This Article Full Text (PDF) Submit a scholarly reply Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Doyle, D. J. Search for Related Content PubMed PubMed Citation Articles by Doyle, D. J.

Embedded Computers in the Operating Room and in the Intensive Care Unit

Toronto, Ontario

Computers are now an important and integral part of modern anesthesia care. In many cases "embedded" computers are integrated "transparently and seamlessly" into medical instruments so that the user never realizes that he or she is using a computer per se.

The incorporation of embedded computer technology into monitoring devices along with recent developments in clinical data management is now starting to provide clinicians with the potential for improved decision-making capability. Computerization has also simplified the use of systems for the target-controlled and closed-loop delivery of intravenous and inhaled drugs. As these developments continue we can expect that the use of computer technology will steadily increase over the next several years.

Indeed, embedded computers are used in virtually all medical instrumentation used in the operating room and intensive care unit, similar to chips in TVs, VCRs, microwave ovens, etc. Such instruments include blood pressure monitors, pulse oximeters, capnographs, anesthetic agent analyzers, and even patient controlled analgesia machines and electrocautery devices. In particular, embedded computer technology has had a profound effect on the evolution of the modern anesthesia machine. Flow control, alarm management, data logging are all examples of activities carried out by the embedded computers present in the latest generations of anesthesia machine. Similarly, the embedded computers present in the latest generations of ventilators carry out data display, flow control, alarm management, and data logging. Finally, many state-of-the-art anesthesia machines will produce an automatic anesthetic record.

Very often, the computers inside these instruments are based on one or more inexpensive single chip systems programmed in "assembly language" (a low-level, tedious programming environment), although some recent designs often involve miniaturized and simplified regular IBM personal computer-type designs, and allow for software development using regular computer languages such as Visual Basic or C++. Such embedded computers often consist of a single chip that can cost as little as a dollar or less. Many instruments contain several embedded computers, each focusing on a particular task.

In general such chips are 8,16 or 32 bit devices with non volatile memory (ROM) containing a control program, as well as read/write memory, analog-to-digital conversion capability and so on. What differentiates medical embedded computer systems from other kinds of embedded computers is the need for high reliability, the need to ensure that the software meets extremely high standards, and the need for detailed documentation. The American Food and Drug Administration (http://www.fda.gov) is active in defining requirements for such medical embedded computers. This partly explains the relatively high cost of medical equipment compared with common household consumer items such as TVs and VCRs.

Despite the advantages offered by these developments, there is much to be done to employ embedded computers more wisely. All too often the need for a clear and intuitive user interface is forgotten in the race to bring the latest generation product to market. For instance, the epidural and intravenous patient-controlled analgesia (PCA) systems used in my hospital have serious ergonomic flaws that make them inappropriately difficult to use. (Example: neither system provides visual feedback to the patient (such as a red/green light) to indicate whether or not the patient is "locked out".) The FDA is also aware of these ergonomic issues and is working to offer guidance regarding the best way to use embedded computers in medical equipment that users are comfortable with.

Similarly, there are a number of inherent problems with current intra-operative monitoring systems based on embedded computer technology. First, these systems are necessarily software based, so that all the problems associated with software (testing for bugs, maintenance, upgrades, documentation, etc.) apply. Secondly, as already alluded to, the complexity of such advanced systems usually requires that careful application of human factors (ergonomic) principles be used to ensure that the system is easy to configure and use. For instance, one must ensure that the alarm systems associated with the system are not so "painful" to use that users disable all alarms to get around a high rate of false alarms. Third, software systems are subject to crashes, sometimes at critical times. With respect to this, special attention should be directed to designing the system to recover quickly from system crashes, should they occur. Of course, special attention should also be directed to preventing system crashes in the first place.

Other "must have" features of systems using embedded computers is that they should allow system components to be added or removed without disturbing operation. As an example, users often turn off airway gas monitoring while patients are on cardiopulmonary bypass - it is thus important that when the airway gas monitor is turned off or back on that this does not crash the system. Another issue is that modern patient monitors are often networked, so that data on one monitor can be viewed remotely, for instance in the anesthesia lounge or in another OR. Some monitors contain sufficiently rich data logging capability that they are easily upgraded to full automatic charting systems. Some monitors will run external software such as drug dosing calculation software or medical diagnostic software to identify acid-base disorders from arterial blood gas data. Other monitors allow seamless transfer of data from bedside monitor to transport monitor, and back to the bedside simply by moving a data module.

The next few years will likely witness a continuing boom in the use of computer technology in the OR, but do not expect that the systems will always be easy to use.

This Article Full Text (PDF) Submit a scholarly reply Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Doyle, D. J. Search for Related Content PubMed PubMed Citation Articles by Doyle, D. J.