Search

Prolonged incorrect sitting postures can lead to various health issues, including musculoskeletal disorders and reduced quality of life. Efficient and accurate methods for monitoring and assessing sitting postures are crucial for promoting overall well-being in our increasingly sedentary society. This study compares the performance of Kolmogorov-Arnold Networks (KANs) and Multilayer Perceptrons (MLPs) in sitting posture recognition, utilizing a multi-task classification approach to simultaneously identify upper and lower body positions. A comprehensive dataset of sitting postures was developed using MediaPipe for skeletal extraction, labeled for both body regions, and encompassing a range of common scenarios. The models were implemented and evaluated using Leave-One-Subject-Out (LOSO) cross-validation to assess generalization capability across different individuals. Results indicated that the KAN model achieved higher accuracy (97.03% for upper body and 92.11% for lower body) with fewer parameters (4,320) compared to the MLP (93.87% and 91.47% respectively, with 7,662 parameters). Although the MLP demonstrated faster inference, both models achieved accuracies exceeding 91% in LOSO validation, suggesting their potential for real-time applications. This study significantly contributes to the field of automatic posture recognition, offering insights into the applicability of KANs and MediaPipe-based pose estimation in ergonomic recognition systems and demonstrating the effectiveness of a multi-task approach for comprehensive sitting posture analysis.

Summary: Background: SARS-CoV-2 infection is considered as a relapsing inflammatory process with a dysregulation of IL-6 signalling. Classic IL-6 signalling is thought to represent a defence mechanism against pathogens. In contrast, IL-6 trans-signalling has pro-inflammatory effects. In severe COVID-19, therapeutic strategies have focused on global inhibition of IL-6, with controversial results. We hypothesized that specific blockade of IL-6 trans-signalling could inhibit inflammatory response preserving the host defence activity inherent to IL-6 classic signalling. Methods: To test the role of the specific IL-6 trans-signalling inhibition by sgp130Fc in short- and long-term consequences of COVID-19, we used the established K18-hACE2 transgenic mouse model. Histological as well as immunohistochemical analysis, and pro-inflammatory marker profiling were performed. To investigate IL-6 trans-signalling in human cells we used primary lung microvascular endothelial cells and fibroblasts in the presence/absence of sgp130Fc. Findings: We report that targeting IL-6 trans-signalling by sgp130Fc attenuated SARS-CoV-2-related clinical symptoms and mortality. In surviving mice, the treatment caused a significant decrease in lung damage. In vitro, IL-6 trans-signalling induced strong and persisting JAK1/STAT3 activation in endothelial cells and lung fibroblasts with proinflammatory effects, which were attenuated by sgp130Fc. Our data also suggest that in those cells with scant amounts of IL-6R, the induction of gp130 and IL-6 by IL-6:sIL-6R complex sustains IL-6 trans-signalling. Interpretation: IL-6 trans-signalling fosters progression of COVID-19, and suggests that specific blockade of this signalling mode could offer a promising alternative to mitigate both short- and long-term consequences without affecting the beneficial effects of IL-6 classic signalling. These results have implications for the development of new therapies of lung injury and endotheliopathy in COVID-19. Funding: The project was supported by ISCIII, Spain (COV-20/00792 to MB, PI23/01351 to MARH) and the European Commission—Next generation EU (European Union) (Regulation EU 2020/2094), through CSIC's Global Health Platform (PTI Salud Global, SGL2103029 to MB). PID2019-110587RB-I00 (MB) supported by MICIN/AEI/10.13039/501100011033/and PID2022-143034OB-I00 (MB) by MICIN/AEI/10.13039/501100011033/FEDER. MAR-H acknowledges support from ISCIII, Spain and the European Commission—Next generation EU (European Union), through CSIC's Global Health PTI.

This study introduces a lightweight storage system for wearable devices, aiming to optimize energy efficiency in long-term and continuous monitoring applications. Utilizing Direct Memory Access and the Serial Peripheral Interface protocol, the system ensures efficient data transfer, significantly reduces energy consumption, and enhances the device autonomy. Data organization into Time Block Data (TBD) units, rather than files, significantly diminishes control overhead, facilitating the streamlined management of periodic data recordings in wearable devices. A comparative analysis revealed marked improvements in energy efficiency and write speed over existing file systems, validating the proposed system as an effective solution for boosting wearable device performance in health monitoring and various long-term data acquisition scenarios.

