Everyone enjoys the challenge of deciphering hidden messages, whether they be on a box of Cracker Jack or in a book like The da Vinci Code. However, biomedical engineers at Duke University have taken the decoder ring to a place that it has never been before: the patterns that are created by bacterial colonies.
Bacteria have a tendency to grow in particular patterns, which can be affected by factors such as the availability of nutrients and the amount of available space. The researchers began by simulating the formation of a bacterial colony in a virtual Petri dish. Next, they adjusted the environmental factors affecting the colony’s growth as well as the number and size of the simulated bacterial dots. Finally, they devised an alphabet based on the appearance of the colony once it had completely filled the Petri dish. This particular method of encoding is referred to as emorfi.
The encoding is not one-to-one because the final simulated pattern that corresponds to each letter does not always look exactly the same. This means that the pattern can vary slightly from time to time. On the other hand, the researchers found that a machine learning program could be taught to tell the difference between the two so that it could correctly figure out which letter was meant.
According to Lingchong You, a professor of biomedical engineering at Duke, “A friend may see many images of me over the course of time, but none of them will be exactly the same.” [Citation needed] “A friend may see many images of me over the course of time, but none of them will be exactly the same.” The friend will be able to recognize me even if they are shown a picture of me that they have never seen before if the images all consistently reinforce what I generally look like.
The encoder will ultimately produce a video consisting of a series of patterns, each of which corresponds to a different letter, in order to encrypt real messages. Although to the untrained eye, they might appear to be the same thing, the computer algorithm is able to tell the difference between them. An intruder should not be able to decipher the code without the assistance of a powerful artificial intelligence of their own, provided that the receiver is aware of the initial conditions that led to the inception of the system.
