Medical EMC

RFI Requirements for Medical Electronic Equipment

Medical electronics can be found in greatly varying electromagnetic environments. The problem is that one set of RFI requirements does not meet the needs of all types of equipment. An engineering perspective may help guide the latest regulatory efforts.

Electronics engineers tend to talk about electromagnetic interference (EMI) in medical electronics in a collective sense. But, in actuality, medical electronic equipment covers such a wide span of equipment types and environments that it is not realistic to apply a single set of requirements across the board. Some highly sensitive equipment cannot bear even a moderate EMI environment, yet other devices must be designed to withstand all the interference modern technology can produce.

In recognition of this disparity, the second draft revision of IEC 60601-1-2 has a much expanded scope, defining a number of equipment categories and environments, and clarifying requirements for each.¹ And there is more to come. Many points are yet to be expanded. Although the revision better defines the requirements, the longer document requires much closer study to determine what parts apply to a given equipment type. The draft revision is not expected to be in effect for at least another year; however, its contents and their implications are relevant now.

The Association for the Advancement of Medical Instrumentation (AAMI) has proposed this document as a national standard. The Food and Drug Administration (FDA) made a significant contribution to this draft, seeking to improve areas containing deficiencies, notably the standard's explanation for determining pass/fail criteria for each equipment type. Although FDA has recognized the current version of IEC 60601-1-2, existing guidance documents still take precedence, and reviewers still make the final decision on acceptability.² This is a bad news, good news situation. The bad news is the risk of inconsistency among reviewers. The good news is the draft's flexibility that allows reviewers to make good judgments in light of such diverse equipment-environment combinations. Such flexibility has been lacking in the first edition. This article considers, but is not limited to, the contents of the draft revision. Although the categories used here are similar to those in IEC 60601-1-2, this article elaborates on key points and provides more detailed explanation. To better understand the context of the draft revision, this article first looks at radio-frequency interference (RFI) and medical electronic equipment from an engineering standpoint. The following discussion is intended to give readers a better insight into what is behind the changes to the standard.

Key EMI Threats

The primary EMI threats to any electronic equipment are RFI (including all field-type interference), electrostatic discharge (ESD), and various power disturbances. In many cases, equipment used outside a clinical environment is battery operated, so power line disturbances are not an issue. The two basic environments involving power disturbances are residential and clinical, with hospitals being the most likely to produce power disturbances.

ESD problems are driven largely by low humidity. Therefore, dry climates are much more likely to experience ESD-related problems than moist climates. This is true in the hospital or walking down the street, so there is no significant pattern in ESD environments. However, ESD is more likely to occur in certain circumstances, such as when an operator slides across the seat of a vehicle and then touches equipment. It is interesting to note that western Europe is basically a maritime climate, and thus has minimal potential to produce much ESD. Consequently, the ESD requirements in IEC 60601-1-2 have been quite modest. The draft increases the ESD limits somewhat. However, although it's a good start, the new levels are not high enough to compensate for ESD problems in drier areas of the United States.

The most prevalent environmental variation is RFI, because the levels range from very low to very high, and equipment sensitivity to RFI also varies greatly. The remainder of this article focuses on the EMI threat of RFI.

Environment

It is important to examine the possible environments that equipment can encounter. Any equipment that accompanies patients in normal life activities will be in uncontrolled environments. IEC 60601-1-2 provides some representative environments, shown in Table I.

Environment	Location	General Characteristics
Typical healthcare	Hospital, large clinic, doctor's office	Partly controlled, covered by the general provisions of this standard.
Residential	Doctor's office, small clinic	Not controlled, healthcare professional present.
Residential	Home	Not controlled, healthcare professional not normally present.
Transport/mobile	Car, aircraft, ambulance	Not controlled, wide variations, critical receivers nearby, harsh environments for ESD, RF, and electric and magnetic fields.
Special	Operating theater, emergency room	Case-by-case examination of environment.

