Browse Articles
Discover research articles across all indexed journals
Reducing annotation burden in physical activity research using vision language models
Abstract Data from wearable devices collected in free-living settings, and labelled with physical activity behaviours compatible with health research, are essential for both validating existing wearable-based measurement approaches and developing novel machine learning approaches. One common way of obtaining these labels relies on laborious human annotation of sequences of images captured by body-worn cameras. The aim of this study was to investigate whether open-source vision-language models could accurately annotate activity intensity classes in wearable camera-based validation studies, thereby reducing the annotation burden. We compared the performance of three vision language models and two discriminative models on two free-living validation studies with 161 and 111 participants, collected in Oxfordshire, United Kingdom and Sichuan, China, respectively, using the Autographer (OMG Life, defunct) wearable camera. We found that the best open-source vision-language model (VLM) and fine-tuned discriminative model (DM) achieved comparable performance when predicting sedentary behaviour from single images on unseen participants in the Oxfordshire study; median F 1 -scores: VLM = 0.89 (0.84, 0.92), DM = 0.91 (0.86, 0.95). Performance declined for light [VLM = 0.60 (0.56, 0.67), DM = 0.70 (0.63, 0.79)], and moderate-to-vigorous intensity physical activity [VLM = 0.66 (0.53, 0.85); DM = 0.72 (0.58, 0.84)]. When applied to the external Sichuan study, performance fell across all intensity categories, with median Cohen’s κ scores falling from 0.54 (0.49, 0.64) to 0.26 (0.15, 0.37) for the VLM, and from 0.67 (0.60, 0.74) to 0.19 (0.10, 0.30) for the DM. Freely available computer vision models could help annotate sedentary behaviour, typically the most prevalent activity of daily living, from wearable camera images within similar populations to seen data, reducing the annotation burden when using cameras as the source of ground-truth.
Assessment of respirable silica in ambient air of Dhaka and evaluation of human health risks
Investigation on the transition depth of open pit to underground mining in Sijiaying iron mine
Autoimmune reactions to ACE2 triggered by SARS-CoV-2 S protein and their histopathological consequences
Oxygen-depleted and ferruginous seawater composition imprinted in Early Cretaceous Fe-rich Al-glauconites in marginal marine deposits
Correction: Morphological and metabolic alterations in different stages of Alzheimer’s diseases: a study using surface-based morphometry (SBM) and amide proton transfer (APT) imaging
Increasing the accessibility to colorectal cancer screening in the community via faecal immunochemical test kits dispensed via automated kiosk
Characterization of air submicrobubbles for water treatment under different generation conditions
Optimized NaOH and KOH treatments for interfacial and mechanical enhancement of agro-marine ash reinforced aluminum composites
Combining AlphaFold—based AI docking with omics technologies preliminarily unveils the mechanism of Guominkang in treating allergic rhinitis
Enhancing queries for code generation with reinforcement learning
Abstract We present a reinforcement learning framework that enhances natural language queries to improve DeepSeek code generation. A parametric refiner (Qwen with LoRA) is trained via REINFORCE while the generator remains fixed, using a scalar reward that can combine text similarity (BLEU-4, ROUGE-L, F1, Overlap) with execution signals (unit tests, syntax/timeout penalties). On the DS1000 benchmark (800 train / 200 test), RL4QE improves the code similarity by 34.3%. Ablations show that BLEU-4 is the most reliable text reward overall (with F1 competitive on a larger scale), and LoRA with rank $$r{=}8$$ outperforms complete fine-tuning on most metrics while being more parameter efficient. The approach is transferred across foundation models (e.g., Qwen1.5/2/2.5 variants), where architecture often matters more than size. RL4QE is easy to integrate in practice (LoRA in attention projections) and supports reproducibility.
Comprehensive evaluation of optimization algorithms for medical image segmentation
Oculomotor chronometry of spoken word structure processing
Reconstruction of Indigenous settlements in ecologically fragile areas through spatial distribution patterns and suitability in Xiaoliangshan, SW China
A mixed-methods cross-sectional study on the associations between internalized stigma, physical activity, and quality of life in people with multiple sclerosis
Proposing a material selection indicator for the design of extended lifespan products
Abstract Material durability has emerged as a key factor influencing product longevity, safety, and cost-effectiveness across various sectors, including the construction, automotive, and electronics sectors. With the growing emphasis on sustainability, there is an increasing demand for materials that not only exhibit high performance but also align with circular economy practices such as reuse, repair, remanufacturing, and repurpose. The challenge lies in the existing material selection frameworks, which frequently fail to comprehensively integrate multiple aspects of durability with environmental impact. Most current indicators predominantly focus on ensuring material durability over time without sufficiently considering the complete lifecycle environmental footprint. To address these challenges, this study introduces the Specific Durability Performance, a quantitative approach that merges mechanical, thermal, and chemical durability with a carbon footprint into a singular value ranging from 0.0 to 1.0. The methodology involves calculating key parameters for each material, such as tensile strength, fatigue resistance, service temperature range, flammability, and chemical resistance, which are weighted according to the criteria that assess the specific use requirements of the product. Each dimension of durability was benchmarked against an “ideal” reference material, and the carbon footprint was also benchmarked. These results were then synthesized into a unique score designed to facilitate material selection by balancing performance and sustainability. This approach was applied to two case studies with distinct design requirements, demonstrating the usefulness of the proposed indicator in identifying materials that offer robust structural performance and reduced environmental impact.