Printer Friendly

Automation: trends in instrumentation, robotics, computers.

In another article in this issue, we explore some of the technologies that are sure to have an impact on the clinical laboratory during the 1990s. Here we will discuss the major forces that should contribute to a more fully automated clinical laboratory environment in the future.

Today's clinical laboratory is being pressed on two fronts. On the one hand, a shortage of medical technologists has made it increasingly difficult to get the work done. On the other hand, growing financial pressures have demanded that labs become more productive. Laboratory automation offers at least a partial solution to this dilemma. In

the past, increased automation was achieved by purchasing an analyzer capable of doing a greater variety of tests at a higher throughput. During the next decade, expect labs to become increasingly automated in nontraditional ways. Here are some of them: * Robotics. Some sections of the clinical laboratory, such as hematology and chemistry, have shown dramatic productivity increases due to greater automa-

tion. But other sections, such as specimen processing, have been relatively untouched by push-button advances and remain manual, labor-intensive areas. This should change during the next decade now that the clinical laboratory is finally being exposed to a selection of affordable and flexible robotic systems.

Robots are mechanical devices that can be programmed to do a variety of tasks with the dexterity

of a human. They have always been an integral part of fully automated analyzers where they are used for such tasks as pipetting specimens and reagents. These types of robots have restricted motion and limited programma- bility; they are usually dedicated to a single function. In contrast, stand-alone robots are fully programmable and have a wide range of three-dimensional motions (some are more configurable than others).

These devices confer several real benefits on the laboratory: They cut costs by allowing unattended operation, eliminate risks associated with handling biohazardous materials, perform repetitive or complex tasks in the same reproducible way each time, reduce the errors that tend to plague relatively boring tasks, and are fast and flexible.

Robots are becoming more commonplace in the lab, often performing sophisticated specimen pipetting functions. A number of manufacturers sell robotic stations capable of preparing, washing, and reading microplates.

Most of the interest has thus far focused on serology and blood grouping and typing. Many independent laboratories use robots to remove specimens from tubes and make multiple specimen and reagent additions for testing as diverse as RIA and serology. Logical extensions of such systems will eliminate practically all human handling on EIA procedures to minimize staff contact with serum. These robots operate in a Cartesian mode with three degrees of freedom-x, y, and z.

In the next decade, robotic sys-

tems will play a larger role in automating specimen handling and processing. For instance, a device may be programmed to aliquot specimens, centrifuge them, or sort them by workstation. This would reduce many of the errors associated with specimen labeling and handling.

Robotic arm systems with four degrees of freedom, which enables them to handle complex repetitive tasks, are popular for automating the front end of certain instrument systems, preparing solutions and performing repetitive aliquoting steps. These devices have gained slow but steady acceptance and are seen in many industries, including the clinical laboratory.

Laboratories are also using robotic arms for specialized tasks, such as moving microtiter plates from location to location on command and automatically presenting specimens to single-sample devices. Look for more automation of the front end of instruments to end the need to move specimens from one location to another.

The most sophisticated systems available have five degrees of freedom. These have yet to see extensive use in the laboratory but when they begin to, the changes will occur rapidly. The devices are not as limited in their field of operation or in their ability to learn complex nonrepetitive tasks.

Recently, a Japanese firm displayed a fully automated sample handling system. Although not perfect, it is a harbinger of what is to come. The system accepts an aliquot list downloaded from the host computer; it then automati-

cally prepares serum aliquots from spun primary collection tubes. Aliquots are prepared for a host of laboratory locations, all without any manual handling of the specimen tubes. This approach will grow in importance as concern about biohazardous materials makes such tasks increasingly undesirable.

The ultimate step in robotics will be full automation of specimen flow from start to finish. Bar-coded specimens will move along conveyer belts and be shunted to the proper location as they pass transfer points. Once the specimens reach their intended loca- tion, they will be lined up for aliquoting into the appropriate tube for an instrument. The instrument will then be loaded automatically by a robot and the completed tray of specimens removed to make room for the next.

Sound farfetched? Systems of that kind have been investigated by a number of reference laboratories and are now in the prototype stage at a number of major teaching hospitals. Beyond these locations, one only has to look at modem warehouses where systems of conveyor belts and robotic arms fill containers and orders with virtually no human intervention.

