RadioLab project: knowledge of radon gas in Italy

RadioLab is an Italian project, addressed to school-age people, and designed for the dissemination of scientific culture on the theme of environmental radioactivity, with particular regards to the importance of knowledge of radon gas exposure. The project is a nationwide initiative promoted by the National Institute of Nuclear Physics- INFN. First tool used by the project, and of immediate impact to assess the public awareness on radon, is the administration of the survey "do you know the radon gas?". In the survey, together with the knowledge of radon and of its sources, information on personal, cultural and territorial details regarding the interviewees are also taken. Reasonably, the survey invests not only young people, but also their relatives, school workers and, gradually, the public. The survey is administrated during exhibitions or outreach events devoted to schools, but also open to the public. The survey is in dual form: printed and online. The online mode clearly leads RadioLab project even outside the school environment. Based on the results of the survey, several statistical analyses have been performed and many conclusions are drawn about the knowledge of the population on the radon risk. The RadioLab benefit and the requirement to carry on the project goals, spreading awareness of environmental radioactivity from radon, emerge. The dataset involves all twenty Italian regions and consists of 28,612 entries covering the 5-year period 2018-2022.

Shelf-life studies of putative probiotic Lacticaseibacillus casei strains in milk and model yogurt

Lacticaseibacillus casei are commonly utilized as probiotic in a wide-range of fermented and unfermented dairy products. The stability of probiotics in fermented dairy products during shelf-life is of concern due to low pH and high level of organic acids. The objective of this study is to evaluate L. casei for their ability to survive in a model yogurt and fluid milk; additionally, their impact on the pH, organic acids, and sensory attributes of these products was examined. The strain-to-strain differences in cell densities in yogurt and milk inoculated at a therapeutic level at the end of shelf-life were 1.2 and 1.4 log CFU/mL, respectively. Five of the strains examined increased the pH of the yogurt, while two strains were observed to reduce the pH. In milk, one strain raised the pH, while eleven strains reduced the pH. The levels of lactate, acetate, and formate in both the yogurt and milk were altered in a strain-specific manner. The results suggested that the metabolism by these strains differed significantly during the shelf-life. Careful strain selection is required to identify probiotic L. casei strains that will survive through shelf-life in either yogurt or fluid milk and not impact product quality.

Adaptive calibration of imaging array detectors

We present two methods for nonuniformity correlation of imaging array detectors based on neural networks; both exploit image properties to supply lack of calibrations and maximize the entropy of the output. The first method uses a self-organizing net that produces a linear correction of the raw data with coefficients that adapt continuously. The second method employs a kind of contrast equalization curve to match pixel distributions. Our work originates from silicon detectors, but the treatment is general enough to be applicable to many kinds of array detectors like those used in infrared imaging or in high-energy physics.

A self-organizing neural network for the traveling salesman problem that is competitive with simulated annealing

Unsupervised learning applied to an unstructured neural network can give approximate solutions to the traveling salesman problem. For 50 cities in the plane this algorithm performs like the elastic net of Durbin and Willshaw (1987) and it improves when increasing the number of cities to get better than simulated annealing for problems with more than 500 cities. In all the tests this algorithm requires a fraction of the time taken by simulated annealing.

Sorting with self-organizing maps

A self-organizing feature map (Von der Malsburg 1973; Kohonen 1984) sorts n real numbers in O(n) time apparently violating the O(n log n) bound. Detailed analysis shows that the net takes advantage of the uniform distribution of the numbers and, in this case, sorting in O(n) is possible. There are, however, an exponentially small fraction of pathological distributions producing O(n2) sorting time. It is interesting to observe that standard learning produced a smart sorting algorithm.