Robin A. Felder, Ph.D. and Sean Graves, Ph.D., The Medical Automation Research Center, University of Virginia, Charlottesville, VA
Dan Robertson and Bill Hulette, Organon Teknika, Durham, NC.
Denis A. Ferkany, CRS Robotics Corporation, Burlington, Ontario, Canada
Correspondence: rfelder@virginia.edu
Laboratory automation is expected to dramatically reshape the cost structure of laboratory medicine in the next decade. Already reengineered laboratories in Japan have demonstrated that labor costs can be reduced over 50% without sacrificing productivity or quality. In North America, many of the 3500 hospital and commercial laboratories are expected to buy automated laboratory systems in the next 5 years (source: Enterprise Analysis Corporation, Stamford, CT). A significant proportion of the automation acquired in North America is likely to be significantly different than the pioneering TLA (total lab automation) systems used in Japan. Systems that feature flexibility and adhere to accepted standards are likely to be the most successful.
Automation will ultimately provide the tools necessary to meet the most rigorous efficiency, quality, and return on investment target. However, both hardware and software products best suited for use in North America are still in the process of evolving. A strategy to consider on the path to implementing automation in your laboratory is a systematic stepwise process involving the coordinated purchase of hardware modules, development of software, and cross training of technologists.
The lowest risk in laboratory automation hardware technology might be the modular analytical workcell. Modular systems can reduce the need for laboratory space and often require only minor facilities modifications. The ideal modular workcell can improve throughput while reducing the requirement for technologist labor to perform specimen analysis and quality control. The modular approach to laboratory automation will reduce the capital costs of laboratory automation to a point where it is affordable by most laboratories.
Laboratory safety and technologist occupational health can also be improved by the use of modular workcells. The robotic arm addresses a key issue in modern labs, the problem of repetitive stress injuries such as sample transport, loading and unloading centrifuges, and loading and unloading tubes from the analyzer.
Automating the specimen processing area is one of the most cost effective automation strategies (1). The pre-analytical area of the laboratory consumes almost 20% of the total laboratory labor budget. Pre-analytical processors are now available from a number of vendors (Table 1). As long as the specimen container is of a standard size (e.g. 12mm X 75mm, or 16mm X 100mm) and shape (e.g. Vacutainer or Sarstedt tube and others) most pre-analytical processors will be able to automatically perform the steps of bar code reading, centrifugation, decapping, aliquotting, and sorting. However, poorly labeled tubes or non-standard containers can negatively impact the overall efficiency of pre-analytical automation in the laboratory. Therefore, features that enhance the versatility of pre-analytical processors and their capability to accommodate a variety of tube sizes and shapes and deal with exceptions to standard containers is important. Automation affordability is highly constrained at the moment due to the large variety of specimen containers that are in use in hospitals. This situation is further complicated by exceptions due to insufficient volume and fibrin clots.
Modular workcells are a relatively new concept in laboratory automation. The Medical Automation Research Center (MARC) at The University of Virginia has created a number of analytical workcells for the diagnostics industry. A recent project that began in 1996 and was centered on the creation of an automated sample manager/processor for the MDA® coagulation analyzer made by Organon Teknika Corporation. This automated system has been trademarked as the MDA® coagAutoLink™, and provides a versatile series of robotic modules that automate coagulation testing.
This project resulted from the cooperative efforts of Organon Teknika Corporation, CRS Robotics Corporation, and the Medical Automation Research Center (MARC). The robotic-module concept allows labs to cost-effectively automate coagulation testing by combining the Multichannel Discrete Analyzer (MDA®) system for coagulation with the robotic modules. The core analytical components are proven technologies, namely the MDA® 180 or MDA® II. Laboratories can enhance their analytical capabilities with a robotic workcell that interfaces with either one or two MDA® instruments. The core automation component of the MDA® coagAutoLink™ workcell is a standard CRS Robotics "human scale arm". The use of a robotic arm was dictated by the need to accommodate the diversity of tasks that technologists are challenged with. In today’s laboratory, the robotic arm can deal with instruments built for human interaction. The modular systems have been configured with enough flexibility to allow for loading and unloading of diagnostic analyzers, centrifuges, and sample transport options. In addition, the workcell system process controller can handle all the timing issues entailed in the multi-task duties it performs.
A further enhancement of the workcell is the use of a mobile robot (MDA coagAutoLink-MR) that provides a cost-effective sample-transport option that connects into the lab grid. The mobile robot delivers samples to the MDA® coagAutoLink™ from anywhere within approximately 1&Mac218;4 mile on one floor. The configurability of the mobile robot route provides a significant increase in flexibility over pneumatic tubes. Furthermore, the mobile robot can be adapted to automatically pick up coagulation tubes and deliver them directly into the modular automation for processing and analysis. The mobile robot can be programmed for speed of delivery, pickup and drop-off locations, delivery routes, and delivery times. The use of voice synthesis even allows the robot to produce "personalized" messages that help to soothe technologists’ fear of this new technology.
For Organon Teknika, it was important to create a series of automation modules that would integrate with the existing MDA® system as economically as possible. The project was initiated to allow the MDA® 180’s or MDA® II’s to be retrofitted into any laboratory in the process of automating without requiring software changes and with minimal hardware upgrades. Therefore the collaborative effort between MARC, CRS Robotics, and Organon Teknika created four different modules.
The MDA coagAutoLink 1 automates a single MDA and the MDA coagAutoLink 2 automates two MDAs.
The MDA coagAutoLink CST is a conveyor sample transport interface.
The MDA coagAutoLink MR is a mobile sample transport robot and Interface.
The MDA coagAutoLink SPIN is an integrated centrifuge interface.
It was clear to Organon Teknika that total lab automation using a variety of conveyor mechanisms was becoming a reality in a number of large-volume labs, and that they had to be able to integrate their MDA® product line into all of these total lab automation systems. Within seven months of initiating the project, a working model was ready to exhibit at clinical trade shows. The rapid prototyping of the concept workcell was made possible by the versatility of the CRS robotic arm. At this time clinical trials are in progress in the core laboratory at the University of Virginia Hospital. The MDA® coagAutoLink™ will be released for general sale from Organon Teknika later in 1999.
The outcome of this project has demonstrated that laboratories have two principal directions in which they can automate: either with total lab automation, ‘TLA’, where the whole package is acquired from beginning to end of the process; or with a modular approach, focused on acquiring task specific workcells. The MDA® coagAutoLink™ provides the versatility to fit almost any laboratory’s coagulation automation needs in the near term, and allows upward compatibility at a future date if test volumes and capital availability require that the workcell be interfaced into a TLA system.
Table 1. Preanalytical Automation
|
2. Graves S, Felder RA. Robotic automation of coagulation analysis. Clin Chem Acta 278(2), 269-279, 1998.
3. Felder RA. Robotics and automated workstations for rapid response testing. Am J Clin Pathol 104 (Suppl 1): S26-S32, 1995.
4. Felder RA. Overview and challenges. in Handbook of Clinical Laboratory Automation and Robotics. Kost GJ (Ed), Wiley and Sons Publishers, NY, pp 3-29, 1996.
5. Felder RA. Automation of Preanalytical Processing and Mobile Robotics. in Handbook of Clinical Laboratory Automation and Robotics. Kost GJ (Ed), Wiley and Sons Publishers, NY, pp 252-282, 1996.