Search Results (1,689)

Parameter tuning based adapter methods have achieved notable success in vision-language models (VLMs). However, they face challenges in scenarios with insufficient training samples or limited resources. While leveraging image modality caching and retrieval techniques can reduce resource requirements, these approaches often overlook the significance of textual modality and cross-modal cues in VLMs. To address this, we propose a Conditional Cross-Modal learning model, which is abbreviated as CoCM. CoCM builds separate cache models for both the text and image modalities and embedding textual knowledge conditioned on image information. It dynamically adjusts the cross-modal fusion affinity ratio and disentangles similarity measures across different modalities. Additionally, CoCM incorporates intra-batch image similarity loss as a regularization term to identify hard samples and enhance fine-grained classification performance. CoCM surpasses existing methods in terms of accuracy, generalization ability, and efficiency, achieving a 0.28% accuracy improvement over XMAdapter across 11 datasets and demonstrating 44.79% generalization performance on four cross-domain datasets. Full article

Background: Promoting physical activity among people living with dementia is critical to maximise physical, cognitive and social benefits; yet the lack of knowledge, skills and confidence among health professionals, informal care partners and people with dementia deters participation. As the initial phase of a larger feasibility study, co-design was employed to develop a new model of community care, ‘Changing the Focus’, to facilitate the physical activity participation of older people living with mild dementia. Methods: Co-design methodology was utilised with nine stakeholders (with experience in referring to or providing physical activity programs and/or contributing to policy and program planning) over three workshops plus individual interviews with four care partners of people with dementia. Insights were gathered on the physical activity for people with mild dementia, referral pathways were explored and ‘personas’ were developed and refined. Materials and resources to support exercise providers and referrers to work effectively with people with mild dementia were finalised. Results: Three ‘personas’ emerged from the co-design sessions, aligned with stages of behaviour change: (1) hesitant to engage; (2) preparing to engage; and (3) actively engaged. Referral pathway discussions identified challenges related to limited resources, limited knowledge, access constraints and individual factors. Opportunities were classified as using champions, streamlining processes, recognising triggers for disengagement, influencing beliefs and attitudes, and means of communication. Conclusion: This study captured the views of physical activity referrers and providers and informal care partners in an inclusive and iterative manner. The use of co-design ensured a robust approach to facilitating participation in formal and informal physical activity options for people living with mild dementia. This study has provided the necessary framework from which to develop and test training and resources for the next stage of intervention (a feasibility trial) to improve physical activity participation for people with dementia. Full article

Detecting binary changes in co-registered bitemporal hyperspectral images (HSIs) using deep learning methods is challenging due to the high dimensionality of spectral data and significant variations between images. To address this challenge, previous approaches often used dimensionality reduction methods separately from the change detection network, leading to less accurate results. In this study, we propose an end-to-end fully connected adversarial network (EFC-AdvNet) for binary change detection, which efficiently reduces the dimensionality of bitemporal HSIs and simultaneously detects changes between them. This is accomplished by extracting critical spectral features at the pixel level through a self-spectral reconstruction (SSR) module working in conjunction with an adversarial change detection (Adv-CD) module to effectively delineate changes between bitemporal HSIs. The SSR module employs a fully connected autoencoder for hyperspectral dimensionality reduction and spectral feature extraction. By integrating the encoder segment of the SSR module with the change detection network of the Adv-CD module, we create a generator that directly produces highly accurate change maps. This joint learning approach enhances both feature extraction and change detection capabilities. The proposed network is trained using a comprehensive loss function derived from the concurrent learning of the SSR and Adv-CD modules, establishing EFC-AdvNet as a robust end-to-end network for hyperspectral binary change detection. Experimental evaluations of EFC-AdvNet on three public hyperspectral datasets demonstrate that joint learning between the SSR and Adv-CD modules improves the overall accuracy (OA) by 5.44%, 10.43%, and 7.52% for the Farmland, Hermiston, and River datasets, respectively, compared with the separate learning approach. Full article

