From the Department of Medicine (M.F.R., R.A., V.B.K.), Boston University Chobanian & Avedisian School of Medicine, MA; Department of Radiology and Biomedical Imaging (M.F.R.), University of California, San Francisco; Department of Neurology (L.C.S., R.A.), Boston University Chobanian & Avedisian School of Medicine; Department of Electrical and Computer Engineering (I.C.P.), Division of Systems Engineering, and Department of Biomedical Engineering; Faculty of Computing and Data Sciences (I.C.P., V.B.K.), Boston University; Department of Anatomy and Neurobiology (R.A.); The Framingham Heart Study, Boston University Chobanian & Avedisian School of Medicine; Department of Epidemiology, Boston University School of Public Health; Boston University Alzheimer's Disease Research Center (R.A.); and Department of Computer Science (V.B.K.), Boston University, MA.
Neurology. 2023 Dec 4;101(23):1058-1067. doi: 10.1212/WNL.0000000000207967.
Recent advancements in generative artificial intelligence, particularly using large language models (LLMs), are gaining increased public attention. We provide a perspective on the potential of LLMs to analyze enormous amounts of data from medical records and gain insights on specific topics in neurology. In addition, we explore use cases for LLMs, such as early diagnosis, supporting patient and caregivers, and acting as an assistant for clinicians. We point to the potential ethical and technical challenges raised by LLMs, such as concerns about privacy and data security, potential biases in the data for model training, and the need for careful validation of results. Researchers must consider these challenges and take steps to address them to ensure that their work is conducted in a safe and responsible manner. Despite these challenges, LLMs offer promising opportunities for improving care and treatment of various neurologic disorders
最近，生成式人工智能，尤其是大型语言模型（LLM），引起了公众越来越多的关注。我们提供了一种观点，即 LLM 可以分析来自病历的大量数据，并深入了解神经科的特定主题。此外，我们还探讨了 LLM 的用例，例如早期诊断、支持患者和护理人员，以及充当临床医生的助手。我们指出了 LLM 带来的潜在伦理和技术挑战，例如对隐私和数据安全的担忧、模型训练数据中潜在的偏差，以及对结果进行仔细验证的必要性。研究人员必须考虑这些挑战，并采取措施加以解决，以确保他们的工作以安全和负责任的方式进行。尽管存在这些挑战，但 LLM 为改善各种神经疾病的护理和治疗提供了有希望的机会。