Table I. Electromagnetic environments derived from the draft revision of IEC 60601-1-2.

The table includes most of the environments in which medical electronic equipment would be found, with the notable exception of devices such as pacemakers, hearing aids, and motorized wheelchairs that patients would use daily. For these cases, industrial, commercial, and pedestrian environments should be added to the list of environments.

RFI Sources. The most common RFI exposure to medical electronics is from portable transmitters such as cellular telephones or handheld radios. These transmitters are generally low power. Most cell phones have a 600-mW transmitter. Most handheld radios have a 1-W transmitter, but some are as high as 5 W or more. Even though the power levels are quite low, these transmitters present one of the worst threats to medical electronics because they can be very close to patients. These low-power transmitters can produce RF fields of 50 V/m, or even more in extreme cases (e.g., a cell phone in the breast pocket of a pacemaker patient). This RFI is also the hardest to control because these portable devices are everywhere—home, office, factory, automobile, hospitals, and sidewalks.

Another primary source of RFI is mobile and vehicle-installed radios. Although these are not as common as cellular phones, they are found in some private automobiles, many commercial vehicles, and all emergency vehicles, including ambulances. These radios not only emit significant exposure to those in the vehicle, but also to people nearby—outside or in adjacent vehicles. Even though these radios are not as prevalent and don't come as close to medical electronics as cell phones, they have significantly higher power, commonly running above 15 W and up to 100 W (even higher for shipboard transmitters). The automotive industry has found that RF fields in vehicle cabs can reach 100 V/m or more.

Commercial broadcast and fixed base transmitters are much higher power, but unlike those discussed previously, these transmitters are isolated from human contact. Kimmel Gerke Associates measured the fields from the main TV transmitting towers in Shoreview, MN (just north of Minneapolis/St. Paul), and found a maximum field of 1.3 V/m on the ground. The measurements were necessarily taken below the main beam. If they had been taken from a high-rise, the fields would have probably been up to 15 V/m, but still not as high as those described above. Fixed base transmitters are often located atop water towers and higher buildings, but are generally low power. Most are 100 W or less, but some (notably maritime radio) reach up to 1 kW.

Radar transmitters often found near airports can generate significant fields, but these pose little threat because they are isolated and tightly beamed slightly upward. Although no evidence indicates any significant threat from these installations, fields near some military installations are known to be substantially higher.

Low-power telemetry units are increasingly found in hospitals. These units pose a potential threat, but the expected power levels are lower than the levels reported for the transmitters above. Telemetry units generate less than 1 V/m, unless the receiving device is almost touching the transmitting antenna.

In addition to communications transmitters, it is important to mention the industrial, scientific, and medical (ISM) sources. The most notable source is the electrosurgical unit (ESU), where the field close to the unit might exceed 30 V/m. Industrial RF heaters may run at 10 kW of power, producing fields that are potentially hazardous to both the operator and nearby electronic equipment. Proximity to this equipment is controlled, but the high field strengths affect electronic equipment at much greater distances than fields from other types of transmitters.

Equipment Types

This section examines a variety of equipment types and assesses the consequences of malfunction. IEC 60601-1-2 cites eight equipment characteristics considered relevant in establishing requirements. These characteristics, however, are not necessarily mutually exclusive.

1.Is the equipment to be used exclusively in a shielded room, or will it be exposed to the ambient RF environment?

2.Is the equipment large enough that it will be permanently installed, or will it be small enough that it can reasonably be transported from one location to another?

3.Does the medical equipment use RF energy for treatment or diagnosis?

4.Does the medical electronic equipment incorporate RF transmitters or receivers?

5.Does the equipment have at least one life-support function, intended to keep the patient alive?

6.Does the medical equipment have at least one part applied to the patient?

7.Is some performance degradation of the medical device allowed during interference?

8.Does the equipment include some very sensitive electronics, especially those monitoring physiological processes?

