Comparison of AI-generated and human-made animated videos for medical education: experts and students preferred AI over humans
Abstract
Objective: This study compared medical students and experts, and evaluated a frames‑to‑video AI‑generated problem‑based learning (PBL) trigger against its scene‑matched human‑made animated counterpart in terms of evaluations and preferences. Study Design: A mixed‑methods study was conducted at a medical school. Two scene‑matched videos were used: an AI‑generated video and an animated (human‑made) video. Students (n210; Years 2–5) viewed both videos in counterbalanced order and rated eight 5‑point Likert items for each; they also indicated their preferred video for engagement, emotional impact, and PBL use. A multidisciplinary expert panel (n=104) evaluated only the AI video on comparable items and provided open‑ended comments. Mann–Whitney-U tests compared experts with students on the AI video; Wilcoxon signed‑rank tests compared students’ ratings across videos. Qualitative data underwent thematic analysis. Results: Students rated the AI‑generated video significantly higher than the animated video on all eight items (all p≤.026) and preferred it for engagement (83.8%), emotional impact (81.0%), and PBL use (79.0%). Experts’ ratings of the AI video were also high and exceeded students’ ratings on visual quality, distraction avoidance, and visual consistency (p≤.001). Qualitative themes highlighted realism, suitability for PBL sessions, and strong engagement, while suggested improvements included micro‑continuity, pronunciation, and body language. Conclusion: Within the PBL context, a frames‑to‑video AI workflow produced a fully synthetic trigger that was preferred by students and endorsed by experts. AI‑generated triggers appear feasible, acceptable, and educationally promising, provided attention is given to fine‑grained audiovisual continuity and communication cues.
Downloads
Metrics
-
Abstract91
-
pdf108
References
Davis MH, Harden RM. AMEE Medical Education Guide No. 15: Problem-based learning: a practical guide. Medical Teacher. 1999, 21, 130–40. doi: 10.1080/01421599979743
Basu Roy R, McMahon GT. Video‐based cases disrupt deep critical thinking in problem‐based learning. Medical Education. 2012, 46, 426–35. doi: 10.1111/j.1365-2923.2011.04197.x
De Leng B, Dolmans D, Van De Wiel M, et al. How video cases should be used as authentic stimuli in problem-based medical education. Med Educ. 2007, 41, 181–8. doi: 10.1111/j.1365-2929.2006.02671.x
Chan LK, Patil NG, Chen JY, et al. Advantages of video trigger in problem-based learning. Medical Teacher. 2010, 32, 760–5. doi: 10.3109/01421591003686260
Budakoğlu Iİ, Coşkun Ö, Özeke V. e-PBL with multimedia animations: a design-based research. BMC Med Educ. 2023, 23, 338. doi: 10.1186/s12909-023-04298-x
Dağcı M, Dost A, Çam F. Potential Impacts of AI-Generated Videos on Nursing Care. Nurse Educator. 2025, 50(5), E254-E258, doi: 10.1097/NNE.0000000000001880
Liu Y, Zhang K, Li Y, et al. Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models. 2024, https://doi.org/10.48550/arXiv.2402.17177
Waisberg E, Ong J, Masalkhi M, et al. Concerns with OpenAI’s Sora in Medicine. Ann Biomed Eng. 2024, 52, 1932–4. doi: 10.1007/s10439-024-03505-0
Contreras I, Hossfeld S, De Boer K, et al. Revolutionising Faculty Development and Continuing Medical Education Through AI-Generated Videos. Journal of CME. 2024, 13, 2434322. doi: 10.1080/28338073.2024.2434322
Shu J, Kou H, Zhang J, et al. AI-Generated Instructional Videos: A Systematic Review of Learning Impacts, Applications, and Research Perspectives. 2025 7th International Conference on Computer Science and Technologies in Education (CSTE). Wuhan, China: IEEE 2025:877–81. 10.1109/CSTE64638.2025.11091997
Netland T, Von Dzengelevski O, Tesch K, et al. Comparing human-made and AI-generated teaching videos: An experimental study on learning effects. Computers & Education. 2025, 224, 105164. doi: 10.1016/j.compedu.2024.105164
Lawson McLean A, Gutiérrez Pineda F. Application of transformer architectures in generative video modeling for neurosurgical education. Int J CARS. 2024, 20, 797–805. doi: 10.1007/s11548-024-03266-0
Yeung JWY, Tang FWK, Cheng HY. Tips for Incorporating AI-Generated Video in Disaster Nursing Education. Nurse Educator. 2025, 50, E151–E151. doi: 10.1097/NNE.0000000000001840
Chan MMK, Wan AWH, Cheung DSK, et al. Integration of Artificial Intelligence in Nursing Simulation Education: A Scoping Review. Nurse Educator. 2025, 50, 195–200. doi: 10.1097/NNE.0000000000001851
Sardesai N, Russo P, Martin J, et al. Utilizing generative conversational artificial intelligence to create simulated patient encounters: a pilot study for anaesthesia training. Postgraduate Medical Journal. 2024, 100, 237–41. doi: 10.1093/postmj/qgad137
Kiger ME, Varpio L. Thematic analysis of qualitative data: AMEE Guide No. 131. Medical Teacher. 2020, 42, 846–54. doi: 10.1080/0142159X.2020.1755030
Leppink J. Data analysis: more expensive does not imply better. Rev Esp Edu Med. 2022, 3. doi: 10.6018/edumed.516731
Arkün-Kocadere S, Çağlar Özhan Ş. Video Lectures With AI-Generated Instructors: Low Video Engagement, Same Performance as Human Instructors. IRRODL. 2024, 25:350–69. doi: 10.19173/irrodl.v25i3.7815
Masters K. Medical Teacher ’s first ChatGPT’s referencing hallucinations: Lessons for editors, reviewers, and teachers. Medical Teacher, 2023; 45, 673–5. doi: 10.1080/0142159X.2023.2208731
Copyright (c) 2025 Servicio de Publicaciones de la Universidad de Murcia
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
The works published in this magazine are subject to the following terms:
1. The Publications Service of the University of Murcia (the publisher) preserves the economic rights (copyright) of the published works and favors and allows them to be reused under the use license indicated in point 2.
2. The works are published under a Creative Commons Attribution-NonCommercial-NoDerivative 4.0 license.
3. Self-archiving conditions. Authors are allowed and encouraged to disseminate electronically the pre-print versions (version before being evaluated and sent to the journal) and / or post-print (version evaluated and accepted for publication) of their works before publication , since it favors its circulation and earlier diffusion and with it a possible increase in its citation and reach among the academic community.