The traditional jobs of specimen processing and sorting are sure to be reduced. Automated tracking systems will also be used. Specimens will be stored in bar-coded locations so that no one will ever have to spend half a day looking for a missing tube. This process will be an adaptation of the tracking systems currently used by such outfits as Federal Express.

These developments will mark

the movement of today's technology into tomorrow's workplace. It is not always an easy process-witness the auto industry's problems in adapting standard technology to automate plants.

An even more exciting future application of robots will be in locations outside the central laboratory. Robots deployed in decen-

tralized testing sites may be responsible for introducing specimen into an instrument. This would allow unmanned satellite labs to operate around the clock; it would also save on labor costs and, in this era of technologist shortages, create more interesting tasks and enhance productivity. Robots will also be used to log specimens and to store and retrieve them automatically.

Look for robots to take over many of the mundane tasks that no one in the laboratory likes to do. This change will be expensive at first since most of the equipment will be custom-developed, but in time standard systems will become available. Installation

will be driven by the need of laboratories to lower costs and their inability to recruit and retain skilled staff. The biosafety aspects will lend additional urgency to acquiring robots.

Robots generally do not perform jobs faster than the laboratory staff. They can, however, perform them in a more reliable, con-

sistent, and accurate way. They also do not suffer from fatigue or boredom when doing repetitive tasks. They will enable the technologist of the future to spend more time monitoring sophisticated instruments and far less time mixing supplies and pouring specimens. * Computers. Perhaps more than any other technology, computers have been responsible for advances in clinical laboratory automation. As a measure of how far and fast this technology has developed, consider that many of today's hand-held calculators have greater processing power than large floor-model systems of only a decade ago-and at a fraction of

the cost. The heart of the computer, the microprocessor, is composed of miniaturized transistors on a silicon chip.

Besides their use in laboratory information systems, computers have had a major impact on automated instruments in th as years. Most advances n instrumentation would not have been

possible without corresponding computer advances.

Microprocessors perform multiple functions on an instrument. They are responsible for coordinating reagent, optical, and liquid handling systems as well as a variety of data management functions, such as reducing raw data optical readings, for example), matching cup positions with spec- imens, and collating demographic information.

They are also responsible for many features that make instruments easier to run in the lab today. These include automatic rerun and dilution, on-line trouble- shooting, reagent inventory management, bidirectional interface

with a host computer, extended calibration, and real-time quality control.

In the 1990s, computers will continue to simplify the user interface and operation of clinical laboratory analyzers. Knowledge-based expert systems will provide computer-assisted support for troubleshooting, maintenance, and quality control activities. New patient data could be interpreted in light of historical data to determine what follow-up action to take. An instrument could also be linked via modem to the manufacturer's central computer in. order to compare data from other laboratories.

If the last decade was one in which computers exerted a significant impact on the design and operation of laboratory instruments, the next decade will be one in which they revolutionize the management, presentation, and analysis of lab data and its integration within a comprehensive health care data bank. New technologies, specifically superconductors and optical disks, will provide enormous advances in computer processing speed and data storage capacity, respectively.

Computers have until now been used to perform two basic functions: system control and data management. System control functions have expanded dramatically with the advent of microprocessors. Data management has focused on moving numbers and results in the most efficient way to make results more readily available to the user. We see the future being different.

Microprocessors will not only control the system functions but also make decisions about system

operation and implement those decisions. Automated troubleshooting, data checking, automatic repeat testing, and rejection of outlier results will be standard features. Much effort will be placed in making the user interface smoother to permit easier operation. This simplified operation will become more important as the national pool of highly qualified medical technologists continues to shrink.

A more exciting trend for computers in the laboratory is in the area of expert systems. These systems have just begun to appear, and they will permit instruments to make decisions that would require multiple interpretive steps by a human. As expert systems become more accurate and commonplace, many of the common interpretive procedures in the laboratory will be replaced-for example, hemoglobin and protein electrophoresis, thin layer chromatography, and other pattern interpretation tests.

An additional instrument feature based on artificial intelligence will be the ability to appropriately order follow-up tests based on the result of the first test. Rather than wait for the clinician to review the result and essentially come to the same conclusion, the instrument in consultation with its own expert system or the laboratory's expert system will order the next test. Conceivably, a series of tests would be presented to the clinician in the same time it now takes to receive the initial result. In the era of DRGs and short lengths of hospital stay, these developments will be a major contribution.