In arid Turkana County, over 90% of the population is food insecure, and wild edible plants (WEPs) provide 12–30% of dietary intake. However, climate change and overexploitation threaten these crucial resources. This study employed sequential qualitative methods to investigate community perceptions, conservation priorities for WEPs, barriers, and necessary actions in Turkana. It combined participatory community workshops and expert validation interviews. The research revealed critical threats to WEP availability, including climate change, shifting cultural practices, and a lack of natural regeneration. Key conservation barriers included intergenerational knowledge gaps, inadequate policy implementation, and conflicts between immediate needs and long-term conservation goals. In developing conservation plans, the stakeholders identified and prioritized WEP species based on food value, medicinal properties, cultural significance, utility, and drought resistance. The co-developed conservation strategy emphasized both in situ protection measures, such as community awareness programs and local policy enforcement mechanisms, and restoration actions that include planting prioritized WEPs in home gardens and community spaces. Collaborative roles for communities, non-governmental organizations, researchers, and government actors were identified to provide training, resources, and technical support. This strategy also emphasizes the need for incentivization through food/cash-for-work programs and small business grants to promote alternative livelihoods. The strategies align with some of the most-utilized conservation frameworks and principles, and present new ideas such as integrating indigenous knowledge. Expert validation confirmed the feasibility of proposed actions, highlighting the importance of multi-stakeholder approaches. This study contributes to expanding our knowledge base on community-based conservation and provides insights for policymakers, emphasizing WEPs’ critical role in food security, cultural preservation, and ecological resilience. The findings could serve as a model for similar initiatives in other arid regions facing comparable challenges. Full article

The authentication of Ziziphi Spinosae Semen (ZSS), Ziziphi Mauritianae Semen (ZMS), and Hovenia Acerba Semen (HAS) has become challenging. The chromatic and textural properties of ZSS, ZMS, and HAS are analyzed in this study. Color features were extracted via RGB, CIELAB, and HSI spaces, whereas texture information was analyzed via the gray-level co-occurrence matrix (GLCM) and Law’s texture feature analysis. The results revealed significant differences in color and texture among the samples. The fire–ice ion dimensionality reduction algorithm effectively fuses these features, enhancing their differentiation ability. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) confirmed the algorithm’s effectiveness, with variable importance in projection analysis (VIP analysis) (VIP > 1, p < 0.05) highlighting significant differences, particularly for the fire value, which is a key factor. To further validate the reliability of the algorithm, Back Propagation Neural Network (BP), Support Vector Machine (SVM), Deep Belief Network (DBN), and Random Forest (RF) were used for reverse validation, and the accuracy of the training set and test set reached 98.83–100% and 95.89–99.32%, respectively. The method provides a simple, low-cost, and high-precision tool for the fast and nondestructive detection of food authenticity. Full article

Climate change caused by greenhouse gas (GHG) emissions is an escalating global issue, with the transportation sector being a significant contributor, accounting for approximately a quarter of all energy-related GHG emissions. In the transportation sector, vehicle emissions testing is a key part of ensuring compliance with environmental regulations. The Vehicle Certification Agency (VCA) of the UK plays a pivotal role in certifying vehicles for compliance with emissions and safety standards. One of the primary methods employed by the VCA to measure vehicle emissions for light-duty vehicles is the Worldwide Harmonized Light Vehicles Test Procedure (WLTP). The WLTP is a global standard for testing vehicle emissions and fuel consumption, and sensors are crucial in ensuring accurate, real-time data collection in laboratories. Using the data collected by the VCA, regression machine learning models were trained to predict CO₂ emissions in light-duty vehicles. Among six regression models tested, the Decision Tree Regression model achieved the highest accuracy, with a Mean Absolute Error (MAE) of 2.20 and a Mean Absolute Percentage Error (MAPE) of 1.69%. It was then deployed as a web application that provides users with accurate CO₂ emission estimates for vehicles, enabling informed decisions to reduce GHG emissions. This research demonstrates the efficacy of machine learning and AI-driven approaches in fostering sustainability within the transportation sector. Full article

