The scientists and engineers who would like to understand the safety problems related with the magnetic resonance imaging.MRI is known to be a safe imaging method but all possible safety problems needs to be considered in order to ensure its safe use. Here two of these safety issues, 1) the gradient induced peripheral nerve stimulation and 2) the implant related safety problems, will be discussed.

The Faraday’s law of induction states that the change in magnetic field generates electric field. It is, therefore, no surprise to observe electric field in the human body due to rapidly changing gradient field [1, 2]. This field is usually very low and has no harmful effect to the body. Commercial MRI scanners are equipped with the gradient coils and amplifiers that can generate electric field in a level that may cause peripheral nerve stimulations (PNS) [3]. Although this may cause an unpleasant sensation in some instances, it is not harmful. The level of magnetic field should be significantly higher than the gradient field generated by the commercial scanners in order to induce pain. Even larger fields are necessary in order to pace the heart [2]. Harm may occur at that level since random pacing may cause ventricular fibrillation or cardiac arrest but again, the commercial scanners does not have capability to generate this level of field. Let’s have closer look at this safety problem.

In addition to physiological electric field, an external electric field pulse may stimulate nerves as long as the strength and duration of the pulse is high enough [4]. Gradient magnetic fields are typically designed as combination of various trapezoidal waveforms. During the ramp-up and ramp-down periods, an electric pulse is generated in the body. Therefore, the duration and amplitude of the ramp are the main elements in relation to the PNS [3]. The strength and duration limitations are determined by two parameters called the rheobase and the chronaxie time, respectively. If the strength of the electric field (which is determined by the slope of the gradient ramp) is lower than the rheobase, independent of the duration of the pulse, stimulation does not occur. In general, no pacing can be observed if the gradient ramp amplitude is divided by the rheobase is less than sum of the chronaxie time and the duration of the ramp.

In finding the value of the rheobase (therefore limits of the gradient waveforms), the design of the gradient coil is important. Fortunately, most of the gradient coil designs are similar in design and based on the requirement of largest volume coverage with acceptable level of magnetic field linearity. If one assumes the human body as a cylindrical uniform object (obviously, this assumption has problems, i.e. the human body is heterogeneous and irregular shaped) and assume an ideal gradient field, the maximum electric field inside the body is observed at the surface of the body and occurs around the position where the coil ends. It can be shown that the magnitude of the electric field at its maximum location is equal to the rate of change in the gradient field times the radial and longitudinal positions of this location. This result is in line with what we observe in human experiments [5, 6]. When, strong gradient field causes sensation not at the landmark position but about 20-30cm away from it. For example, if the landmark is at heart, twitching sensation around nose of the patient may be observed [5]. Since body is in not homogenous and its shape is irregular, it is not possible to accurately predict with a simple analytical formulation. Although it is possible to conduct some simulation studies to figure out the electric field distribution in the body, further complication comes from the fact that humans have varying sizes and shapes. Since, this effect is not harmful, a number of people are tested for this effect and observed some level of variation between the stimulation thresholds but a threshold is developed for keeping most of the patients away from the sensation.

Most of the MRI safety studies are carried out assuming that patients does not carry any device or implant in their body during MRI scanning. On the hand, growing number of patients have some sort of device or implant in their body. In this part, the effect of the implants on MRI safety will be discussed.

There are three types of magnetic fields in MRI, i.e. static magnetic field, gradient magnetic field and the RF magnetic field. Each of these may interact with implants in the body. The effect of the static magnetic field to the patient with implants is rather trivial but need some discussion. Static field may pull, or cause torque on an implant if it contains some ferromagnetic materials in it [7]. Second, if the implant contains electronic circuits that may be affected from high static field (inductors with ferrous core or reed relay) the implant may malfunction in the MRI scanner. High magnetic field may cause permanent damage to some electronic circuit elements. Newer implants are designed by considering this problem and they are designed to have no ferromagnetic materials in it.

There are some implant related safety issues related with the gradient magnetic field as well [8]. First, the gradient field may induce eddy current on an implant that contain an electronic circuit and it may affect its operation during MRI scanning and possibly cause permanent damage to the circuit. Second, the metallic implants may concentrate the electric field to certain regions in the body and therefore the gradient field may cause nerve stimulation although it would not normally cause that in the absence of implant [9]. This issue may be very critical for the cardiac pacemakers. Since the electric field concentration may occur at the tip of the implant. This effect can be eliminated by designing the implant with high electrode to case impedance.

For the patients with medical implants, the RF safety issue is the most important one since it is rather difficult understand completely and proposed solutions to this problem is complex [10]. This is especially important for active implantable medical decides such as cardiac pacemakers, deep brain stimulators. There are well-documented reports on brain damage due to the excessive heating in the brain during MRI scanning of a patient wearing a deep brain stimulator [11, 12] including a very severe one [13]. It is now known that the induced current on the implant lead has a characteristic similar to a transmission line and it has a damped wave pattern [14]. When its effective length become close to half of wavelength the current on the wire becomes maximum. Maximum heating occurs on the position where the spatial change in the current is maximum. This is typically at the electrodes of the leads [15]. In case of the deep brain stimulators, this correspond to a point deep inside the brain.

There are methods to alleviate this problem. There are some studies on the development of implants that safe to use in MRI scanners [16-18]. In addition, there are studies that modifies that MRI scanner transmission system such that no implant heating occurs [19-23].

Understanding of the safety problems are critical for continuation of the safe usage of magnetic resonance imaging in patients. In this text, the gradient induced peripheral nerve stimulation is discussed as well as the safety issues related with the patients with medical implants. As long as the safety recommendations are strictly followed, MRI will continue to be safe medical imaging method.