Chat-GPT 4 and the power of generative AI in medicine
In the ever-evolving world of medicine and technology, a recent experiment conducted by physician-researchers at Beth Israel Deaconess Medical Center (BIDMC) has shed light on the potential of generative AI in the field of complex diagnostic reasoning. Published in JAMA, their findings reveal the impressive capabilities of Chat-GPT 4 to diagnose challenging medical cases accurately.
Generative AI, a branch of artificial intelligence that enables computers to generate new content based on patterns and training data, has gained recognition for its applications in various domains such as creative industries, education, and customer service. However, its potential within the clinical setting has remained largely unexplored, particularly in the realm of intricate diagnostic reasoning.
“Recent advances in artificial intelligence have led to generative AI models that are capable of detailed text-based responses that score highly in standardized medical examinations,” said Adam Rodman, co-director of the Innovations in Media and Education Delivery (iMED) Initiative at BIDMC and an instructor in medicine at Harvard Medical School. “We wanted to know if such a generative model could ‘think’ like a doctor, so we asked one to solve standardized complex diagnostic cases used for educational purposes. It did really, really well.”
To evaluate the diagnostic prowess of Chat-GPT 4, Rodman and his team employed clinicopathological case conferences (CPCs), which present intricate patient cases with relevant clinical and laboratory data, imaging studies and histopathological findings for educational purposes. The researchers assessed 70 CPC cases and discovered that the generative AI accurately matched the final CPC diagnosis in 39% of the cases. Furthermore, in 64%, the AI’s list of potential diagnoses were spot on.
Co-author Byron Crowe, an internal medicine physician at BIDMC and an instructor in medicine at Harvard Medical School, emphasizes that this study contributes to a growing body of literature showcasing AI technology’s promising capabilities. Further investigation is required to fully understand how these new AI models can transform healthcare delivery.
“Our study adds to a growing body of literature demonstrating the promising capabilities of AI technology,” he said. “Further investigation will help us better understand how these new AI models might transform health care delivery.”
The implications of this research extend beyond the confines of the medical field. As generative AI chatbots continue to advance, their potential impact on creative industries, education, customer service and other domains becomes increasingly apparent. By harnessing the power of natural language processing, these chatbots can mimic human-like language and improve interactions between technology and users.
Despite a lot of “AI is going to take over the world” talk, the fusion of artificial intelligence and healthcare actually holds immense promise. While there are still hurdles to overcome, such as ensuring privacy and refining the optimal uses of this technology, the findings from the study provide a glimpse into the transformative potential of generative AI in the realms of diagnosis and patient care.
It is important to note chatbots powered by generative AI cannot replace trained medical professionals.
“While chatbots cannot replace the expertise and knowledge of a trained medical professional, generative AI is a promising potential adjunct to human cognition in diagnosis,” said first author Zahir Kanjee and assistant professor of medicine at Harvard Medical School. “It has the potential to help physicians make sense of complex medical data and broaden or refine our diagnostic thinking.”