Metabolic flexibility describes the capability to switch between oxidative fuels depending on their availability during diet or exercise. In a previous study, we demonstrated that in response to training, marathon (DUhTP) mice, paternally selected for high treadmill performance, are metabolically more flexible than unselected control (DUC) mice. Since exercise-associated metabolic flexibility can be assessed by indirect calorimetry or partially by circulating lactate concentrations, we investigated these parameters in DUhTP and DUC mice. Therefore, males of both lines completed a three-week high-speed treadmill training or were physically inactive (sedentary) before being placed in a metabolic cage for three days (one day of acclimatization, two days with monitoring), measuring CO₂ and O₂ to calculate respiratory quotient (RQ) and fatty acid oxidation (FATox). Circulating blood lactate concentrations were determined. Training resulted in a lower RQ in DUhTP and an increased RQ in DUC mice compared to their sedentary counterparts. Increased FATox rates and lower lactate concentrations were observed in exercised DUhTP but not in DUC mice, indicating a shift to oxidative metabolism in DUhTP and a glycolytic one in DUC mice. Therefore, improved metabolic flexibility in DUhTP mice is verifiable up to three days after training. Full article

Background: Caffeine and sodium bicarbonate individually enhance muscular endurance by delaying fatigue, but their combined effects have scarcely been studied. Objectives: This study aimed to evaluate the acute effects of co-ingesting caffeine and sodium bicarbonate on muscular endurance at different loads in bench press and back squat exercises. Methods: Twenty-seven recreationally trained participants (female/male: 14/14; age: 23 ± 3.6 years) were randomized to four conditions in a double-blind, crossover design: (a) sodium bicarbonate and caffeine (NaHCO₃ + CAF); (b) sodium bicarbonate (NaHCO₃); (c) caffeine (CAF); (d) placebo (PLA); ingesting 0.3 g/kg NaHCO₃, 3 mg/kg caffeine or placebo (maltodextrin). Participants performed two muscle endurance tests on bench press and back squat exercises at 65% and 85% 1RM, performing as many repetitions as possible in one set until task failure. Results: CAF increased the number of repetitions (p < 0.001; η_p² = 0.111), mean velocity (V_mean, p = 0.043, η_p² = 0.16), and mean power output (W_mean, p = 0.034, η_p² = 0.15) compared to placebo. These effects were observed in back squat exercise at 65%1RM in V_mean (3.7%, p = 0.050, g = 1.144) and W_mean (5.2%, p = 0.047, g = 0.986) and at 85%1RM in V_mean (5.4%, p = 0.043, g = 0.22) and W_mean (5.5%, p = 0.050, g = 0.25). No ergogenic effects were found in NaHCO₃ + CAF) or NaHCO₃ conditions. Conclusions: CAF increased muscular endurance performance in male and female participants by increasing the number of repetitions, mean velocity, and power output; however, when NaHCO₃ was ingested, these effects were not detected. Full article

In the last decade, the size and complexity of datasets have expanded significantly, necessitating more sophisticated predictive methods. Despite this growth, limited research has been conducted on the effects of autocorrelation within widely used regression methods. This study addresses this gap by investigating how autocorrelation impacts the predictive accuracy and efficiency of six regression approaches: Artificial Neural Network (ANN), Ordinary Least Squares (OLS), Cochrane–Orcutt (CO), Prais–Winsten (PW), Maximum Likelihood Estimation (MLE), and Restricted Maximum Likelihood Estimation (RMLE). The study evaluates each method’s performance on three datasets characterized by autocorrelation, comparing their predictive accuracy and variability. The analysis is structured into three phases: the first phase examines predictive accuracy across methods using Mean Squared Error (MSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE); the second phase evaluates the efficiency of parameter estimation based on standard errors across methods; and the final phase visually assesses the closeness of predicted values to actual values through scatter plots. The results indicate that the ANN consistently provides the most accurate predictions, particularly in large sample sizes with extensive training data. For GDP data, the ANN achieved an MSE of 1.05 × 10⁹, an MAE of 23,344.64, and an MAPE of 81.66%, demonstrating up to a 90% reduction in the MSE compared to OLS. These findings underscore the advantages of the ANN for predictive tasks involving autocorrelated data, highlighting its robustness and suitability for complex, large-scale datasets. This study provides practical guidance for selecting optimal prediction techniques in the presence of autocorrelation, recommending the ANN as the preferred method due to its superior performance. Full article

