The huge number of images shared on the Web makes effective cataloguing methods for efficient storage and retrieval procedures specifically tailored on the end-user needs a very demanding and crucial issue. In this paper, we investigate the applicability of Automatic Image Annotation (AIA) for image tagging with a focus on the needs of database expansion for a news broadcasting company. First, we determine the feasibility of using AIA in such a context with the aim of minimizing an extensive retraining whenever a new tag needs to be incorporated in the tag set population. Then, an image annotation tool integrating a Convolutional Neural Network model (AlexNet) for feature extraction and a K-Nearest-Neighbours classifier for tag assignment to images is introduced and tested. The obtained performances are very promising addressing the proposed approach as valuable to tackle the problem of image tagging in the framework of a broadcasting company, whilst not yet optimal for integration in the business process.
DOCUMENT
Communication between healthcare professionals and deaf patients has been particularly challenging during the COVID-19 pandemic. We have explored the possibility to automatically translate phrases that are frequently used in the diagnosis and treatment of hospital patients, in particular phrases related to COVID-19, from Dutch or English to Dutch Sign Language (NGT). The prototype system we developed displays translations either by means of pre-recorded videos featuring a deaf human signer (for a limited number of sentences) or by means of animations featuring a computer-generated signing avatar (for a larger, though still restricted number of sentences). We evaluated the comprehensibility of the signing avatar, as compared to the human signer. We found that, while individual signs are recognized correctly when signed by the avatar almost as frequently as when signed by a human, sentence comprehension rates and clarity scores for the avatar are substantially lower than for the human signer. We identify a number of concrete limitations of the JASigning avatar engine that underlies our system. Namely, the engine currently does not offer sufficient control over mouth shapes, the relative speed and intensity of signs in a sentence (prosody), and transitions between signs. These limitations need to be overcome in future work for the engine to become usable in practice.
DOCUMENT
This method paper presents a template solution for text mining of scientific literature using the R tm package. Literature to be analyzed can be collected manually or automatically using the code provided with this paper. Once the literature is collected, the three steps for conducting text mining can be performed as outlined below:• loading and cleaning of text from articles,• processing, statistical analysis, and clustering, and• presentation of results using generalized and tailor-made visualizations.The text mining steps can be applied to a single, multiple, or time series groups of documents.References are provided to three published peer reviewed articles that use the presented text mining methodology. The main advantages of our method are: (1) Its suitability for both research and educational purposes, (2) Compliance with the Findable Accessible Interoperable and Reproducible (FAIR) principles, and (3) code and example data are made available on GitHub under the open-source Apache V2 license.
DOCUMENT
With the proliferation of misinformation on the web, automatic misinformation detection methods are becoming an increasingly important subject of study. Large language models have produced the best results among content-based methods, which rely on the text of the article rather than the metadata or network features. However, finetuning such a model requires significant training data, which has led to the automatic creation of large-scale misinformation detection datasets. In these datasets, articles are not labelled directly. Rather, each news site is labelled for reliability by an established fact-checking organisation and every article is subsequently assigned the corresponding label based on the reliability score of the news source in question. A recent paper has explored the biases present in one such dataset, NELA-GT-2018, and shown that the models are at least partly learning the stylistic and other features of different news sources rather than the features of unreliable news. We confirm a part of their findings. Apart from studying the characteristics and potential biases of the datasets, we also find it important to examine in what way the model architecture influences the results. We therefore explore which text features or combinations of features are learned by models based on contextual word embeddings as opposed to basic bag-of-words models. To elucidate this, we perform extensive error analysis aided by the SHAP post-hoc explanation technique on a debiased portion of the dataset. We validate the explanation technique on our inherently interpretable baseline model.
