Objective:Acknowledging study limitations in a scientific publication is a crucial element in scientific transparency and progress. However, limitation reporting is often inadequate. Natural language processing (NLP) methods could support automated reporting checks, improving research transparency. In this study, our objective was to develop a dataset and NLP methods to detect and categorize self-acknowledged limitations (e.g., sample size, blinding) reported in randomized controlled trial (RCT) publications.Methods:We created a data model of limitation types in RCT studies and annotated a corpus of 200 full-text RCT publications using this data model. We fine-tuned BERT-based sentence classification models to recognize the limitation sentences and their types. To address the small size of the annotated corpus, we experimented with data augmentation approaches, including Easy Data Augmentation (EDA) and Prompt-Based Data Augmentation (PromDA). We applied the best-performing model to a set of about 12K RCT publications to characterize self-acknowledged limitations at larger scale.Results:Our data model consists of 15 categories and 24 sub-categories (e.g., Population and its sub-category DiagnosticCriteria). We annotated 1090 instances of limitation types in 952 sentences (4.8 limitation sentences and 5.5 limitation types per article). A fine-tuned PubMedBERT model for limitation sentence classification improved upon our earlier model by about 1.5 absolute percentage points in F1 score (0.821 vs. 0.8) with statistical significance (). Our best-performing limitation type classification model, PubMedBERT fine-tuning with PromDA (Output View), achieved an F1 score of 0.7, improving upon the vanilla PubMedBERT model by 2.7 percentage points, with statistical significance ().Conclusion:The model could support automated screening tools which can be used by journals to draw the authors’ attention to reporting issues. Automatic extraction of limitations from RCT publications could benefit peer review and evidence synthesis, and support advanced methods to search and aggregate the evidence from the clinical trial literature.
MULTIFILE
Traces of condom lubricants in fingerprints can be valuable information in cases of sexual assault. Ideally, not only confirmation of the presence of the condom but also determination of the type of condom brand used can be retrieved. Previous studies have shown to be able to retrieve information about the condom brand and type from fingerprints containing lubricants using various analytical techniques. However, in practice fingerprints often appear latent and need to be detected first, which is often achieved by cyanoacrylate fuming. In this study, we developed a desorption electrospray ionization mass spectrometry (DESI-MS) method which, combined with principal component analysis and linear discriminant analysis (PCA-LDA), allows for high accuracy classification of condom brands and types from fingerprints containing condom lubricant traces. The developed method is compatible with cyanoacrylate (CA) fuming. We collected and analyzed a representative dataset for the Netherlands comprising 32 different condoms. Distinctive lubricant components such as polyethylene glycol (PEG), polydimethylsiloxane (PDMS), octoxynol-9 and nonoxynol-9 were readily detected using the DESI-MS method. Based on the analysis of lubricant spots, a 99.0% classification accuracy was achieved. When analyzing lubricant containing fingerprints, an overall accuracy of 90.9% was obtained. Full chemical images could be generated from fingerprints, showing the distribution of lubricant components such as PEG and PDMS throughout the fingerprint, while still allowing for classification. The developed method shows potential for the development of DESI-MS based analyses of CA treated exogenous compounds from fingerprints for use in forensic science.
MULTIFILE
This Professional Doctorate (PD) project explores the intersection of artistic research, digital heritage, and interactive media, focusing on the reimagining of medieval Persian bestiaries through high dark fantasy and game-making. The research investigates how the process of creation with interactive 3D media can function as a memory practice. At its core, the project treats bestiaries—pre-modern collections of real and imaginary classifications of the world—as a window into West and Central Asian flora, fauna, and the landscape of memory, serving as both repositories of knowledge and imaginative, cosmological accounts of the more-than-human world. As tools for exploring non-human pre-modern agency, bestiaries offer a medium of speculative storytelling, and explicate the unstable nature of memory in diasporic contexts. By integrating these themes into an interactive digital world, the research develops new methodologies for artistic research, treating world-building as a technique of attunement to heritage. Using a practice-based approach, the project aligns with MERIAN’s emphasis on "research in the wild," where artistic and scientific inquiries merge in experimental ways. It engages with hard-core game mechanics, mythopoetic decompressed environmental storytelling, and hand-crafted detailed intentional world-building to offer new ways of interacting with the past that challenges nostalgia and monumentalization. How can a cultural practice do justice to other, more experimental forms of remembering and encountering cultural pasts, particularly those that embrace the interconnections between human and non-human entities? Specifically, how can artistic practice, through the medium of a virtual, bestiary-inspired dark fantasy interactive media, allow for new modes of remembering that resist idealized and monumentalized histories? What forms of inquiry can emerge when technology (3D media, open-world interactive digital media) becomes a tool of attention and a site of experimental attunement to cosmological heritage?
Organisations are increasingly embedding Artificial Intelligence (AI) techniques and tools in their processes. Typical examples are generative AI for images, videos, text, and classification tasks commonly used, for example, in medical applications and industry. One danger of the proliferation of AI systems is the focus on the performance of AI models, neglecting important aspects such as fairness and sustainability. For example, an organisation might be tempted to use a model with better global performance, even if it works poorly for specific vulnerable groups. The same logic can be applied to high-performance models that require a significant amount of energy for training and usage. At the same time, many organisations recognise the need for responsible AI development that balances performance with fairness and sustainability. This KIEM project proposal aims to develop a tool that can be employed by organizations that develop and implement AI systems and aim to do so more responsibly. Through visual aiding and data visualisation, the tool facilitates making these trade-offs. By showing what these values mean in practice, which choices could be made and highlighting the relationship with performance, we aspire to educate users on how the use of different metrics impacts the decisions made by the model and its wider consequences, such as energy consumption or fairness-related harms. This tool is meant to facilitate conversation between developers, product owners and project leaders to assist them in making their choices more explicit and responsible.