Our test kits rely on taking samples from rice to provide input. However, it is crucial to note that samples taken from different positions within the rice may contain varying concentrations of toxins. Therefore, it is imperative for us to establish a reliable framework for representing the sampling position within the rice. Bacterial distribution in the food matrix often demonstrates unpredictable and complex mechanisms, taking into account bacteria-food interactions, bacterial colony growth, and environmental factors, etc. Our project tackles cereulide, a toxin produced by a gram-positive rod-shaped bacteria named B. Cereus. Contamination of rice product is reported from time to time. This is more prominent in one of the most consumed staple foods around the globe, rice. After cooking, B. cereus dies due to high temperature. Nonetheless, without proper storage, B. cereus from the atmosphere may aggregate on the surface of rice, elevating the risk of food contamination due to bacterial proliferation and spore germination in the matrix.
In investigating how B.cereus is distributed in rice after cooking and under the condition of improper storage, experiment is first performed to simulate the actual bacterial movement in rice. Initially, stacking of biological images was considered, which entails the process of taking every slice image of a 3D structure, and intensity of each point is obtained using the methods like maximum intensity projection, average intensity projection, or summation slices projection. Yet, due to the lack of expertise in our team in doing such stacking technique and the lack of related equipment in our institution, we came up with an alternative.
      Beyond doubt, test kits cannot escape from sampling frailties. Perfect representation simply does not exist. A negative result shown in a sample does not necessarily evince the absence of hazard in the vast source from which the sample is obtained. However accurate and well-considered the test kit behaves, the sampling method is certainly the key that determines the trustworthiness of the result.