For every test run, the Delvotest® Accelerator Smart (DAS) is able to automatically figures out the right 'Control Time' per plate or even 'plate-strips' (i.e., every 96-well Delvotest® plate can be broken into 6 'strips' of 16 wells each), accurately determining whether each sample is either 'Negative' or 'Positive'. So, how's it able to all this so effortlessly? It's all in great part thanks to how its algorithm works, its so called 'Working Principle'. In this article, we explore that logic in detail, highlighting some of the key implications to keep in mind when testing.

## Z-values, Statistical Significance, and Z = -10

Every minute during an incubation, the DAS takes a picture of your Delvotest® plate or plate-strips. It then converts this image into a 'digital map', translating the colour of every well into a colour value. This 'colour value' is based on a proprietary three-dimensional colour scale, the so-called 'Z-scale' (i.e., three axes, 'X', 'Y' and 'Z'), from which 'Z-values' are derived.

However, the first problem the DAS has to deal with is that measuring 'absolute colour values' is not a reliable, easily-reproduceable approach for determining the difference between a 'Positive' and 'Negative' result in microbial inhibition broad-spectrum tests like Delvotest®. Instead, it does the next best thing: it tries to figure out colour changes over time between wells in *relation* to one another.

In practice, that means the DAS begins this relative measurement journey by first presuming every well will yield a 'Negative' result — i.e., that within a certain window of time, colour values should change from 'purple' to 'yellow'. However, as some of these results may be 'Positive', it closely follows a statistically significant number of all wells being measured (i.e., usually 10% of the total, rounding up to the next integer) which are producing this 'purple to yellow change curve' it expects to see. At some point, as the average 'Z-values' of this 'representative cohort of wells' will pass a certain 'Z-value threshold' (usually Z = -10, [/// add name of this parameter//]). The moment this threshold is reached by this 'representative cohort' is called the 'Control Time,' which the DAS uses as a reference point to determine the results for each remaining well based on their individual 'Z-values' at that point in time. Wells which yield a value below [/// add name of this parameter//] Z = -4 are considered 'Negative', and all other above are considered 'Positive'. Incidentally, effectively by working with these two key parameters (i.e., [/// add name of this parameter//] and [/// add name of this parameter//]) and others (e.g., the total time it takes to run an incubation), it is possible for our experts to help you create customised methods for local validation frameworks (🔗) capable of testing dairy matrices other than co-mingled cow milk (e.g., milk with preservatives, other species and dairy matrices, etc.).

## Four Implications

**The first implication // no empty wells. It is important to make sure no wells are left empty. Filling them with negative control milk can improve the algorithm’s accuracy in determining the correct incubation time. For a detailed explanation of the working principle, refer to the graph below:**

The second implication is all about **verifiability**. Even though users can 'tag' whether a well has either a 'Negative' or 'Positive' reference control added to it, the DAS will not take this fact at face value. Instead, it will also measure these wells, determining if they yielded 'True Negative' or 'True Positive' results. That means the DAS is a great system for continuously monitoring and verifying your reference control samples are performing as expected — and for this reason, we advise you to add at least one 'Negative' and one 'Positive' reference control sample to every plate or plate-strip(s) you test. They are not there to be the 'base' from which results are determined (i.e., you can in principle do without them); rather, they're simply there to help you make sure things are running as they're expected. So, if there is ever a problem with your reference controls, this level of continuous verifiability helps you more easily arrive at a Root Cause Analysis (RCA) about their performance.

The third implication is about **statistical significance**. The higher the number of known or expected 'Positive' samples are tested on a plate or plate-strip(s), the greater the need is to ensure a minimum number of known 'Negative' reference controls samples are present to safeguard the operational principle of statistical significance. For most cases, this '10% of the total' applies — with two exceptions. Imagine this: you want to incubate a whole Delvotest® plate (i.e., 96 wells) with as many known/expected 'Positive' samples as possible. 10% of 96, rounded up to the next integer equals '10' which is a reasonable representative cohort in relation to 96. That means at least 10 of the wells being incubated must be 'Negative' reference controls samples in order for the DAS to be able to successfully establish the correct 'Control Time' for all the other 86 wells. However, what if you were trying to achieve the same goal (i.e., incubate as many known/expected 'Positive' samples as possible) but instead with a single plate-strip (i.e. 16 wells)? 10% of 16, rounded up to the next integer equals '2', which is not reasonably a reliable representative cohort in relation to 16. A more reliable but still reasonably practical size for a representative cohort would be 4 wells — which is precisely the number the wells being incubated which must have a 'Negative' reference controls samples in order for the DAS to be able to successfully establish the correct 'Control Time' for all the other 12 wells. For such cases, we therefore recommend the following minimum number of 'Negative' reference controls per number of plate-strips:

No. of Wells | No. of Plate 'Strips' | Min. No. of Negative Controls Required | Min. No. of Positive Controls Advised |

96 | 6 i.e., full plate | 10 | 1 |

80 | 5 | 8 | 1 |

64 | 4 | 7 | 1 |

48 | 3 | 5 | 1 |

32 | 2 | 4 | 1 |

16 | 1 | 4 | 1 |

**Fourth implication about Z-values not being absolute, so though it can be an indication for 'how negative' or 'how positive', it's not that simple.**