The Expanding Role of ChatGPT (Chat-Generative Pre-Trained Transformer) in Neurosurgery: A Systematic Review of Literature and Conceptual Framework

Abstract

The objective of this study is to explore the use of ChatGPT (Chat-Generative Pre-Trained Transformer) in neurosurgery and its potential impact on the field. The authors aim to discuss, through a systematic review of current literature, how this rising new artificial intelligence (AI) technology may prove to be a useful tool in the future, weighing its potential benefits and limitations. The authors conducted a comprehensive and systematic literature review of the use of ChatGPT and its applications in healthcare and different neurosurgery topics. Through a systematic review of the literature, with a search strategy using the databases such as PubMed, Google Scholar, and Embase, we analyzed the advantages and limitations of using ChatGPT in neurosurgery and evaluated its potential impact. ChatGPT has demonstrated promising results in various applications, such as natural language processing, language translation, and text summarization. In neurosurgery, ChatGPT can assist in different areas such as surgical planning, image recognition, medical diagnosis, patient care, and scientific production. A total of 128 articles were retrieved from databases, where the final 22 articles were included for thorough analysis. The studies reviewed demonstrate the potential of AI and deep learning (DL), through language models such as ChatGPT, to improve the accuracy and efficiency of neurosurgical procedures, as well as diagnosis, treatment, and patient outcomes across various medical specialties, including neurosurgery. There are, however, limitations to its use, including the need for large datasets and the potential for errors in the output, which most authors concur will need human verification for the final application. Our search demonstrated the potential that ChatGPT holds for the present and future, in accordance with the studies’ authors’ findings herein analyzed and expert opinions. Further research and development are required to fully understand its capabilities and limitations. AI technology can serve as a useful tool to augment human intelligence; however, it is essential to use it in a responsible and ethical manner.

Accuracy of ChatGPT in Neurolocalization

“…AI has a promising future in medical applications, and ChatGPT is the first step in this long journey. While ChatGPT can assist in decision-making and reduce the workload for neurosurgeons [ 27 ], it is not yet equipped to handle real-life cases with complex, intertwining variables where the superior judgment of the human brain is evident. Additionally, an inherent limitation of ChatGPT is its inability to perform physical examinations or process radiological images [ 28 ].…”

Ethical Concerns About ChatGPT in Healthcare: A Useful Tool or the Tombstone of Original and Reflective Thinking?

“…In addition, very recent studies suggest that this chatbot would also have the potential to reduce work hours for clinicians and assist in limiting cases of clinician burnout by contributing to medical bureaucracy writing, such as preauthorization letters to insurance companies, discharge summaries, or other documentation papers [11,14]. Moreover, ChatGPT can offer healthcare professionals the opportunity and the time to focus on the most creative part of medicine, as shown in the systematic review of Roman et al [15], focusing on the applications of ChatGPT in neurosurgery issues and taking into account the potential risks [15].…”

Natural language processing augments comorbidity documentation in neurosurgical inpatient admissions

“…ML and NLP applications in neurosurgery and neuroimaging are numerous and varied [15,[35][36][37][38][39] and have only increased in breadth in depth since the widespread popularization of large language models (LLMs) such as ChatGPT [40][41][42][43][44]. A variety of methods utilizing either radiology reports, raw images, or both, have been successfully applied in a variety of clinical applications [45,46].…”