Related Experiment Video
Updated: Jan 11, 2026
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Evaluating Conferences as a Tool to Improve Medical Artificial Intelligence Comprehension among Healthcare
Nafisa Islam1, Abdel Rahman Osman1, Laveinia Godfrey2
1Barking, Havering and Redbridge University Hospitals NHS Trust Queen's Hospital, Romford, United Kingdom.
Introduction:
The integration of artificial intelligence (AI) in the United Kingdom's National Health Service may enhance patient care and alleviate systemic pressures. However, adoption of medical AI is challenged by limited educational access, low confidence among staff, and concerns regarding transparency and ethics. We evaluated the National Health Service National AI Conference on its impact on the understanding and attitudes of health care professionals regarding AI.
Methods:
A before-and-after study design was employed using anonymized surveys distributed to conference attendees. The survey assessed participants' roles, prior experience with medical AI, and perceptions of AI's risks, benefits, and applications. Using a 5-point Likert scale, responses were analyzed via Wilcoxon's signed-rank test, after trialing McNemar's test, with statistical significance defined as p < 0.05.
Results:
The survey was completed by 43 attendees. Most were clinical professionals (53.49%), with 65.12% having never attended a similar conference. Most participants (pre-conference: 69.77% vs. post-conference: 85.05%, p = 0.0868) understood the benefits and uses of AI, agreed that AI has the potential to improve patient care (93.02% vs. 95.35%, p = 1.00), were interested in pursuing a career in medical AI (62.79% vs. 67.44%, p = 0.824), and were concerned about the use of AI in health care (65.12% vs. 53.49%, p = 0.419). We observed an increase in understanding of AI after the conference among participants (p = 0.0367). Participant confidence and empowerment increased from 53.49% to 69.77% (p = 0.00319) and from 51.16% to 67.44% (p = 0.00596), respectively; these increases, alongside the increase in understanding of AI, reached statistical significance when analyzed using Wilcoxon's test, but not when dichotomized and analyzed with McNemar's test.
Conclusions:
Our conference may increase AI understanding, confidence, and empowerment among health care professionals, encouraging further research into targeted medical AI education. A national AI curriculum, transparent governance, and robust information technology infrastructure are recommended to support the adoption of AI internationally.