Background: Although cervical cancer (CC) is highly preventable through appropriate screening methods like the Papanicolaou (Pap) test, which enables early detection of malignant and precancerous lesions, access to such screening has not been equitable across social groups. Sex workers and people with records of incarceration are among the most under-screened populations in Ontario. Little is known about the acceptability and feasibility of HPV self-sampling (HPV-SS) as an alternative cervical cancer screening method for these groups. This online, community-based mixed-methods pilot study aimed to address this knowledge gap. Methods: Eighty-four under- and never-screened sex workers and ex-prisoners aged 25–69 years and residing in the Greater Toronto Area, were recruited by community peer associates. Participants completed an online survey and viewed short videos about CC and screening with Pap and HPV-SS. Those who opted for HPV-SS conducted the test at one of two collaborating organizations. Results: The median age of participants was 36.5 years. Most had limited knowledge about CC and screening. Approximately 13% identified as non-binary, and 5% as two-spirit or trans men, with the majority having completed secondary education. Of the participants, 88% chose HPV-SS, and one-third tested positive for high-risk HPV types. The ability to self-sample without judgment from healthcare providers was noted as a key advantage. However, there was a need for training on proper HPV-SS techniques. Conclusions: To improve cervical cancer screening among sex workers, increasing awareness through participatory community co-creation of sexual health education is essential. Additionally, offering HPV-SS as a screening option is crucial, given its demonstrated acceptability and feasibility within this population, many of whom lack a primary care provider and face discriminatory attitudes in healthcare settings. Full article

Digital transformation offers transformative potential for public service delivery, yet many local administrative organizations (LAOs) in Thailand struggle with integrating digital tools effectively into their workflows. This study investigates the integration of social listening tools (SLTs) to enhance the efficiency and responsiveness of public service delivery in Thailand’s LAOs. The primary goal is to redesign traditional, manual workflows through the development of a digital-by-design framework, addressing inefficiencies in public engagement and service provision. Employing a mixed-method approach, this research combines interviews and focus groups with municipal staff from four municipalities in Northeast Thailand to identify challenges and co-create solutions. The redesigned workflow integrates digital practices into existing organizational structures and achieves a significant 282% improvement in efficiency, measured in transactions per manpower-hour. Additionally, the new process enhances operational speed, responsiveness, and public engagement. To ensure sustainability, this study recommends a phased implementation strategy and consistent staff training. This research contributes to the public administration literature by providing a practical, scalable framework for digital transformation in local governance. It underscores the potential of SLTs to modernize public sector workflows, enabling more dynamic, responsive, and citizen-centric interactions between LAOs and the communities they serve. Full article

Background: The Black Country (BC) is an area of the United Kingdom covering Dudley, Sandwell, Walsall, and Wolverhampton. The area is ethnically, culturally and religiously diverse. One-fifth of the total population is in the lowest socioeconomic quintile, with an uneven distribution of wealth. The area manifests unmet needs and as perceived underserved community groups. Objectives and Methods: To better understand the situation and inform future provision, listening events were organised across the BC to engage with local underserved communities. A mixed-methods design was employed, using collaborative workshops. The workshops enabled stakeholders to explore priorities, perceived barriers and solutions to mental health services’ access within the BC. Results: Sixty participants verbally consented and signed in to attend the three workshops. There were nine groups that provided 247 statements on the topic, yielding a total of 12 codes and six themes (priorities). The top identified priorities were inappropriate periodisation of accessible funded healthcare needs (n = 42, 18.03%), barriers to appropriate healthcare (n = 49, 21.03%) and limited resources for training, health promotion, preventative care and support networks (n = 62, 26.61%). Conclusions: Addressing the identified priorities will require location and community-specific solutions to establish those communities’ trust and engagement. Cultural stigma should not be viewed as the only barrier to access healthcare but should be considered in combination with the population’s reluctance to reach out to healthcare services due to loss of trust between community groups and lack of co-design of culturally and religiously appropriate services for the community. Full article

