ChatGPT Virtual Assistant for Breast Reconstruction: Assessing Preferences for a Traditional Chatbot versus a Human AI VideoBot
View abstract on PubMed
Summary
This summary is machine-generated.Patients found AI VideoBots more engaging than traditional chatbots for plastic surgery information. While both AI tools were equally trustworthy, most preferred the human-like avatar for a more captivating telehealth experience.
Area Of Science
- Medical Informatics
- Artificial Intelligence in Healthcare
- Human-Computer Interaction
Background
- Artificial intelligence (AI) is transforming telehealth, with AI chatbots like Chat Generative Pretrained Transformer (ChatGPT) demonstrating potential in medicine.
- ChatGPT can provide basic patient education for plastic and reconstructive surgery (PRS) procedures.
- Patient preference between AI VideoBots and traditional chatbots in PRS remains unexamined.
Purpose Of The Study
- To compare patient perceptions of AI VideoBots versus traditional chatbots for plastic and reconstructive surgery (PRS) information.
- To evaluate the effectiveness, trust, and user preference for AI-driven telehealth tools.
Main Methods
- Developed an AI VideoBot by integrating ChatGPT with Synthesia, a human AI avatar platform.
- Created an interactive experience answering common breast reconstruction questions via the VideoBot on Tolstoy.
- Utilized Zapier for a ChatGPT-integrated chatbot and surveyed 396 female participants on trust using a validated 16-item scale.
Main Results
- Perceptions of truthfulness, believability, expertise, ease of use, and safety were similar between the VideoBot and chatbot.
- A majority of participants preferred the VideoBot (63.5%) over the traditional chatbot (28.1%) due to its captivating nature.
- 77% of participants expressed a preference for an avatar matching their gender and race.
Conclusions
- AI VideoBots and text-based chatbots exhibit comparable effectiveness, usability, and trust in telehealth.
- The human-like qualities of VideoBots enhance patient interactivity and engagement.
- Future research should investigate the impact of race and gender concordance in telehealth for personalized patient experiences.

