Search

The COVID‐19 pandemic revealed the pivotal role of digital learning and online lecture videos, leading a shift toward blended and flipped classrooms in the post‐pandemic era. This shift calls for development and (or) refreshment of novel online educational resources, such as Voice‐Over PowerPoint (VOPPT) presentations, specifically designed for asynchronous or synchronous learning methods. However, resource limitations often impede the timely delivery of high‐quality instructional materials. In this descriptive article, the use of AI‐voice cloning technology to automate the creation of VOPPT presentation has been explored. Descripts' Overdub tool, an AI‐voice cloning program, was trained on the 15‐min voice sample of an anatomy professor, and the synthesized voice was used to narrate presentations on inguinal canal and extrahepatic biliary anatomy. The educational use of this novel approach was evaluated based on the second‐year undergraduate medical students' qualitative feedback. Voice similarity analysis using Resemblyzer, an open‐source Python tool, showed a high similarity score of 0.92 between the cloned and original voices. Despite this, students raised concerns about the robotic voice, quick pace, punctuation problems, pronunciation difficulties, and expressed reservations about AI‐generated lecture narration. While the cloned voice closely matched the original, the AI‐generated narration fell short of capturing the nuanced details needed for an effective anatomy instruction. This uncovers AI's current limitations in the educational contexts but establishes a foundation for future progress in AI‐voice cloning technologies aimed at enhancing online educational resources for anatomy and medical education. [ABSTRACT FROM AUTHOR]

Large Language Models (LLMs) have the potential to improve education by personalizing learning. However, ChatGPT‐generated content has been criticized for sometimes producing false, biased, and/or hallucinatory information. To evaluate AI's ability to return clear and accurate anatomy information, this study generated a custom interactive and intelligent chatbot (Anatbuddy) through an Open AI Application Programming Interface (API) that enables seamless AI‐driven interactions within a secured cloud infrastructure. Anatbuddy was programmed through a Retrieval Augmented Generation (RAG) method to provide context‐aware responses to user queries based on a predetermined knowledge base. To compare their outputs, various queries (i.e., prompts) on thoracic anatomy (n = 18) were fed into Anatbuddy and ChatGPT 3.5. A panel comprising three experienced anatomists evaluated both tools' responses for factual accuracy, relevance, completeness, coherence, and fluency on a 5‐point Likert scale. These ratings were reviewed by a third party blinded to the study, who revised and finalized scores as needed. Anatbuddy's factual accuracy (mean ± SD = 4.78/5.00 ± 0.43; median = 5.00) was rated significantly higher (U = 84, p = 0.01) than ChatGPT's accuracy (4.11 ± 0.83; median = 4.00). No statistically significant differences were detected between the chatbots for the other variables. Given ChatGPT's current content knowledge limitations, we strongly recommend the anatomy profession develop a custom AI chatbot for anatomy education utilizing a carefully curated knowledge base to ensure accuracy. Further research is needed to determine students' acceptance of custom chatbots for anatomy education and their influence on learning experiences and outcomes. [ABSTRACT FROM AUTHOR]

The article provides information on a transformer-based clustering model named ClusterBot developed to generate a curriculum map. Topics discussed include the potential of generative artificial intelligence (Gen-AI) to revolutionise medical education, how the accuracy of ClusterBot in identifying relevant text was evaluated, and lessons learned from the human-AI hybrid approach.

Aim: To investigate whether a digital sleep intervention improves child and care giver sleep and psychosocial outcomes. Methods: A total of 120 families with children aged 2–13 years, reporting moderate to severe child behavioural sleep problems, were recruited from a hospital sleep clinic waitlist or the community. Children from non‐English speaking families, with known intellectual disability (IQ < 70) or severe medical problems excluded. Tailored behavioural sleep strategies were delivered to primary care givers via a smart phone app and complementary website. Eligible families completed a baseline questionnaire and child 'sleep check' then received the digital sleep intervention for 5 weeks, and then completed a post questionnaire. Outcomes: care giver report of child sleep as no/mild versus moderate/severe problem over past month (primary outcome); problem child sleep patterns (Brief Infant Sleep Questionnaire or Child Sleep Habits Questionnaire), child temperament, care giver mental health (Kessler 6), care giver sleep, health service use for their child's sleep and time off work/activities to access services. Results: At follow up, care givers reported fewer moderate/severe child sleep problems (84.6–40.7%), improved problem child sleep patterns, better temperament and improved care giver mental health. Care giver sleep quality and quantity remained unchanged. Health service use (averaged over a 6‐month period pre‐ and post‐intervention) fell from 18.9% pre‐ to 14.1% post‐intervention. Conclusion: A digital sleep intervention appears promising in improving sleep in children with moderate/severe behavioural sleep problems, and care giver mental health. It may be a useful alternative to face‐to‐face management of behavioural sleep problems. [ABSTRACT FROM AUTHOR]