In the rapidly evolving domain of conversational agents, the integration of Large Language Models (LLMs) into Chatbot Development Platforms (CDPs) is a significant innovation. This study compares the efficacy of employing generic and fine-tuned GPT-3.5-turbo models for designing dialog flows, focusing on the intent and entity recognition crucial for dynamic conversational interactions. Two distinct approaches are introduced: a generic GPT-based system (G-GPT) leveraging the pre-trained model with complex prompts for intent and entity detection, and a fine-tuned GPT-based system (FT-GPT) employing customized models for enhanced specificity and efficiency. The evaluation encompassed the systems’ ability to accurately classify intents and recognize named entities, contrasting their adaptability, operational efficiency, and customization capabilities. The results revealed that, while the G-GPT system offers ease of deployment and versatility across various contexts, the FT-GPT system demonstrates superior precision, efficiency, and customization, although it requires initial training and dataset preparation. This research highlights the versatility of LLMs in enriching conversational features for talking assistants, from social robots to interactive chatbots. By tailoring these advanced models, the fluidity and responsiveness of conversational agents can be enhanced, making them more adaptable and effective in a variety of settings, from customer service to interactive learning environments.

IL-6 is one of the major mediators of the hyper-inflammatory responses with complex biological functions as it can signal via different modes of action. IL-6 by classical signalling has anti-inflammatory and antibacterial activities, while trans-signalling mediates pro-inflammatory effects. The net biological effect of IL-6 is established by multiple factors beyond its absolute concentration. Here, we assess the relationship between IL-6 signalling variables [IL-6, soluble IL-6R (sIL-6R) and soluble gp130 (sgp130)] and outcomes in a cohort of 366 COVID-19 patients. The potential trans-signalling was evaluated by a ratio between the pro-inflammatory binary IL-6:sIL-6R complex and the inactive ternary IL-6:sIL-6R:sgp130 complex (binary/ternary complex) and the fold molar excess of sgp130 over sIL-6R (FME). Our data provide new evidence that high levels of IL-6, sIL-6R, sgp130, binary/ternary complex ratio, and low FME are independent predictors of COVID-19 severity in survivor patients (without death), and the combination of IL-6 + sIL-6R + sgp130 exhibited the most robust classification capacity. Conversely, in a subgroup of patients with a very poor prognosis, we found that high levels of IL-6 and low levels of sIL-6R, sgp130, and binary/ternary complex ratio were predictors of death. In this context, the highest predictive capacity corresponded to the combined analysis of IL-6 + FME + lymphopenia + creatinine. Herein, we present IL-6 signalling variables as a helpful tool for the early identification and stratification of patients with clear implications for treatment and clinical decision-making.

Cognitive deficits are very difficult to diagnose during the initial stages; tests typically consist of a patient performing punctual dual-task activities, which are subjectively analyzed to determine the cognitive decline impact on gait. This work supports novel and objective diagnosis methods by stating a baseline on how neurotypical aging affects dual tasks while using a smartphone on the move. With this aim, we propose a twofold research question: Which mobile device tasks performed on the move (dual tasking) have characteristic changes in gait parameters, and which are especially characteristic at older ages? An experiment was conducted with 30 healthy participants where they performed 15 activities (1 single task, 2 traditional dual-tasks and 12 mobile-based dual-tasks) while walking about 50 m. Participants wore a wireless motion tracker (15 sensors) that made the concise analysis of gait possible. The results obtained characterized the gait parameters affected by mobile-based dual-tasking and the impact of normal cognitive decline due to aging. The statistical analysis shows that using smartphone-based dual-tasking produces more significant results than traditional dual-tasking. In the study, 3 out of 10 gait parameters were very significantly affected (p < 0.001) when using the traditional dual tasks, while 5 out of 10 parameters were very significantly affected (p < 0.001) in mobile-based dual-tasking. Moreover, the most characteristic tasks and gait parameters were identified through the obtained results. Future work will focus on applying this knowledge to improve the early diagnosis of MCI.