MEDICAL EMC

Mapping EMI in the Modern Hospital

Basile Spyropoulos, Dimitris Glotsos, Dimitris Batistatos, and Ioannis Marneris

An on-line EMI distribution mapping system pinpoints potential EMI threats and provides information for determining preventive action.

The electromagnetic environment of hospitals has changed dramatically over the past 20 years, and such change is ongoing. New, sophisticated electronic medical systems are being introduced continually to modern hospitals to facilitate healthcare services in the forms of diagnosis and therapy.

But without proper precautions, such benefits can be counterbalanced by potential electromagnetic interference (EMI) to the normal operations of these systems. Electromagnetic compatibility, or EMC, refers to the capability of all electronic equipment to function properly within its electromagnetic environment.1 Medical systems, like all electronic equipment, produce electromagnetic fields around them, the characteristics of which determine the type of potential interference with their electromagnetic environment. EMI-related issues can manifest not only in the form of visually recognizable disturbances, but also in the most complex and dangerous forms of "obscure opaque influence" that can create life-threatening conditions for patients.2

Consequently, considerable research has centered on gathering representative data for the determination of hospital EMI levels.3-5 Research groups have also focused their attention on EMI-critical equipment, including defibrillators, electrosurgical units, and wireless telecommunications.6-8 Their work has shown that potentially hazardous EMI can be analog or digital. Analog signals can be affected by time-varying electromagnetic fields in the surrounding environment, and digital packets of data can be disturbed by unwanted signals entering the transportation media (e.g., cables) and embedding themselves within the packets, as "physiological" parts.

To prevent this type of interaction, equipment manufacturers have begun to consider issues of electromagnetic immunity during design and production. They have also published several directives, of which the most common and most widely accepted is international standard IEC 60601-1-2, which has been adopted as harmonized European standard EN 60601-1-2.9,10

The standard refers to two causes of EMC-related problems: the existence of source equipment whose emissions must be limited, and the existence of susceptible equipment that must be adequately shielded in order to be protected from the disturbances in its environment. IEC 60601-1-2 addresses issues relating to tests, measurement techniques, and safety levels of emissions and susceptibility for medical equipment. It also proposes installation and mitigation guidelines to achieve controllable emissions and immunity of equipment.

Figure 1. On-line navigation maps linked with relevant information.

It is important to identify critical EMI situations, particularly in healthcare where patient safety is a factor. The research outlined in this article addresses EMI concerns in a hospital environment.11,12 The study centers on an integrated on-line electromagnetic distribution mapping system created and used to research the sources, types, and levels of EMI in the departments of the General Distinct Anti-Cancer Hospital (Piraeus, Metaxa, Greece) and the magnetic resonance imaging (MRI) facility of a private clinic in Patras, Asklipios (see Figure 1). The goals were to recognize and understand potential EMI threats and to facilitate problem solving within the hospital environment. Following is a discussion of the mapping system and its implementation and results.

Figure 2. ECG self-interfering problem (left); measuring equipment (right).

Materials and Methods

Using specialized equipment, the researchers carried out magnetic-field measurements in the General Distinct Anti-Cancer Hospital and in the clinic's MRI facility. A Metron Norway QA5 measuring device yielded the magnetic-field intensity measurements. In addition, an appropriate home-developed device measured the root-mean-square (rms) value of the induced electromotive force (EMF) by the presence of a time-varying electromagnetic field on a coil of one turn, enclosing an area of about 0.1 m2 (see Figure 2).

The researchers documented this information in the form of electromagnetic distribution layouts that were superimposed on plots of the buildings' structures, providing a visual representation of the magnetic-field distribution covering all departments within the hospital. The abbreviation OR that appears in the electromagnetic distribution layouts and in Tables I-IV stands for over range, meaning that the reading was higher than the useful range of the instrument.

The information database identified magnetic-field distribution areas, including the contribution of each specific medical-equipment source of electromagnetic radiation to the formation of the surrounding distribution. Equipment was tested in all modes of operation, from standby to the various frequency ranges of electrosurgical units (ESU). These navigation maps were capable of isolating risk areas; that is, areas that presented an increasing probability of EMI events according to IEC 60601-1-2, and thus indicated susceptible equipment and potentially questionable installations. Additional device-specific information accompanied the maps. These data, in layouts and charts, indicated the magnetic-field profiles around the devices and their frequency from several distances and directions.