Boundary contour determination during seedling image segmentation is critical for accurate object detection and morphological characterization in agricultural machine vision systems. The traditional manual annotation for segmentation is labor-intensive, time-consuming, and prone to errors, especially in controlled environments with complex backgrounds. These errors can affect the accuracy of detecting phenotypic traits, like shape, size, and width. To address these issues, this study introduced a method that integrated image features and a support vector machine (SVM) to improve boundary contour determination during segmentation, enabling real-time detection and monitoring. Seedling images (pepper, tomato, cucumber, and watermelon) were captured under various lighting conditions to enhance object–background differentiation. Histogram equalization and noise reduction filters (median and Gaussian) were applied to minimize the illumination effects. The peak signal-to-noise ratio (PSNR) and the structural similarity index measure (SSIM) were used to select the clip limit for histogram equalization. The images were analyzed across 18 different color spaces to extract the color features, and six texture features were derived using the gray-level co-occurrence matrix (GLCM) method. To reduce feature overlap, sequential feature selection (SFS) was applied, and the SVM was used for object segmentation. The SVM model achieved 73% segmentation accuracy without SFS and 98% with SFS. Segmentation accuracy for the different seedlings ranged from 81% to 98%, with a low boundary misclassification rate between 0.011 and 0.019. The correlation between the actual and segmented contour areas was strong, with an R² up to 0.9887. The segmented boundary contour files were converted into annotation files to train a YOLOv8 model, which achieved a precision ranging from 96% to 98.5% and a recall ranging from 96% to 98%. This approach enhanced the segmentation accuracy, reduced manual annotation, and improved the agricultural monitoring systems for plant health management. The future direction involves integrating this system with advanced methods to address overlapping image segmentation challenges, further enhancing the real-time seedling monitoring and optimizing crop management and productivity. Full article

During the ignition process of a solid rocket motor, the pressure changes dramatically and the ignition process is very complex as it includes multiple reactions. Successful completion of the ignition process is essential for the proper operation of solid rocket motors. However, the measurement of pressure becomes extremely challenging due to several issues such as the enormity and high cost of conducting tests on solid rocket motors. Therefore, it needs to be investigated using numerical calculations and other methods. Currently, the fundamental theories concerning the ignition process have not been fully developed. In addition, numerical simulations require significant simplifications. To address these issues, this study proposes a solid rocket motor pressure prediction method based on bidirectional long short-term memory (CBiLSTM) combined with adaptive Gaussian noise (AGN). The method utilizes experimental pressure data and simulated pressure data as inputs for co-training to predict pressure data under new operating conditions. By comparison, the AGN-CBiLSTM method has a higher prediction accuracy with a percentage error of 3.27% between the predicted and actual data. This method provides an effective way to evaluate the performance of solid rocket motors and has a wide range of applications in the aerospace field. Full article

Background: In karate, the ability to execute high-velocity movements, particularly kicks and punches, is heavily dependent on the strength and power of the lower limb muscles, especially the knee extensors. As such, this study aimed to evaluate the effects of an 8-week eccentric training program utilizing the reverse Nordic exercise (RNE) integrated into karate training compared with regular karate training only on measures of physical fitness in youth karate athletes. Methods: Twenty-seven youth karatekas were recruited and allocated to either RNE group (n = 13; age = 15.35 ± 1.66 years; 7 males and 6 females) or an active control group ([CG]; n = 14; 7 males and 7 females; age = 15.30 ± 1.06 years). To track the changes in measures of physical fitness before and after training, tests to assess linear sprint speed (i.e., 10 m), change of direction (CoD) speed (i.e., modified 505 CoD), vertical jumping (i.e., countermovement jump [CMJ] height) and horizontal jumping distance (i.e., standing long jump [SLJ]), and lower-limb asymmetry score (i.e., the difference between SLJ-dominant and non-dominant legs) were carried out. Results: The results indicated significant group-by-time interactions in all measures of physical fitness (effect size [ES] = 1.03 to 2.89). Post-hoc analyses revealed significant changes in the RNE group across all performance measures (effect size [ES] = 0.33 to 1.63). Additionally, the asymmetry score exhibited a moderate decrease from pre to posttest (∆46.96%, ES = 0.64). In contrast, no significant changes were observed in the CG across all fitness measures. Moreover, the individual response analysis indicated that more karatekas from the RNE group consistently achieved improvements beyond the smallest worthwhile change threshold across all fitness measures. Conclusions: In summary, RNE training is an effective approach to enhance various physical fitness measures besides lower-limb asymmetry scores in youth karatekas and is easy to incorporate into regular karate training. Practitioners are therefore encouraged to consistently integrate RNE training to enhance essential physical fitness components in young karatekas. Full article