| Prototype active CMOS biochip under development at Columbia Univ. DNA probe is immobilized directly on the chip surface on one of the 256 pixel sites. Active pixel sensors detect the hybridization of fluorescently labeled target. Such a chip could one day be used in a handheld diagnostic device. |
Handheld diagnostic devices--most of which are still largely at the research stage--are poised to benefit from new chip technologies.
Due to advances in medicine and the biological sciences, where genetic and other biochemical information is crucial, sensitive diagnostic devices will be needed to detect specific molecular agents and extract their information. Miniaturized instruments using lab-on-a-chip technologies are likely to play a key role in this area, with handheld technology being ideal in field applications. Researchers are now investigating the best kind of chip technology that will enable handheld-sized diagnostic devices."Handheld technologies are currently limited since very few array technologies, except for electrochemical array detection, lend themselves well to miniaturization," says Jennifer Dent, director of business development at CombiMatrix Corp., Washington, D.C. "Most array systems--spotted, or generated by other chemicals methods--require large or delicate imaging systems to gather data, where the image of the chip is analyzed, rather than the actual event."
These "photonic" approaches can work in a research scenario, however they are not as sensitive as electrochemical detection, nor suitable for handheld devices. New technologies from CombiMatrix allow on-chip detection of assay results by using electronic circuits on the chip. Samples are labeled with reagents that generate chemical signals, which are then detected by on-chip electrochemical methods. These fluorescent-free assays are similar to fluorescent-based ones; however, assay signal intensities are not captured by optical detection instruments, but rather by a small computer interface.
But is CMOS the boss?
"Examples of handheld biosensor technologies offered commercially exist but are not widespread," says Ken Shepard, associate professor of electrical engineering at Columbia Univ., N.Y. "There are handheld biosensors that use CMOS chips as part of the overall system design, such as in data storage or read-out; the most prominent example is glucose monitoring in diabetic patients." However, these do not take advantage of the potentially massive parallelism available with CMOS-based direct detection in which thousands of analytes are monitored concurrently. Devices such as DNA and protein microarrays are prime candidates for near-term development of analogous CMOS-based device platforms in which biological molecules are directly coupled to CMOS-sensing structures.Shepard, along with team member Rasti Levicky, is developing a fully CMOS-integrated microarray device. This array contains active pixel sensor fluorescent detection and electrical field control of hybridization through oxide-passivated surfaces. The chip adds electronics with photodiode detection for pixel-level analog-to-digital conversion of fluorescent light intensity. Each pixel site has a 搈eshed?metal pad that lets light pass through while establishing an electric field to control hybridization between probe and target species.
"Self-contained operation simplifies use, as well as reduces cost by eliminating expensive analytical hardware," says Shepard. "These advantages are expected to be critical, for instance, for routine use of biosensor technology in most medical offices, where large capital investments in depreciable hardware are not well justified by the limited requirements on sample throughput."
Future in their hands?
Research suggests that within ten years, commercial CMOS-based biochips may appear, with the initial commercial application likely to focus on genotyping, which has tremendous clinical potential."The availability of low-cost, even disposable, diagnostic CMOS-based devices could make personalized medicine a routine reality," says Shepard. 揅ontinued investment by the life sciences community into connecting genetic information with clinically relevant behavior and response is essential to drive these applications."
