Generative Healthcare Intelligence (gHi) System：

The primary objective of the gHi system is to facilitate clinical healthcare professionals in the creation and maintenance of medical records, utilizing advanced Natural Language Processing (NLP) technologies, including Large Language Models (LLM) and Automatic Speech Recognition (ASR).

In the contemporary healthcare landscape, healthcare professionals are tasked with diagnosing, treating, monitoring, and aiding in the recovery of patients' ailments. Concurrently, they are mandated to create and update medical records in compliance with legal stipulations and institutional guidelines. Medical records are not merely crucial for evaluating and documenting patients' health status, but they also form the bedrock for ensuring superior healthcare services and ongoing enhancement. Hence, it is a task of considerable administrative significance.

However, the human capacity for physical and mental exertion is finite, and the time and focus required for medical record creation may detract from patient care. The pursuit of high-quality documentation may compromise the time allocated for patient care, and vice versa. Moreover, given that a majority of medical documents in Taiwan are typically composed in English and necessitate a high degree of professional expertise, they are often difficult to delegate to general administrative assistants. Additionally, inaccuracies in documentation can potentially result in medical errors or legal complications. Consequently, the task of medical record creation has historically imposed a substantial strain on healthcare professionals, potentially leading to extreme fatigue or burnout.

Therefore, our aim is to develop pertinent tools to ameliorate these conditions. In consideration of the clinical documentation needs of healthcare professionals, we posit that the current state of Large Language Models and Artificial Intelligence technologies such as speech recognition can provide substantial assistance in medical record creation. These technologies can aid in structuring and transcribing textual data. Based on user feedback, the gHi system has been shown to reduce the documentation burden in medical record creation by 75%.