4 research outputs found
Interplay between leadership and patient safety in dentistry: a dental hospital-based cross-sectional study
No full textObjectives: The study aimed to study the association of leadership practices and patient safety culture in a dental hospital. Design: Hospital-based, cross-sectional study Setting: Riphah Dental Hospital (RDH), Islamabad, Pakistan. Participants: All dentists working at RDH were invited to participate. Main outcome measures: A questionnaire comprised of the Transformational Leadership Scale (TLS) and the Dental adapted version of the Medical Office Survey of Patient Safety Culture (DMOSOPS) was distributed among the participants. The response rates for each dimension were calculated. The positive responses were added to calculate scores for each of the patient safety and leadership dimensions and the Total Leadership Score (TLS) and total patient safety score (TPSS). Correlational analysis is performed to assess any associations. Results: A total of 104 dentists participated in the study. A high positive response was observed on three of the leadership dimensions: inspirational communication (85.25%), intellectual stimulation (86%), and supportive leadership (75.17%). A low positive response was found on the following items: ‘acknowledges improvement in my quality of work’ (19%) and ‘has a clear sense of where he/she wants our unit to be in 5 years’ (35.64%). The reported positive responses in the patient safety dimensions were high on three of the patient safety dimensions: organisational learning (78.41%), teamwork (82.91%), and patient care tracking/follow-up (77.05%); and low on work pressure and pace (32.02%). A moderately positive correlation was found between TLS and TPSS (r=0.455, p<0.001). Conclusions: Leadership was found to be associated with patient safety culture in a dental hospital. Leadership training programmes should be incorporated during dental training to prepare future leaders who can inspire a positive patient safety culture
Exploring the factors affecting career progression in informal faculty mentoring sessions within mentor and mentee relationships: a qualitative study
No full textAbstract Background Mentoring plays a pivotal role in mentees’ professional advancement. However, the factors that affect career progression in informal mentoring relationships, especially with respect to faculty members, have not been extensively explored. This study aimed to explore the factors that affect career progression in informal faculty mentorings within mentor and mentee relationships. Methodology A Qualitative Exploratory Study was designed and conducted from May to October 2023. Faculty members with informal mentoring relationships were recruited through purposive sampling. Seven faculty mentors and eight faculty mentees from various institutions participated in the study. Semi-structured interviews were used to collect data, which were audio-recorded and verbatim transcribed. The transcripts were then analysed using NVivo Software and coded. Braun and Clark’s framework was used for the thematic analysis. The study adhered to the COREQ (Consolidated Criteria for Reporting Qualitative Research) checklist to ensure comprehensive and transparent reporting of the qualitative research process. Results A total of 76 codes emerged which were classified into six themes: (1) an ideal mentee, (2) an ideal mentor, (3) factors promoting the relationship, (4) the role of gender, (5) factors deteriorating the relationship, (6) overcoming barriers / trust-building strategies. Within each theme, mentor–mentee needs and behaviours were identified, which could lead to positive or deteriorating outcomes. Conclusion Factors affecting career progression in informal faculty mentoring sessions include mentees' positive mindset, internal motivation and clarity of vision, mentors' skills, reputation, and role modelling. Effective communication, trust, respect, and clear goals are also essential. Challenges, such as busy schedules, cross-gender mentoring, and societal biases, affect these relationships. Overcoming these barriers involves sharing experiences, psychosocial support, empathy, active listening, and feedback
Jervell and Lange‐Nielsen Syndrome (JLNS) in a 13‐Year‐Old Girl: A Rare Case Report
No full textABSTRACT JLNS is a rare genetic disorder characterized by congenital sensorineural hearing loss and a prolonged QTc interval, leading to life‐threatening arrhythmias. Early diagnosis, beta‐blocker therapy, lifestyle modifications, and consideration of ICD surgery are critical in managing sudden cardiac death risk
Inferencia de lenguaje natural: un caso en español
No full textThis thesis examines Natural Language Inference (NLI), or Recognizing Textual Entailment (RTE), focusing on the under-explored Spanish language context with a multi-genre configuration accounting for contrasting, entailment, reasoning and neutral semantic relationship types between sentences.
All the code and models can be found in this private repository: https://zenodo.org/records/14219405This thesis examines Natural Language Inference (NLI), or Recognizing Textual Entailment (RTE), focusing on the under-explored Spanish language context. While English NLI has been extensively researched, Spanish NLI efforts are limited. To address this, a comprehensive multi-genre dataset is developed, incorporating unique labels for contrast, entailment, reasoning, and neutrality, particularly highlighting reasoning to account for causal text relationships. The dataset’s effectiveness is evaluated using BERT-based models, including stress tests and arti- fact assessments. Additionally, the study explores generalization across various genre combinations and compares results with established datasets like XNLI. Finally, the performance of Large Language Models through zero and few-shot prompts is assessed. The findings reveal that BERT-based models achieve over 60% accuracy and demonstrate robustness, while combining the new dataset with XNLI offers above 70% accuracy. In contrast, Large Language Models underperform, achieving less than 50% accuracy in most configurations, but achieve a close performance when testing on human validated datasets.MaestríaNatural Language ProcessingNatural Language UnderstandingNatural Language Inferenc
