Unveiling ChatGPT's Potential across Biomedicine and Patient Care: From Text to Treatment

Abstract

Nowadays, cutting-edge Artificial intelligence systems demonstrate exceptional capabilities in comprehending, generating, and manipulate human language, ushering in a new era of understanding, collaboration, and innovation. This systematic review explores the varied applications of ChatGPT within the realm of biomedicine and patient care, aiming to uncover its potential in transforming textual information into actionable insights for improved healthcare outcomes. To conduct this study, we followed PRISMA guidelines that specifically pertain to systematic reviews by targeting two dataset sources “Google Scholar” and “PubMed” and by using the keywords “Chatgpt, Biomedicine, and Health care”. According to our literature surveys conducted from January 2021 to January 2024, we introduce ChatGPT, as a powerful language model, implicated in various tasks, such as interpreting medical literature, extracting meaningful patterns from vast datasets, and enhancing communication between healthcare professionals. It also investigates the model's ability to assist in clinical decision-making, generating hypotheses, and facilitating personalized treatment plans. However, the integration of ChatGPT into healthcare frameworks may bridge ethical concerns, and potential obstacles such as data privacy, bias, hallucination, and interpretability, emphasizing the need for responsible implementation to ensure patient safety and data security.