Results

The researchers developed magnetic-field distribution layouts for all departments in the hospital, including the intensive care unit (ICU), operating theater, and hospital wards, and for the private clinic's MRI facility. A virtual tour consisting of HTML distribution maps is presented on a controlled-access Web site, http://www.bmtl.bme.teiath.gr/ Electromagnetic%20Compatibility/index.htm. A restricted area of the site includes digitized layouts based on all of the data that were acquired, documented, and superimposed to the buildings' structural plots. These layouts reconstruct the exact mobile and nonmobile medical equipment installation, indicating susceptible areas—areas that present increasing probability of EMI events—via hot-spot links. Accompanying information consists of charts and spreadsheets citing all measurements obtained from each device. The readings documented in Tables I-IV are given in telsa (T) and converted to volt/meter (V/m), assuming a far-field relationship of E (V/m) = 377 W X H (A/m), and 1 T = 795,000 A/m. The measurements were categorized according to their direction and mode of operation. In the following study, only the maximum magnetic-field intensity (Bmax) values were presented at a distance of 10 cm.

In Vitro Diagnostic Laboratories. The in vitro diagnostic laboratory environment is comprised of a variety of different sources, including sensitive computerized analyzers and such electrical motor components as centrifuges and stirring devices (see Table I). Although no disturbances were observed, electrical motors showed significant spreading of magnetic-field profiles capable of interfering with sensitive microprocessor-based components.

ICU and Surgery. The equipment used in these areas (ESUs and defibrillators, for example), exhibited strong magnetic- and electric field intensity characteristics (see Table II). The units' electric-field extended a few centimeters and exceeded 20 V/m—an intensity value capable of causing EMI problems in the near vicinity. An outmoded electrocardiograph's (ECG's) self-interfering power supply and a gamma-counter software malfunction were among the EMI disturbances detected.

Figure 3. Layout of radiotherapy department.

But these were not the only issues. To support a patient hospitalized in the ICU, many of these medical systems had to be in operation mode and installed near the patient. It is not uncommon for multiple instruments such as ventilators, suction pumps, dosimetric pumps, vital-signs monitors, and external pacemakers to be connected to just one patient. In this hospital, the close proximity of all these devices in a limited space around the nursing bed complicated the electromagnetic environment in this area.

Radiotherapy. During this study, no interference was observed in this department. It should be noted, however, that mobile dosimetric systems (e.g., rate meters, calibration chambers, etc.) must be carefully checked for EMI problems because the electronics embedded within them may record noise and alter their output (see Figure 3).

Medical Imaging. A gamma-counter unit demonstrated susceptibility symptoms (see Table III). When strong electromagnetic emitters (e.g., ordinary centrifuges, mixers, etc.) approached the unit, several software errors were recorded. These errors disappeared when the emitters were drawn away. Also, mobile x-ray units were found to interfere with outmoded ECG units, altering output.

Wards. Medical systems in the hospital wards were mostly mobile devices assisting in-bed examination of patients. Among the most common were mobile resorptions, mobile x-ray units, and ECGs, the latter of which were found to be the most sensitive within their respective electromagnetic environment (see Figure 2). The complexity of these specialized areas increased when visitors brought in external mobile sources such as cell phones. Such devices were measured in order to determine whether their usage could influence in-bed hospital examination equipment (see Table IV).

Outpatient Departments. As in other areas of the hospital, some of the old-fashioned ECG units showed susceptibility symptoms. When strong electromagnetic emitters, such as electrosurgical devices, approached the unit, the output was altered by noise contamination. In some cases, the useful signal was totally overlapped and could not be distinguished from the interference. When the ESUs were drawn away, these errors disappeared.

Auxiliary Facilities. Several specialized facilities within the hospital environment were studied. Facilities included areas such as the nutrition department, the medicine storage area, the medical physics laboratory, the well-hole area, the washing machines area, and the administration offices. In contrast to other auxiliary facilities examined, the intercommunication center and the power supply substation were of particular interest. These two areas were heavily equipped with transformers that produced significant levels of electromagnetic radiation. However, although these high-emission areas were adjacent to the in vitro diagnostic laboratories, the equipment in the labs was unaffected.

MRI Facility. The private clinic's MRI installation was also tested. The unit used a 0.5-T magnet, which was continuously in operation mode due to the magnet's superconductive type. Cooling facilities were located in the vicinity, within an adjacent and isolated room containing the liquid nitrogen and helium installations. The superconductive facility was shielded by a Faraday cage constructed with high-specification RF enclosures. Therefore, no significant spread of the magnetic field was recorded.

Discussion

Magnetic-field measurements carried out both in the hospital and in the MRI facility covered all medical equipment from a variety of angles, directions, and distances. An information system created to administer the acquired information was ultimately translated into electromagnetic distribution layouts superimposed to structural plots of the buildings. The electromagnetic distribution layouts were designed using maximum intensity values with the distance from the source. An accessible Web site made the data available to the various hospital departments. The information facilitated recognition, understanding, and handling of potential EMI events, which could cause inconvenience and even life-threatening situations. Digitized layouts indicated high-risk areas in the form of hot-spot link points and were accompanied by additional technical information, including the initial raw measurements for each piece of equipment during several modes of its operation.

