A breakthrough study has emerged in the medical community recently, showing thatArtificial Intelligence ModelChatGPTexistMedicalThe researchers, from Uppsala University Hospital, Uppsala University, Den Dred Hospital and University Hospital Basel in Switzerland, wrote the report in the journal Acta Orthopaedica.
Source Note: The image is generated by AI, and the image is authorized by Midjourney
The medical community has long debated how to make healthcare more efficient. Researchers at Uppsala University Hospital and Uppsala University say that with advances in generative artificial intelligence and language modeling, there is now an opportunity to reduce the administrative burden on medical professionals, freeing doctors up to spend more time with patients.
The aim of the study was to evaluate the quality and effectiveness of the ChatGPT tool in generating medical records. The researchers used six virtual cases that mimicked real cases in both structure and content. The discharge document for each case was generated by an orthopedic surgeon, and ChatGPT-4 was then asked to generate the same record. The quality assessment was performed by a panel of 15 experts who were unaware of the origin of the documents. As a secondary measure, the time required to create the document was compared.
The results show that the transcripts generated by ChatGPT-4 are comparable to those generated by humans in terms of overall quality, but ChatGPT-4 is 10 times faster than doctors. This result shows that advanced large language models like ChatGPT-4 have the potential to change the way we work in medical administrative tasks.
In the future, the researchers plan to launch an in-depth study to collect 1,000 medical case records. Again using ChatGPT to generate similar administrative records. This will be an interesting and resource-intensive project involving multiple partners. They are already actively working to meet all data management and confidentiality requirements in order to conduct this study.