Multipurpose monitoring system for edible insect breeding based on machine learning.

The Tenebrio molitor has become the first insect added to the catalogue of novel foods by the European Food Safety Authority due to its rich nutritional value and the low carbon footprint produced during its breeding. The large scale of Tenebrio molitor breeding makes automation of the process, which is supported by a monitoring system, essential. Present research involves the development of a 3-module system for monitoring Tenebrio molitor breeding. The instance segmentation module (ISM) detected individual growth stages (larvae, pupae, beetles) of Tenebrio molitor, and also identified anomalies: dead larvae and pests. The semantic segmentation module (SSM) extracted feed, chitin, and frass from the obtained image. The larvae phenotyping module (LPM) calculated features for both individual larvae (length, curvature, mass, division into segments, and their classification) and the whole population (length distribution). The modules were developed using machine learning models (Mask R-CNN, U-Net, LDA), and were validated on different samples of real data. Synthetic image generation using a collection of labelled objects was used, which significantly reduced the development time of the models and reduced the problems of dense scenes and the imbalance of the considered classes. The obtained results (average [Formula: see text] for ISM and average [Formula: see text] for SSM) confirm the great potential of the proposed system.

Species Composition and Trichothecene Genotype Profiling of Fusarium Field Isolates Recovered from Wheat in Poland.

Fusarium head blight (FHB) of cereals is the major head disease negatively affecting grain production worldwide. In 2016 and 2017, serious outbreaks of FHB occurred in wheat crops in Poland. In this study, we characterized the diversity of Fusaria responsible for these epidemics using TaqMan assays. From a panel of 463 field isolates collected from wheat, four Fusarium species were identified. The predominant species were F. graminearum s.s. (81%) and, to a lesser extent, F. avenaceum (15%). The emergence of the 15ADON genotype was found ranging from 83% to 87% of the total trichothecene genotypes isolated in 2016 and 2017, respectively. Our results indicate two dramatic shifts within fungal field populations in Poland. The first shift is associated with the displacement of F. culmorum by F. graminearum s.s. The second shift resulted from a loss of nivalenol genotypes. We suggest that an emerging prevalence of F. graminearum s.s. may be linked to boosted maize production, which has increased substantially over the last decade in Poland. To detect variation within Tri core clusters, we compared sequence data from randomly selected field isolates with a panel of strains from geographically diverse origins. We found that the newly emerged 15ADON genotypes do not exhibit a specific pattern of polymorphism enabling their clear differentiation from the other European strains.

The use of computer-assisted image analysis in the evaluation of the effect of management systems on changes in the color, chemical composition and texture of m. longissimus dorsi in pigs.

The effect of management systems on selected physical properties and chemical composition of m. longissimus dorsi was studied in pigs. Muscle texture parameters were determined by computer-assisted image analysis, and the color of muscle samples was evaluated using a spectrophotometer. Highly significant correlations were observed between chemical composition and selected texture variables in the analyzed images. Chemical composition was not correlated with color or spectral distribution. Subject to the applied classification methods and groups of variables included in the classification model, the experimental groups were identified correctly in 35-95%. No significant differences in the chemical composition of m. longissimus dorsi were observed between experimental groups. Significant differences were noted in color lightness (L*) and redness (a*).