P0867 - The Future of IBD Management: Insights From AI Innovations With ChatGPT and Google Bard

Raj Mohan Ram Mohan, MD, Sai Reshma Magam, MD, Sai Greeshma Magam, MD, Melvin Joy, MD, Dilman Natt, MD, Winghang Lau, MD, Abhishek Tadikonda, MD, Leeza Pannikodu, MD, Jiten Desai, MD, Rucha Jiyani, MD, Saher Sheikh, MD, Krishnaiyer Subramani, MD, Paul Mustacchia, MD, MBA
Nassau University Medical Center, East Meadow, NY

Introduction: Artificial Intelligence (AI) has notably transformed healthcare, especially in diagnosing and treating Inflammatory Bowel Disease (IBD) and other digestive disorders. Essential AI tools, such as ChatGPT and Google Bard, can interpret endoscopic imagery, analyze diverse samples, simplify administrative duties, and assist in assessing medical images and device automation. By individualizing treatments and forecasting adverse reactions, these AI applications have notably enhanced the management of digestive diseases. Integrating AI technologies empowers patients to make knowledgeable choices by evaluating customized treatment alternatives.

Methods: The study's objective was to evaluate the precision of two widely used Chat Bots, Chat GPT and Google BARD, in responding to queries related to medical management. Each bot was tasked with answering a series of questions, and the responses were scored on a 1-10 Likert scale, with 1 denoting low and 10 being high accuracy. To uphold impartial judgment, two unbiased assessors reviewed each bot's replies. The purpose of this research was to illuminate the competencies of these chatbots by systematically analyzing their performance in aspects of accuracy and dependability. Employing two evaluators and utilizing the Likert scale approach helped mitigate potential biases, corroborating the findings.

Results: Our study compared the proficiency of Chat GPT and Google BARD in medical management. Chat GPT demonstrated superior performance, achieving 71% accuracy as opposed to Google BARD’s 39% (p = 0.032) and 58% versus 29% in the reliability of medical information (p = 0.021). This research underscores the importance of accuracy and reliability in creating IBS chatbots. While Chat GPT showed commendable results, signifying its promise as a trustworthy resource, it also highlighted the need for ongoing research and advancement in this domain.

Discussion: AI has significantly reshaped healthcare, particularly aiding in diagnosing and managing IBD. ChatGPT and Google Bard are key AI tools that analyze medical images, assess samples, and optimize tasks. Our research evaluated the precision and dependability of these chatbots in medical management. ChatGPT surpassed Google Bard, securing 71% accuracy and 58% in reliability of information. While ChatGPT showcases substantial potential as a reliable AI instrument in IBD management, the results also accentuate the continual necessity for exploration and advancement in this area



Disclosures:


Raj Mohan Ram Mohan, MD, Sai Reshma Magam, MD, Sai Greeshma Magam, MD, Melvin Joy, MD, Dilman Natt, MD, Winghang Lau, MD, Abhishek Tadikonda, MD, Leeza Pannikodu, MD, Jiten Desai, MD, Rucha Jiyani, MD, Saher Sheikh, MD, Krishnaiyer Subramani, MD, Paul Mustacchia, MD, MBA. P0867 - The Future of IBD Management: Insights From AI Innovations With ChatGPT and Google Bard, ACG 2024 Annual Scientific Meeting Abstracts. Philadelphia, PA: American College of Gastroenterology.