Abstract
The International Conference on Automation, Robotics and Artificial Intelligence Applied to Analytical Chemistry and Laboratory Medicine has become the premier meeting dedicated to expanding the knowledge base in the rapidly evolving field of laboratory automation. This year was the fifth consecutive year for the International Conference, and we experienced impressive growth in registrations and exhibitions. The conference, held at the Sheraton Hotel and Marina in San Diego, California, featured 51 exhibitors, 14 sponsors, 54 oral presentations, 41 posters, and over 600 registrants and exhibitor associates.
In the last issue of LAN, we covered the first day of scientific sessions which focused on artificial intelligence as well as the final session on novel technology. In this second part of a three-part article, we summarize Part 1 of the parallel scientific session of the conference.
SCIENTIFIC PROGRAM: DAY TWO
On Wednesday, January 18 the scientific program was presented in two parallel sessions. The first dealt with clinical laboratory automation, while the second was concerned with analytical, environmental, and pharmaceutical automation.
PARALLEL SESSION I -CLINICAL LABORATORY AUTOMATION
by Stephen Middleton
The incorporation of laboratory automation will not only change the way in which a laboratory operates, but also how laboratory professionals perceive their work. It is a paradigm shift. Stephen Middleton presented Auto Lab Systems' solutions. He claims that automation decisions must be based on the goals and role of the laboratory. Laboratory automation solutions have deeper roots than simply managing test tubes.
by Margarita Palutke
Hematology is becoming one of the fastest growing automated disciplines within the clinical laboratory. Margarita Palutke showed, for the first time in an international conference, data that demonstrate the advantages of automation in reducing costs in hematology. She demonstrated that the Sysmex hematology system is providing substantial cost savings to her institution.
by Jeff Quint
Beckman Instruments' studies have quantitated the costs of processing a medical specimen. Jeff Quint showed data that demonstrated that the preanalytical phase of a medical sample's life is the most expensive due to the high labor requirement for processing. Laboratories can realize a 40% reduction in labor and reporting time simply by consolidating workstations and providing an electronic interface to their laboratory information system (LIS).
by Carl Teplitz
One of the largest hospital-based automation systems will soon be installed at the Beth Israel hospital in New York. Carl Teplitz explained that their system is designed to accommodate the thousands of specimens that will result from the consolidation of several hospitals in New York, As they continue to consolidate and expand their hospital base, the system should be able to accommodate the increased numbers of specimens without the need for additional hardware. Teplitz indicated that they expect a three-year return on investment. Beth Israel has also installed one of the first voice recognition systems for documenting surgical pathology cases. Although voice recognition software still requires training, the Beth Israel system has already proven more efficient than written surgical pathology reports.
by Rodney Markin
Only one American-made clinical laboratory automation system (LAS) is currently available. Rod Markin, with the help of business associates, has created a modular conveyor belt system for transporting medical specimens throughout the laboratory. His company, Lablnterlink (Omaha, NE) not only sells the conveyor system with unique specimen carriers, but also has developed software for Hospital Information System - LIS - LAS interconnectivity. Lablnterlink will soon offer process control software. Markin has offered payback figures for the Lablnterlink system: cost $2.8 million, payback five years, FTE reduction 11.66 across all shifts.
by Stanley Bauer
Stan Bauer from Beth Israel Hospital (New York, NY) presented issues related to planning the installation of the Coulter/IDS laboratory automation system. The Beth Israel installation will encompass all of the high-throughput workstations in their current laboratory. The first US installation of the Hitachi Clinical Laboratory Automation System (CLAS) is scheduled for a hospital in Indianapolis. The CLAS system is a turnkey, third generation LAS which is functioning in many Japanese hospitals. In the Gunma hospital, they are saving 5 FTEs by processing 800 samples per day with the CLAS. Similarly, the Tohoku hospital processes 1200 samples per day (900 serum/ 300 hematology) with only 5/ 6 FTEs to operate the system. The Indianapolis installation is projected to process 1800 tubes per day using Hitachi 911, 917 and 747 analyzers. The system is expected to operate 24 hours a day.