DOCUMENT
A common strategy to assign keywords to documents is to select the most appropriate words from the document text. One of the most important criteria for a word to be selected as keyword is its relevance for the text. The tf.idf score of a term is a widely used relevance measure. While easy to compute and giving quite satisfactory results, this measure does not take (semantic) relations between words into account. In this paper we study some alternative relevance measures that do use relations between words. They are computed by defining co-occurrence distributions for words and comparing these distributions with the document and the corpus distribution. We then evaluate keyword extraction algorithms defined by selecting different relevance measures. For two corpora of abstracts with manually assigned keywords, we compare manually extracted keywords with different automatically extracted ones. The results show that using word co-occurrence information can improve precision and recall over tf.idf.
DOCUMENT
A common strategy to assign keywords to documents is to select the most appropriate words from the document text. One of the most important criteria for a word to be selected as keyword is its relevance for the text. The tf.idf score of a term is a widely used relevance measure. While easy to compute and giving quite satisfactory results, this measure does not take (semantic) relations between words into account.
DOCUMENT
Background Prehabilitation offers patients the opportunity to actively participate in their perioperative care by preparing themselves for their upcoming surgery. Experiencing barriers may lead to non-participation, which can result in a reduced functional capacity, delayed post-operative recovery and higher healthcare costs. Insight in the barriers and facilitators to participation in prehabilitation can inform further development and implementation of prehabilitation. The aim of this review was to identify patient-experienced barriers and facilitators for participation in prehabilitation. Methods For this mixed methods systematic review, articles were searched in PubMed, EMBASE and CINAHL. Articles were eligible for inclusion if they contained data on patient-reported barriers and facilitators to participation in prehabilitation in adults undergoing major surgery. Following database search, and title and abstract screening, full text articles were screened for eligibility and quality was assessed using the Mixed Method Appraisal Tool. Relevant data from the included studies were extracted, coded and categorized into themes, using an inductive approach. Based on these themes, the Capability, Opportunity, Motivation, Behaviour (COM-B) model was chosen to classify the identified themes. Results Three quantitative, 14 qualitative and 6 mixed methods studies, published between 2007 and 2022, were included in this review. A multitude of factors were identified across the different COM-B components. Barriers included lack of knowledge of the benefits of prehabilitation and not prioritizing prehabilitation over other commitments (psychological capability), physical symptoms and comorbidities (physical capability), lack of time and limited financial capacity (physical opportunity), lack of social support (social opportunity), anxiety and stress (automatic motivation) and previous experiences and feeling too fit for prehabilitation (reflective motivation). Facilitators included knowledge of the benefits of prehabilitation (psychological capability), having access to resources (physical opportunity), social support and encouragement by a health care professional (social support), feeling a sense of control (automatic motivation) and beliefs in own abilities (reflective motivation). Conclusions A large number of barriers and facilitators, influencing participation in prehabilitation, were found across all six COM-B components. To reach all patients and to tailor prehabilitation to the patient’s needs and preferences, it is important to take into account patients’ capability, opportunity and motivation.
DOCUMENT
Study selection: Randomized controlled trials published after 2007 with (former) healthcare patients ≥ 21 years of age were included if physical activity was measured objectively using a wearable monitor for both feedback and outcome assessment. The main goal of included studies was promoting physical activity. Any concurrent strategies were related only to promoting physical activity. Data extraction: Effect sizes were calculated using a fixed-effects model with standardized mean difference. Information on study characteristics and interventions strategies were extracted from study descriptions. Data synthesis: Fourteen studies met the inclusion criteria (total n = 1,902), and 2 studies were excluded from meta-analysis. The overall effect size was in favour of the intervention groups (0.34, 95% CI 0.23–0.44, p < 0.01). Study characteristics and intervention strategies varied widely. Conclusion: Healthcare interventions using feedback on objectively monitored physical activity have a moderately positive effect on levels of physical activity. Further research is needed to determine which strategies are most effective to promote physical activity in healthcare programmes. Lay Abstract Wearable technology is progressively applied in health care and rehabilitation to provide objective insight into physical activity levels. In addition, feedback on physical activity levels delivered by wearable monitors might be beneficial for optimizing their physical activity. A systematic review and meta-analysis was conducted to evaluate the effectiveness of interventions using feedback on objectively measured physical activity in patient populations. Fourteen studies including 1902 patients were analyzed. Overall, the physical activity levels of the intervention groups receiving objective feedback on physical activity improved, compared to the control groups receiving no objective feedback. Mostly, a variety of other strategies were applied in the interventions next to wearable technology. Together with wearable technology, behavioral change strategies, such as goal-setting and action planning seem to be an important ingredient to promote physical activity in health care and rehabilitation. LinkedIn: https://www.linkedin.com/in/hanneke-braakhuis-b9277947/ https://www.linkedin.com/in/moniqueberger/
MULTIFILE
Purpose Self-injury is common in forensic psychiatric settings. Recent research offers some insights into the functions and management of self-injurious behaviour but generally focusses on either the experiences of staff or patients. This study aims to explore the experiences of both staff and patients with non-suicidal self-injury in a Dutch forensic psychiatric hospital. Design/methodology/approach In total, 6 patients and 11 staff members were interviewed about the functions they ascribe to self-injurious behaviour, the emotional experience provoked by this behaviour and the management of self-injurious behaviour. The interviews were transcribed and analysed using a thematic analysis. Findings Four main themes resulted from the analysis: functions; emotional distancing; patient needs; and management. Overall, findings illustrate that staff reports limited knowledge of the different functions of self-injury. To circumvent potential automatic stereotypical judgement, staff should proactively engage in conversation about this topic with their patients. In managing self-injurious behaviour, clarity and uniformity among staff members should be promoted, and collaboration between the staff and patients is desirable. Staff recognised the potential benefit of a management guideline. Staff may find detached coping strategies to be effective but should be vigilant to not let this evolve into excessive detachment. Practical implications Increased knowledge and awareness of self-injury functions among staff can allow for better understanding and evaluation of self-injury incidents. Circumvention of automatic, stereotypical judgement of self-injurious behaviour is warranted, and more accessible explanations of the variety of functions of self-injury should be used. More proactive engagement in conversations about functions of self-injury by staff, can facilitate this. Detached coping can help staff to remain resilient in their job, but requires vigilance to prevent this from turning into excessive detachment. Clarity and uniformity among staff when managing self-injury incidents is considered beneficial by both patients and staff. A guideline may facilitate this. When imposing restrictions on patients, staff should strive to establish collaboration with the patient in determining the course of action and ensure the restriction is temporary. Originality/value The impact of self-injurious behaviour on all those involved can be enormous. More research is needed into experiences of both patients and staff members regarding the impact, motivations, precipitants and functions of self-injurious behaviour, and effective treatment of it.
LINK
Adverse Outcome Pathways (AOPs) are conceptual frameworks that tie an initial perturbation (molecular initiat- ing event) to a phenotypic toxicological manifestation (adverse outcome), through a series of steps (key events). They provide therefore a standardized way to map and organize toxicological mechanistic information. As such, AOPs inform on key events underlying toxicity, thus supporting the development of New Approach Methodologies (NAMs), which aim to reduce the use of animal testing for toxicology purposes. However, the establishment of a novel AOP relies on the gathering of multiple streams of evidence and infor- mation, from available literature to knowledge databases. Often, this information is in the form of free text, also called unstructured text, which is not immediately digestible by a computer. This information is thus both tedious and increasingly time-consuming to process manually with the growing volume of data available. The advance- ment of machine learning provides alternative solutions to this challenge. To extract and organize information from relevant sources, it seems valuable to employ deep learning Natural Language Processing techniques. We review here some of the recent progress in the NLP field, and show how these techniques have already demonstrated value in the biomedical and toxicology areas. We also propose an approach to efficiently and reliably extract and combine relevant toxicological information from text. This data can be used to map underlying mechanisms that lead to toxicological effects and start building quantitative models, in particular AOPs, ultimately allowing animal-free human-based hazard and risk assessment.
DOCUMENT
