RESUMEN
Industrial statistics plays a major role in the areas of both quality management and innovation. However, existing methodologies must be integrated with the latest tools from the field of Artificial Intelligence. To this end, a background on the joint application of Design of Experiments (DOE) and Machine Learning (ML) methodologies in industrial settings is presented here, along with a case study from the chemical industry. A DOE study is used to collect data, and two ML models are applied to predict responses which performance show an advantage over the traditional modeling approach. Emphasis is placed on causal investigation and quantification of prediction uncertainty, as these are crucial for an assessment of the goodness and robustness of the models developed. Within the scope of the case study, the models learned can be implemented in a semi-automatic system that can assist practitioners who are inexperienced in data analysis in the process of new product development.
RESUMEN
Juvenile psoriatic arthritis (JPsA) is a relatively rare condition in childhood as it represents approximately 5% of the whole Juvenile Idiopathic Arthritis (JIA) population. According to International League of Associations of Rheumatology (ILAR) classification, JPsA is defined by the association of arthritis and psoriasis or, in the absence of typical psoriatic lesions, with at least two of the following: dactylitis, nail pitting, onycholysis or family history of psoriasis in a first-degree relative. However, recent studies have shown that this classification system could conceal more homogeneous subgroups of patients differing by age of onset, clinical characteristics and prognosis. Little is known about genetic factors and pathogenetic mechanisms which distinguish JPsA from other JIA subtypes or from isolated psoriasis without joint involvement, especially in the pediatric population. Specific clinical trials testing the efficacy of biological agents are lacking for JPsA, while in recent years novel therapeutic agents are emerging in adults. In this review, we summarize the clinical features and the current evidence on pathogenesis and therapeutic options for JPsA in order to provide a comprehensive overview on the clinical management of this complex and overlapping entity in childhood.
RESUMEN
Polyploidy is a widespread phenomenon across angiosperms, and one of the main drivers of diversification. Whilst it frequently involves hybridisation, autopolyploidy is also an important feature of plant evolution. Minority cytotypes are frequently overlooked due to their lower frequency in populations, but the development of techniques such as flow cytometry, which enable the rapid screening of cytotype diversity across large numbers of individuals, is now providing a more comprehensive understanding of cytotype diversity within species. Senecio doronicum is a relatively common daisy found throughout European mountain grasslands from subalpine to almost nival elevations. We have carried out a population-level cytotype screening of 500 individuals from Tête Grosse (Alpes-de-Haute-Provence, France), confirming the coexistence of tetraploid (28.2%) and octoploid cytotypes (71.2%), but also uncovering a small number of hexaploid individuals (0.6%). The analysis of repetitive elements from short-read genome-skimming data combined with nuclear (ITS) and whole plastid DNA sequences support an autopolyploid origin of the polyploid S. doronicum individuals and provide molecular evidence regarding the sole contribution of tetraploids in the formation of hexaploid individuals. The evolutionary impact and resilience of the new cytotype have yet to be determined, although the coexistence of different cytotypes may indicate nascent speciation.
RESUMEN
Ecosystems are at increasing risk from the global pollination crisis. Gaining better knowledge about pollinators and their interactions with plants is an urgent need. However, conventional methods of manually recording pollinator activity in the field can be time- and cost-consuming in terms of labour. Field-deployable video recording systems have become more common in ecological studies as they enable the capture of plant-insect interactions in fine detail. Standard video recording can be effective, although there are issues with hardware reliability under field-conditions (e.g. weatherproofing), and reviewing raw video manually is a time-consuming task. Automated video monitoring systems based on motion detection partly overcome these issues by only recording when activity occurs hence reducing the time needed to review footage during post-processing. Another advantage of these systems is that the hardware has relatively low power requirements. A few systems have been tested in the field which permit the collection of large datasets. Compared with other systems, automated monitoring allows vast increases in sampling at broad spatiotemporal scales. Some tools such as post-recording computer vision software and data-import scripts exist, further reducing users' time spent processing and analysing the data. Integrated computer vision and automated species recognition using machine learning models have great potential to further the study of pollinators in the field. Together, it is predicted that future advances in technology-based field monitoring methods will contribute significantly to understanding the causes underpinning pollinator declines and, hence, developing effective solutions for dealing with this global challenge.