by Marc Simmons
In Japan, the scope of automatic clinical laboratory systems ranges from dedicated chemistry applications to fully-integrated chemistry, hematology, urinalysis, coagulation and immunoassay. Boehringer Mannheim has the marketing rights to the Hitachi Laboratory Automation System, the only turnkey solution in the United States. Marc Simmons claims that with only five to six FTEs, up to 1800 specimens per hour can be processed. In the future, 1000–1200 primary tubes per hour and 1000–1200 aliquot tubes per hour will be processed.
by Alain Laugier
Alain Laugier showed the results of the first artificial neural network (ANN) module used to classify pathologic blood slides. Leukemia classification was successfully completed using this approach with correct classifications on 1470 of 1500 slides. Results were encouraging and motivated a laboratory evaluation which will complete the ANN training set and confirm the accuracy of the diagnosis.
by Pierangelo Bonini
Pierangelo Bonini gave a detailed account of the operation of a typical Italian hospital clinical laboratory. In Italy, the breakdown of labor commitment by task is: 20% pre-analytical extra-laboratory, 37% pre-analytical infra-laboratory, 25% analytical, 13% post analytical and 5% miscellaneous. There is a substantial need for laboratory automation due to the high cost of labor and a 0.3–0.5% error rate.
by Kevin Bennet
The Mayo Clinic (Rochester, MN), a major hospital as well as a commercial laboratory, uses automation to achieve the following goals: reduction of costs, increase of sample throughput, elimination of biohazards, and reduction of error rates. They do not implement point-of-care automation systems because of the high cost involved. However, the clinic has applied modular automation with some success over the last several years. Kevin Bennet received the 1996 Boehringer Mannheim award for his work on cost analysis at the Mayo clinic.
by Robin Felder
Robin Felder continued the theme of clinical laboratory automation and introduced the concept of automation of point-of-care analysis. He suggested that there are three major ways to automate the clinical laboratory: turnkey, modular, and point-of-care. Currently, one turnkey solution to laboratory automation is the Boehringer Mannheim/Hitachi CLAS system. Modular approaches, which may be less expensive, include the Coulter/IDS system or the Lablnterlink system. While initial costs may be higher for a turnkey system when compared to a modular system, the overall costs of a modular system may be greater because it will take longer to effect labor reductions. Felder described the evolution of laboratories from today's trend toward large, consolidated laboratories to a time in the not too distant future when such laboratories will be replaced with affordable point-of-care. Emerging technologies, such as the Remote Automated Laboratory System (RALS) at the University of Virginia Medical Center, will make point-of-care affordable, provide professional supervision for each result, eliminate the need for training, and yet provide the essential rapid turnaround time.
by William Godolphin
Bill Godolphin described his thoughts on information technology and the need for privacy. When total electronic recording of patient medical histories becomes widely available, privacy protection schemes will have to be developed. Fair information practices should include having only a limited collection of information contained in each storage device with complete notification to the patient of the intended use for the information.
Consent for any secondary uses should be obtained from the patient as well as rights to access and connection. Each use of private information should be documented, so that there is complete accountability and security.
by Ursula Spichiger
Ursula Spichiger described the general principals of biosensors and mentioned their applications to the continuous monitoring of reactions and process control. Biosensors have three basic components, the first of which is a molecular recognition site in contact with the analyte. Material science is actively involved with identifying substances which can provide a positive barrier to the solute but allow selective recognition of the analyte. The second part of a biosensor transduces the presence of the analyte into a quantifiable signal. Many schemes have been developed to provide signal transduction including optical, electrical, magnetic, thermal, mechanical, and chemical techniques. The third part transforms the electrical signal into a measurable result.
by Rainer Schäfer
Rainer Schäfer discussed the use of neural networks to predict a “hook effect” in immunoassays. The hook effect is a condition in an immunoassay when erroneous data is produced because the antibody is not present in sufficient excess to allow proper competitive binding to occur. Through the use of neural networks, hook effects can be better predicted. It is anticipated that neural network algorithms may become part of all clinical immunoassay analyzers in the future.