Each of these characteristics has some ramification for the electronic equipment and its ability to withstand interference in different environments.³

Shielded Equipment. Some medical electronic equipment is intended to be contained in a shielded enclosure—a notable case being a magnetic resonance imaging (MRI) system. Because shielded enclosures provide excellent isolation between the inside and outside, EMI immunity to outside interference does not apply to the contained equipment. Accordingly, emissions and RF susceptibility requirements could be waived. It is still necessary, however, to ensure that all components within the enclosure meet appropriate EMI requirements so that they are mutually compatible.

Large Equipment. Equipment such as a computed tomography scanner is considered large if it is big enough to be permanently installed in a fixed location. Because such equipment is installed in a hospital or large clinic, the environment is reasonably controllable. The equipment can be installed away from very sensitive or very noisy equipment. Large equipment is handled on a case-by-case basis. The RFI requirements are tailored to the particular equipment, and special installation requirements are levied.

RF Source. If equipment uses RF energy for diagnosis or treatment, containment of the energy is difficult. If RF energy is substantial, then shielding will most likely be necessary (as with the MRI). If RF energy is moderate or if equipment is used only sporadically, then a work-around solution may be possible. An ESU used in an operating room is very noisy and impossible to shield, at least from other nearby equipment. It is generally accepted that nearby equipment is not to operate when the ESU is operating, but must return to normal mode immediately afterward.

RF Transmitters and Receivers. For equipment that uses RF transmitters or receivers for its operation—as with increasingly common telemetry equipment—it is not possible to contain the RF energy. The transmitting antenna must be separated from the sensitive equipment. Telemetry is low power and does not pose a threat to most equipment. It can, however, affect very sensitive analog sensors. Likewise, receiving devices necessarily receive any interference operating within the pass band, so performance degradation may be allowed.

Life-Support Equipment. This category includes any equipment that has at least one life-support function. Such equipment might be employed in a hospital's intensive care unit (in which case the environment is fairly controllable). Some life-support devices, however, are used by patients in daily life (in which case the environment is essentially uncontrolled). Pacemakers and portable infusion pumps might encounter considerable interference, and equipment failure could have serious impact. Although patients can be warned to avoid certain environments such as industrial radio sources, it is impossible to isolate patients from vehicular or handheld transmitters encountered in normal situations such as walking down the street or shopping in a mall.

Patient-Monitoring Equipment. Monitoring equipment is connected directly to a patient either for a brief period to take measurements or for extended periods to determine when a patient's condition has stabilized. When continuous monitoring is used to observe vital signs, the equipment is categorized similarly to life-support equipment, since a malfunction could have serious consequences. However, for a monitoring device designed for use for a brief period—such as an electrocardiograph (ECG) in the doctor's office—failure would be a nuisance only.

Signal outputs from physiological processes are usually very weak, requiring sensitive receiving devices in order to work. These sensitive receivers are notoriously vulnerable to RFI. For devices such as ECGs used in clinics to gather data, interference is no more than a nuisance. The interfering RF is readily discernible and, in most cases, lasts only momentarily. Corrupted tests can simply be repeated. Equipment such as a sleep apnea monitor, however, may be monitoring patient vital signs. This equipment must function as designed, even if exposed to significant RFI. Because these devices are intended to have an operator in attendance, the environment is more controllable, but only slightly so. Precautions could include specifying that patients themselves refrain from using cell phones or handheld transmitters during monitoring. If a residence is in an apartment building, however, it is impossible to control neighbors' use of cell phones on the other side of the wall.

Safety concerns with patient-monitoring equipment conflict with taking steps to ensure RF immunity. Patients must be protected from microshock, in which leakage currents from electrical power sources can become severe enough to interfere with vital physiological functions (specifically the heart). This safety requirement makes it difficult to filter or shield undesired interference.