Conclusion

Although the researchers detected EMI in the hospital environment, it proved of minor significance under real-world conditions in this particular hospital. The most considerable disturbance was the self-interfering power supply of an old-fashioned ECG device. The on-line EMI distribution mapping system was tested successfully under real-world conditions, assisting the interested parties in identifying the three elements that contributed to the creation of EMI events to enable elimination of at least one of them.

It has been shown, therefore, that the system offered a cost-effective digital alternative for assessing EMI-related threats. Hospitals and other environments where EMI events pose a threat are advised to note such information concerning potential EMI events and take appropriate precautions when selecting and installing equipment.13-15

Acknowledgments

The authors would like to express their appreciation to Kostas Panagiotopoulos, PhD, head of the biomedical engineering department of the Regional General Distinct Anti-Cancer Hospital of Piraeus, Metaxa, Greece, who enabled, encouraged, and supported our measurements in all hospital units.

References

1. William D Kimmel and Daryl D Gerke, Electromagnetic Compatibility in Medical Equipment: A Guide for Designers and Installers (New York: Interpharm Press, 1995).
2. B Spyropoulos, "Introduction to Hospital Organization and Biomedical Technology," in Educational Handbook [on-line] (Patra, Greece, 1997); available from Internet: http://www.bmtl. teiath.gr and http://dsg.harvard.edu/public/courses.
3. RJ Hoff, "EMC Measurements in Hospitals," in Proceedings of the IEEE International Symposium on EMC (San Antonio, TX: IEEE EMC Society, 1975).
4. KS Tan and I Hinberg, "Radio Frequency Susceptibility Tests on Medical Equipment," in Proceedings of the 16th Annual International Conference of IEEE Engineering in Medicine and Biology Society, Baltimore, 1994.
5. UA Frank and RT Londer, "The Hospital Electromagnetic Interference Environment," Association for the Advancement of Medical Instrumentation 5, no. 4 (1971): 246-254.
6. RM Nelson and H Ji, "Electric and Magnetic Fields Created by Electrosurgical Units," IEEE Transactions on Electromagnetic Compatibility 41, no. 1 (1999): 55-63.
7. PS Ruggera and R Elder, "Electromagnetic Radiation Interference with Cardiac Pacemakers," DHEW Publication BRH DEP (1971): 71-75.
8. RE Schlegel et al., ''Electromagnetic Compatibility Study of the In-Vitro Interaction of Wireless Phones with Cardiac Pacemakers," Biomedical Instrumentation and Technology/Association for the Advancement of Medical Instrumentation 32, no. 6 (1998): 645-655.
9. IEC 60601-1-2, 1993, "Medical Electrical Equipment; Part I: General Requirements for Safety; 2. Collateral Standard: Electromagnetic Compatibility; Requirements and Tests."
10. EN 60601-1-2, 1993, "Medical Electrical Equipment; Part I: General Requirements for Safety; 2. Collateral Standard: Electromagnetic Compatibility; Requirements and Tests," (IEC 60601-1-2, 1993).
11. J Hamilton, "Electromagnetic Interference Can Cause Hospital Devices to Malfunction, McGill Group Warns," Canadian Medical Association Journal 154 (1996): 373-375; available from Internet: http://www.cma.ca/cmaj/vol-154/0373e.htm.
12. JL Silberberg, "Performance Degradation of Electronic Medical Devices Due to Electromagnetic Interference," Compliance Engineering 10, no. 5 (1993): 25-39.
13. S Kirk, "Solving Electromagnetic Problems in Medical Equipment," Medical Device Technology 3, no. 1 (1992): 27-30.
14. S Kirk, "Designing Medical Equipment for Electromagnetic Compatibility," Medical Device Technology 3, no. 4 (1992): 42-49.
15. D Davis, B Segal, and T Pavlasek, "Can Minimum Separation Ensure Electromagnetic Compatibility in Hospitals? An Experimental Study," Biomedical Instrumentation and Technology/Association for the Advancement of Medical Instrumentation 33, no. 5 (1999): 411-416.

Basile Spyropoulos, PhD, Dimitris Glotsos, MSc, and Dimitris Batistatos, BSc, are with the department of medical instrumentation at the Technological Institute of Athens (Athens, Greece). Ioannis Marneris, PhD, is at the Brookhaven National Laboratory (Upton, NY). Please direct queries to Spyropoulos. He can be reached via e-mail at basile@teiath.gr.