After testing with our device we found that there is a relatively large offset when compared to the low standard deviation, suggesting non-ideal accuracy. This is accentuated when looking at small vessels such as the one shown in trial 6 (error of 4.84%). However, we estimate that the offset is not an issue with our measurement capabilities. This is because the known measurement was made before the cylinders were color adjusted to be viable with our imaging system. By painting each cylinder we artificially introduced some offset, and so we believe our accuracy is likely much higher. Furthermore, the standard deviation of our measurements is 0.01 mm which is equivalent to that of a digital caliper. This indicates that the actual precision of our measurement system could be greater than shown, as the error may have been introduced by the known measurement determined by the digital caliper rather than through our measurement system.

For these reasons we do not believe that this testing has fully described the measurement capabilities of our device. However, we are hopeful that precision will increase or remain constant, and that accuracy will be largely increased given further testing. Furthermore, the offset can also be adjusted on a small scale with alterations of the image processing script which can provide fine tuning of the accuracy.

In order to provide certainty as to the capabilities of our device we want to perform testing using the laser tool provided by the DVJ lab which can reveal the limits of our diameter measurement with its high accuracy and precision.

In order to improve the research capabilities of our device we want to implement mechanical automation along with a pressure sensor. Currently we have no access to a viable pressure sensor due to cost and size constraints, and so instead we estimate pressure using recently taken pressure-diameter data. With a pressure sensor we would not have to rely on this relationship. Furthermore, it would enable automation through access to constant time data.