Introduction - Hardware/Software Co-Design and Optimization for Cyberphysical Integration in Digital Microfluidic Biochips - page 18

1.2.3 Error Recovery

For biomedical applications including clinical diagnostics, it is necessary to ensure
the accuracy of the on-chip fluidic operations when the biochip is being used [ 48 ].
The accuracy of the fluid-handling operations can be monitored by examining
various parameters, such as the volumes and concentrations of product droplets.
If an error occurs during the execution of the bioassay, such as the volume of an
intermediate product droplet exceeding the normal value, the assay outcomes can
be incorrect, and the entire experiment must be repeated. Therefore, it is important
to detect such errors as early as possible and repeat the corresponding fluid-handling
operations to obtain correct outcomes from the bioassay.
In [ 48 ], the authors proposed a conceptual mechanism to monitor the intermedi-
ate products of the bioassay and implement error-recovery operations to minimize
the influence of erroneous operations. A sequence of checkpoints is inserted into the
initial protocol of the bioassay. At these checkpoints, sensors are used to check
the qualities of the intermediate droplets. If the droplets fail to meet the quality
requirements, the detector sends an interrupt to the control software, and some fluid-
handling operations are repeated. In this way, error recovery can be implemented
automatically.

1.2.4 Pin-Assignment Methods

The electrical I/O interface for digital microfluidics poses challenges. If each
electrode were controlled by an independent pin and each pin were to have an input
pad fabricated on the chip, the area required for the biochip would be extremely
large, and the cost of fabrication would be high. Thus, several design optimization
techniques have been proposed and analyzed in the literature [ 43 , 49 - 51 ] with the
goals of reducing the number of control pins and controlling the digital microfluidic
array without significantly affecting concurrent droplet operations.
These methods reduce the number of pins in two different ways. One way is to
reduce the number of pins by designing the biochip with some special structures.
For example, in the n-bus-phase scheme, every nth electrode is connected to the
same control pin. Another example is the cross-referencing biochip proposed in
[ 52 ]. Figure 1.14 shows the cross-sectional views of the cross-referencing biochip.
The rows and columns of electrodes in the biochip are fabricated on the top and
bottom plates, respectively. A cell is activated when a row of electrodes are charged
opposite polarities with a column of electrodes. When moving a droplet in the
x-direction, the electrode columns on the top plate serve as the “ground electrode”,
and the electrode rows on the bottom plate serve as “driving electrodes” (Fig. 1.14 a).
Similarly, when moving a droplet in the y-direction, the electrode columns on the
bottom plate serve as the “ground electrode”, and the electrode rows on the top plate
serve as “driving electrodes” (Fig. 1.14 b).