Performance Degradation. In some cases—usually with analog signal monitoring—performance degradation can be permitted. A modest level of interference piggybacking onto an ECG plot may be tolerable, provided the magnitude is not high enough to mask the desired signal. Digital errors, by their go/no-go nature, are rarely acceptable and are certainly unacceptable when a critical health decision is at stake.

Patient-Assist Devices. These devices include motorized wheelchairs and scooters. Most are found anywhere outside the home, but are not likely in a noisy industrial environment. The primary sources of interference to such devices are radio interference from police cars, and handheld radios in the mall. Obviously, losing control of these motorized vehicles could have serious consequences.

Diagnostic Equipment. For diagnostic equipment such as blood analyzers, operational failure is simply a nuisance, as long as the failure is detected so that tests can be repeated. Serious consequences could result from equipment that produces undetected erroneous results. Because there is some flexibility in where diagnostic equipment is located, the noisiest environments can be identified and avoided.

Assessment

As noted in several cases described above, the sensitivity of equipment varies greatly depending on its function. Many physiological processes produce very tenuous electrical signals. We continue to marvel that anyone even discovered some of the phenomena, much less came up with a way to measure them.

Some equipment will never be able to operate in a noisy environment. At a minimum, it must be recognized that any interference that lies in the pass band of the measuring device cannot be suppressed. If process bandwidth includes 60 Hz, for example, interference is very difficult to suppress unless all components can be shielded or interfering fields can be canceled. Even if a noisy environment can be avoided, the more common problem of RFI must still be addressed. Is it realistic to expect an ECG looking for microvolts to tolerate an RF source of volts?

Conclusion

With so many environments and such varying equipment sensitivities, it makes sense to tailor medical equipment to the application. This is already done for the military, avionics, and automotive arenas. One major difference is that the medical community encompasses a wider array of needs. It would be a challenge to address all of them, but doing so is necessary, even if it means creating a much more voluminous standard. Developing a standard that is too general exposes patients to risks out of proportion to actual failure. An overly complex standard, however, could increase equipment prices or drive sensitive equipment off the market if it couldn't be modified to meet excessive requirements.

It is highly unlikely that an ECG will ever achieve an immunity of 3 V/m, nor is it necessary. A motorized wheelchair must be able to operate at greater than 10 V/m, and it is realistic to achieve this. The revised IEC 60601-1-2 is taking a significant step in recognizing these operating requirements, but it has a long way to go to account adequately for all of the environment and sensitivity combinations.

References

1. IEC 60601-1-2, draft 2nd ed., "Medical Electrical Equipment, Part 1: General Requirements for Safety. 2. Collateral Standard: Electromagnetic Compatibility, Requirements and Test," International Electrotechnical Commission (IEC), Geneva.

2. IEC 60601-1-2, "Medical Electrical Equipment, Part 1: General Requirements for Safety. 2. Collateral Standard: Electromagnetic Compatibility, Requirements and Test," IEC, Geneva, April 1993.

3. William D Kimmel and Daryl D Gerke, "Electrical Interference in the Hospital Environment," Medical Device & Diagnostic Industry 17, no. 5 (1995): 97–101.

Bill Kimmel, PE, and Daryl Gerke, PE, are cofounders of the engineering consulting firm Kimmel Gerke Associates Ltd., specializing in design, troubleshooting, and training in EMI/EMC. The authors have been full-time consultants since 1987. They are both registered professional engineers and NARTE-certified ESD and EMC engineers. Kimmel has more than 35 years of experience in various EMC-related design and management positions at Sperry Corp. and Control Data Corp. He holds a BSEE from the University of Minnesota. Gerke has more than 30 years of experience in EMC-related positions in design, field engineering, and marketing at the University of Nebraska and is an avid ham operator. They can be reached at 888/EMI-GURU or via their Web site at http://www.emiguru.com. They can also be contacted by e-mail at bkimmel@emiguru.com or dgerke@emiguru.com.

Back to November/December Table of Contents