The laboratory will play a ma- jor role not only in the diagnosis of the patient via end results but also in actually guiding the entire diagnostic process. As algorithms for diagnosis become more and more complex, physicians will no longer have to rely on their memories to help them select the logical next steps. The laboratory will do that.

One can envision hand-held microcomputers given to phlebotomists before their early morning draw. These will be programmed to print bar codes as the patient's wrist band is scanned. Such systems are in fact available and used at some major teaching hospitals. The future difference: The laboratory's data system will have in part made the test selection and will order the appropriate next test -based on previous results. The entire feedback loop will be dramatically shortened. Look for the pathologist to lead the charge toward more rational testing.

Expert systems incorporated into instruments will also generate computer-aided instruction for the first-time user. Rather than presenting a series of canned lessons or steps, the system will find out what the new trainee has learned and frame the lessons appropriately. This will decrease training costs and also make training more consistent across laboratory sections.

Instruments and laboratory information systems will incorporate many of today's E-Mail (electronic mailbox) systems. Physicians will be linked to the laboratory electronically and rely less and less on printed reports. Their beepers will be called automatically with critical values, and telephone test requests will be han-

dled by computer operators. These changes could occur in the next five years as the cost of personnel soars and the velocity of information accelerates. The explosive growth of fax machines is an indication of how quickly a technology can be introduced into a receptive market.

Computers will no longer produce mere numeric reports. Instead, graphic presentation of data will be the norm. Predictive values of results will appear alongside the numeric value, guiding the interpretation. Where necessary, interpretive and diagnostic data will advise the clinician of his choices and possible conclusions.

Under certain circumstances, not only text but images will be transmitted within a local area computer network. For instance, electrophoresis tracings or cytology or surgical pathology slide images could be sent to the nursing station as part of the patient's electronic record.

An offshoot of traditional image analysis methods is telepathology-the combination of highresolution video monitors, robotically controlled microscopes, microcomputers, and satellite communications that allows a pathologist to view a specimen located elsewhere. The system works by sending a slide image from the camera to a space satellite; on the other end, a satellite receiving station captures the analog signal, decodes and digitizes it, and displays it on a high-resolution monitor. A telephone line is used for voice communication and remote microscope control. When the nation's current phone network is upgraded with fiber optic lines,

telepathology transmission could become easier and less costly.

Telepathology is a major advance that will become more widespread in the 1990s. It offers a simple and obvious benefit-long-distance consultation. With this technology, the practicing pathologist in a remote area can confer with specialists to quickly resolve diagnostic problems. It is unlikely that this technology will replace pathologists at distant sites, but it will serve as an adjunct tool.

Instruments will no longer be placed in categories such as chemistry, hematology, or immunology. Rather, they will be multidis- ciplinary test processors (MTPS) that can take a single specimen and perform a variety of tests linked only by some common handling and reading steps. As instruments become more sophisticated, the end user will be able to take advantage of their features by

performing more complex tests.

Reagents will increasingly come in prepackaged form. The intelligence in most cases will be in the reagent test packs themselves, and the instruments will be common processors-much like the sophisticated software that manipulates personal computers. Although costlier than traditional methods, this approach will permit e sophisticated testing to be performed with fewer and less skilled individuals. It will also permit distributed testing in different areas of the hospital.

In addition to MTPS, expect more image-analysis-based applications, due to the advances in computer technology. Current instruments have been used to automate the microscopic urinalysis and histologic or cytologic slides for, as an example, DNA ploidy analysis.

Make way for the new wave of automation.
COPYRIGHT 1989 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:medical laboratories
Author:Lifshitz, Mark S.; De Cresce, Robert P.
Publication:Medical Laboratory Observer
Date:Jul 1, 1989
Previous Article:New lab consolidations: competing on a cost basis.
Next Article:How illness demographics will affect the lab.

Related Articles
A new crisis in medical technology.
Automated immunoassay systems: a new frontier.
A laboratory automation platform: the next robotic step.
Preparing for the lab of the 21st century.
Leveling the playing field: the economics of robotics in the hospital clinical labs.
Automation of diagnostic testing.
Modular stepwise automation and the future of diagnostic testing.
Reading the future: the increased relevance of laboratory medicine in the next century.
Efficient laboratory design.
Robotics in the laboratory: where we have been and where we are going.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters