The global evidence on the risk of symptoms of Long Covid in general populations infected with SARS-CoV-2 compared to uninfected comparator/control populations remains unknown. We conducted a systematic literature search using multiple electronic databases from January 1, 2022, to August 1, 2024. Included studies had ≥100 people with confirmed or self-reported COVID-19 at ≥28 days following infection onset, and an uninfected comparator/control group. Results were summarised descriptively and meta-analyses were conducted to derive pooled risk ratio estimates. 50 studies totaling 14,661,595 people were included. In all populations combined, there was an increased risk of a wide range of 39 out of 40 symptoms in those infected with SARS‑CoV‑2 compared to uninfected controls. The symptoms with the highest pooled relative risks were loss of smell (RR 4.31; 95% CI 2.66, 6.99), loss of taste (RR 3.71; 95% CI 2.22, 7.26), poor concentration (RR 2.68; 95% CI 1.66, 4.33), impaired memory (RR 2.53; 95% CI 1.82, 3.52), and hair loss/alopecia (RR 2.38; 95% CI 1.69, 3.33). This evidence synthesis, of 50 controlled studies with a cumulative participant count exceeding 14 million people, highlights a significant risk of diverse long-term symptoms in individuals infected with SARS-CoV-2, especially among those who were hospitalised.
This thesis presents a cognitive poetic exploration of ekphrasis as a creative writing practice of responding to visual art. The study examines the production and reception of ekphrasis from a new perspective by considering the influence of both an artwork and an ekphrastic text on interpretation. This thesis uses naturalistic and experimental data to investigate the techniques of writing an ekphrastic text alongside participants’ reported experiences of ekphrasis. To examine ekphrastic writing, this thesis employs three collections of contemporary ekphrastic texts published in The Ekphrastic Review. The collections are assembled in response to three paintings: A Blind Girl Reading by Ejnar Nielsen, The Dream by Frida Kahlo, and Roofscape by Gustave Caillebotte. To analyse readerly experience of ekphrasis, this thesis utilises reader-response data generated during the discussions of selected ekphrases by two groups of participants. The resulting findings are consolidated in the model of ekphrastic intervention, which selectively adopts the tools of Text World Theory, visual grammar, the model of narrative interrelation, and the framework of conceptual integration. To examine ekphrastic writing, the model outlines ways of responding to art by focussing on how language potentially can closely or distantly align the representations of an image and a text. To examine ekphrastic reading, the model elicits two cognitive acts involved in interpreting ekphrasis: ekphrastic interrelation and blending. This thesis offers the following contributions. First, ekphrasis is approached from a novel holistic angle, including both visual and verbal into consideration. Second, ekphrasis is framed as a creative writing practice, providing useful tools for enhancing ekphrastic writing skills. Third, the introduced model offers a systematic way of analysing ekphrasis, approaching both readerly interpretations and textual features that may elicit such responses. Finally, the empirical examination of ekphrasis diversifies the tradition of reader response, validating cooperative and resistant behaviours in participants’ reported experiences.
This paper presents a novel algorithm that leverages cutting-edge machine-learning techniques to accurately and efficiently detect AI-generated texts. Rapid advancements in natural language processing models have led to the generation of text closely resembling human language, making it increasingly difficult to differentiate between human and AI-generated content. However, misuse of such texts presents a serious and imminent threat to the quality of academic publishing. This underscores the urgent need for robust detection mechanisms to ensure information quality, maintain trust, and preserve the integrity of research publications. Our proposed model outperformed existing algorithms for accuracy with less computational complexity. The proposed model is a feature-based hybrid deep learning network that leverages part-of-speech tagging and integrates Bidirectional Long Short-Term Memory (BiLSTM) networks with Attention modules. The initial module extracts local contextual features using convolutional layers, followed by BiLSTM layers that capture long-term dependencies from past and future sequences. An attention mechanism highlights critical sequence components, enhancing the model’s focus on relevant data. The outputs from the attention and initial modules are concatenated through a residual connection, ensuring comprehensive feature representation. This combination is then fed into dense layers for final classification, effectively balancing feature richness and computational efficiency. The proposed model was evaluated on two benchmark datasets, achieving 85.00% and 88.00% accuracy, respectively.
Roma people in Europe experience racism and various forms of discrimination: exclusion from the formal labour market, limited access to education and healthcare, and housing deprivation and segregation. At the same time, they respond to this situation by displaying a variety of creative micro-practices and daily forms of resistance, skilfully navigating constraints and opportunities. In this article we focus on how Romanian Roma migrants experience and challenge housing deprivation and segregation in Italy, where thousands of destitute Roma live in either informal settlements, constantly targeted by forced evictions, or within state-funded camps. More specifically, we ask how these adverse housing conditions are navigated differently by Roma women and men. Drawing on the concepts of ‘intersectionality’ and ‘social navigation’, we analyse 24 life stories collected in Italy, as part of a research project investigating the effects of public policies on destitute Roma migrants in France, Italy, and Spain. We argue that, despite both Roma women and men experience housing deprivation and segregation, they navigate these unfavourable circumstances through strategies shaped by their gendered social location and its intersection with age and parental responsibilities. Our analysis shows that Roma migrant women predominantly reconcile informal economic activities with a performance of feminine domesticity and obedience. Especially young women resort to marriage and separation as tools to navigate different housing arrangements. Finally, mothers navigate state institutions and informal settlements in the attempt to preserve both their children’s safety and family unity. In contrast, Roma migrant men leverage their relatively privileged status in the public space, by more easily negotiating access to the formal labour market, including at a younger age. Men with parental responsibilities also prioritise their children’s safety, but the targeting of male-specific socialising practices in state institutions can make them reassert the relative behavioural freedom that they enjoy in informal settlements.
