost of wireless MRI a lot of work is still to be done. Simply showing that the technology works is not sufficient. And this is where the final problem of wireless MRI comes up: how to convince the management of an MRI manufacturer that this should be placed sufficiently high on the list of priorities that it gets funded and the work gets done?

Disclaimer: this ISMRM contribution cannot be construed in any way as indicative of future product development plans or the absence thereof by any manufacturer of MRI systems.

Acknowledgements

No acknowledgement found.

References

No reference found.