Biomedical Engineering Reference
In-Depth Information
electrodes, the broadcast-addressing method achieves low input bandwidth
while providing high throughput. These three methods provide a com-
prehensive framework for the automated design of pin-constrained digital
microfluidic biochips.
We have also presented a comprehensive fault-model library that consists
of not only physical defects but also malfunctions. Efficient structural test-
and-diagnosis methods have been proposed based on parallel manipulation
of multiple test droplets in a scan-like manner. The proposed method can
be used in both online and off-line scenarios. We have also introduced the
concept of functional testing for digital microfluidic biochips. A set of test-
ing techniques have been presented to verify the functionality of on-chip
fluidic modules, such as mixers, splitters, and dispensing reservoirs.
This topic has also identiied the need of design-for-testability tech-
niques for digital microfluidic biochips. Testability considerations have been
addressed in the synthesis flow. Appropriate modifications have also been
made in testing methods to increase their effectiveness.
We have applied the proposed design and optimization methods to a
real-life protein crystallization assay. The successful design of a low-cost,
easily manufacturable, high-throughput, and robust chip for protein crys-
tallization has resulted from the optimization algorithms developed in this
topic. This topic has therefore led to powerful design tools for application- and
technology-guided chip design, and served as a bridge between theory and
realistic applications.
7.2 Future Work
Despite the progress that has been reported in this topic, numerous chal-
lenges remain to be tackled for real-chip design automation. Subsection 7.2.1
outlines a physical-constrained-guided synthesis tool based on the frame-
work described in Chapter 2. This section describes how design parameters
and physical constraints, as derived from the fabrication process, can be
incorporated into the synthesis flow. It also proposes a synthesis flow that
is guided by physical constraints. Subsection 7.2.2 describes how errors that
occur during bioassay execution can be handled. To ensure system depend-
ability, many bioassays must be monitored during execution at several
“checkpoints” using sensor-based feedback. When a malfunction is detected,
or the outcome of an assay step deviates from the expected outcome, certain
fluidic operations must be reexecuted. Possible future research directions for
developing a feedback-based synthesis tool that integrates control flow and
conditional “if-then-else” operations are outlined.
