Search the repository

A+ | A- | | SLO | ENG

Abstract: Introduction: Sarcopenia is a multifaceted condition affecting between 10 and 16% of the global population, and although multiple classification algorithms exist, no prevalence has been reported for a representative sample of the Slovenian population. Furthermore, multiple behavioural factors, such as malnutrition, physical inactivity, sedentary lifestyle and lower cognitive function, can contribute to the risk of sarcopenia. This study aims to: a) determine sarcopenia prevalence among Slovenian older adults according to different classification algorithms, b) compare the agreement among the algorithms and c) evaluate the relationship between proposed risk factors and sarcopenia. Methods: 654 participants (≥60 years, 30.4% males) have been classified into sarcopenia groups according to eight algorithms, and agreement (Fleiss K) between them was calculated. Additionally, age, sex, nutritional status, physical activity, sedentary levels and cognitive function were assessed as sarcopenia risk/protective factors. Results: The prevalence of sarcopenia according to EWGSOP2 was 4.1%, ranging from 2.1% to 15.3%, when classified by all eight algorithms. Overall agreement between algorithms was weak (K=.429; 95% CI .414 to .444) with 0.6% of participants classified as sarcopenic by all eight algorithms. Adequate nutrition and physical activity were identified as protective factors, while age, lower cognitive function and sedentary lifestyle were considered risk factors. Conclusion: Sarcopenia prevalence among the Slovenian general population was lower than in the global population. We can conclude that different sarcopenia algorithms lead to a different prevalence of sarcopenia. It is of great importance to be cautious when comparing prevalences among studies and to further validate the classification algorithms.
Keywords: sarcopenia, prevalence, epidemiology, classification algorithms, risk factors
Published in DiRROS: 03.03.2025; Views: 225; Downloads: 108
Full text (294,19 KB)
This document has many files! More...

Abstract: The ecological significance of vibroscape has been largely overlooked, excluding an important part of the available information from ecosystem assessment. Insects rely primarily on substrate-borne vibrational signalling in their communication, which is why the majority of terrestrial insects are excluded from passive acoustic monitoring. The ability to monitor the biological component of the natural vibroscape has been limited due to a lack of data and methods to analyse the data. In this paper, we evaluate the use of deep learning models to automatically detect and classify vibrational signals from field recordings obtained with laser vibrometry. We created a dataset of annotated vibroscape recordings of meadow habitats, containing vibrational signals categorized as pulses, harmonic signals, pulse trains, and complex signals. We compared different deep neural network architectures for the detection and classification of vibrational signals, including convolutional and transformer models. The PaSST transformer architecture, which was fine-tuned from a pre-trained checkpoint demonstrated the highest performance on all tasks, achieving an average precision of 0.79 in signal detection. For signals with more than one hour of annotated data, the classification models achieved instance-based F1-scores above 0.8, enabling automatic analysis of activity patterns. In our case study, where 24-hour field recordings were analysed, the trained models (even those with lower precision) revealed interesting activity patterns of different species. The presented study, together with the dataset we publish with this paper, lays the foundation for further analysis of the vibroscape and the development of automated methods for ecotremological monitoring that complement passive acoustic monitoring and provide a comprehensive approach to ecosystem assessment.
Keywords: vibroscape, ecotremology, deep learning, automatic classification, biotremology, insects, zoology, laser vibrometry, ecosystem assessment
Published in DiRROS: 21.01.2025; Views: 291; Downloads: 158
Full text (3,46 MB)
This document has many files! More...

Keywords: honey bees, temperature, phenotypic plasticity, image analysis, abdominal coloration, image classification
Published in DiRROS: 29.10.2024; Views: 591; Downloads: 821
Full text (3,72 MB)

Abstract: Higher-level classifications often must account for monotypic taxa representing depauperate evolutionary lineages and lacking synapomorphies of their better-known, well-defined sister clades. In a ranked (Linnean) or unranked (phylogenetic) classification system, discovering such a depauperate taxon does not necessarily invalidate the rank classification of sister clades. Named higher taxa must be monophyletic to be phylogenetically valid. Ranked taxa above the species level should also maximize information content, diagnosability, and utility (e.g., in biodiversity conservation). In spider classification, families are the highest rank that is systematically catalogued, and incertae sedis is not allowed. Consequently, it is important that family level taxa be well defined and informative. We revisit the classification problem of Orbipurae, an unranked suprafamilial clade containing the spider families Nephilidae, Phonognathidae, and Araneidae sensu stricto. We argue that, to maximize diagnosability, information content, conservation utility, and practical taxonomic considerations, this “splitting” scheme is superior to its recently proposed alternative, which lumps these families together as Araneidae sensu lato. We propose to redefine Araneidae and recognize a monogeneric spider family, Paraplectanoididae fam. nov. to accommodate the depauperate lineage Paraplectanoides. We present new subgenomic data to stabilize Orbipurae topology which also supports our proposed family-level classification. Our example from spiders demonstrates why classifications must be able to accommodate depauperate evolutionary lineages, e.g., Paraplectanoides. Finally, although clade age should not be a criterion to determine rank, other things being equal, comparable ages of similarly ranked taxa do benefit comparative biology.
Keywords: classification, family rank, phylogenomics, systematics, monophyly, spider phylogeny, zoology
Published in DiRROS: 12.07.2024; Views: 681; Downloads: 414
Full text (1,63 MB)
This document has many files! More...

Abstract: Sensor-based human activity recognition is becoming ever more prevalent. The increasing importance of distinguishing human movements, particularly in healthcare, coincides with the advent of increasingly compact sensors. A complex sequence of individual steps currently characterizes the activity recognition pipeline. It involves separate data collection, preparation, and processing steps, resulting in a heterogeneous and fragmented process. To address these challenges, we present a comprehensive framework, HARE, which seamlessly integrates all necessary steps. HARE offers synchronized data collection and labeling, integrated pose estimation for data anonymization, a multimodal classification approach, and a novel method for determining optimal sensor placement to enhance classification results. Additionally, our framework incorporates real-time activity recognition with on-device model adaptation capabilities. To validate the effectiveness of our framework, we conducted extensive evaluations using diverse datasets, including our own collected dataset focusing on nursing activities. Our results show that HARE’s multimodal and on-device trained model outperforms conventional single-modal and offline variants. Furthermore, our vision-based approach for optimal sensor placement yields comparable results to the trained model. Our work advances the field of sensor-based human activity recognition by introducing a comprehensive framework that streamlines data collection and classification while offering a novel method for determining optimal sensor placement.
Keywords: human activity recognition, multimodal classification, privacy preservation, real-time classification, sensor placement
Published in DiRROS: 11.12.2023; Views: 898; Downloads: 417
Full text (6,40 MB)
This document has many files! More...