SNMMI: Can ChatGPT start an IV line? | AuntMinnie
Tyler Bradshaw, PhD, a medical physicist from the University of Wisconsin-Madison, discussed the emergence of AI algorithms and large language models (LLMs) in nuclear medicine at the Society of Nuclear Medicine and Molecular Imaging annual meeting in a recent interview with AuntMinnie.com. Bradshaw highlighted that AI applications in radiology are also evident in nuclear medicine, covering tasks like image enhancement, disease detection, pathology segmentation for quantification, and workflow improvement.
According to Bradshaw, the presentation at the meeting showcased the maturation of these tools, indicating a significant advancement in the field. While ChatGPT may not be capable of starting an IV line, Bradshaw used this question to spark discussions among nuclear medicine technologists on LLMs' concepts and history during a session on June 8. He foresees the development of "chatbots" driven by LLMs that clinicians in nuclear medicine can engage with to optimize protocols and navigate complex electronic medical records.
Looking ahead, Bradshaw anticipates that LLMs will evolve into truly "multimodal models" in the coming years. This evolution is not limited to imaging modalities but rather emphasizes the models' ability to process text, images, and audio. He envisions this as a significant technological advancement that will revolutionize clinical practices and introduce capabilities beyond current imagination into the field of nuclear medicine.
