Systems incorporating machine vision are speeding the production of medical products.
In the United States, the health-care market represents a vast, enticing opportunity for providers of medical products and services. Total health-care expenditures reached $1.7 trillion in 2003, and the health sector is one of the largest industry sectors of the national economy, employing over 10 million people and equaling approximately 15% of GDP. According to Espicom Business Intelligence (Tangmere, England), the U.S. medical-device market alone was $71 billion in 2003, and is a highly attractive market for systems developers because it accounts for over half the total world market. In addition, the strength of U.S. medical-device manufacturing companies and the stringent regulatory requirements of the U.S. Food and Drug Administration (FDA) mean that foreign manufacturers have difficulty penetrating the market.
Machine vision has proved to be a critical tool for U.S. medical-device manufacturing companies in their quest to ensure quality, lower costs, and automate device production. Catheters, for example, are often used to deliver stents or open and repair constricted blood vessels. To automate the production of catheter manufacturing, companies such as Boston Scientific (Natick, MA), Conceptus (San Diego, CA), and Medtronic (Minneapolis, MN) rely on system integrators such as Engineering By Design (EBD; San Jose, CA) to develop specialized manufacturing equipment.
According to EBD president Dale Henson, precise control of process variables is important because of the size of the components used in manufacturing these intravascular devices. In the manufacture of miniature wire-wound catheters, programmable wind tensions of less than 1 gram and wind-angle accuracies of better than 1/10° require the use of high-performance control systems and machine-vision systems that measure the wind angle without touching the filament winding.
Other automated systems are combining robots and vision systems for parts feeding and automated assembly operations of products such as pregnancy test kits. David Stroup, chairman and CTO of Prodosyst Automation (San Diego, CA), a systems integrator, says that today’s automated assembly systems for diagnostic-device manufacturing must accommodate frequent product changes and new product introductions.
“Historically,” says Stroup, “disposable pharmaceutical products such as pregnancy test kits were manufactured using product-specific discrete pick-and-place machines. However, with the introduction of shorter production runs and specialized diagnostic testing products, companies are demanding systems capable of manufacturing a number of different products.” Prodosyst has already developed such an automated system for Applied Biotech (San Diego, CA) based on four Cobra s600 SCARA robots from Adept Technology (Livermore, CA), cameras from Panasonic Vision Systems (Secaucus, NJ), and Adept’s vision software. The system has eliminated errors found in the manual assembly of pregnancy test kits, increased product quality, and eliminated manual labor costs.
From abroad
The competitive U.S. market for high-quality, low-cost medical devices has also driven European manufacturers to machine-vision-based automated assembly. Ypsomed (Burgdorf, Switzerland), for example, manufactures disposable pharmaceutical needles and needed to ensure the sharpness and accuracy of needle placement in a plastic housing known as a cannula. To meet quality demands from the USA’s FDA while maintaining a throughput of approximately 480 parts/minute, Ypsomed turned to sortimat Technology (Winnenden, Germany) to build an inspection system, which in turn used imaging software from NeuroCheck (Remseck, Germany) to check each needle while maintaining production throughput. Ypsomed is now expected to purchase three of the production-and-inspection systems from sortimat at a cost of between €2 million and €3 million; machine vision ensures the cannulas comply with the FDA’s specifications.
Manufacturers of medical devices are increasingly turning to automated manufacturing systems and machine vision. This cycle of market pressure will ensure the ongoing development of novel machine-vision systems and offer those using medical devices a more productive life.
CONARD HOLTON is editor in chief of Vision Systems Design; e-mail: [email protected].