Advanced search

Search results      


Analyzing ChatGPT’s Proficiency In Multiple Choice Osteopathic Manipulative Medicine Questions: Insights and Implications

Journal: Journal of Osteopathic Medicine Date: 2023/12, 123(12):Pages: A113-A114. doi: Subito , type of study: descriptive study

Full text    (https://www.degruyter.com/document/doi/10.1515/jom-2023-2000/html)

Keywords:

AI [1511]
artifical intelligence [2]
ChatGPT [1]
descriptive study [37]
OMT [2951]
osteopathic manipulative treatment [2973]

Abstract:

Statement of Significance: Chat Generative Pretrained Transformer (ChatGPT) is a new language model developed to learn and converse interactively with users. Current medical research using ChatGPT has been done through the allopathic lens. Our research project shifts this viewpoint to focus on the Osteopathic holistic beliefs by asking ChatGPT questions about Osteopathic Manipulative Medicine (OMM). This research project will add to ChatGPT’s knowledge base and provide physicians with more insight into ChatGPTs abilities. To assess ChatGPT abilities in multiple-choice osteopathic manipulative medicine questions. Research Methods: The ChatGPT-3 model was utilized for this study. ChatGPT was asked 5 multiple-choice OMM questions in each of the following regions: Cervical, Thoracic, Cranial, Sacral/Innominate, Upper extremities, and lower extremities. The questions for these body regions included questions regarding diagnosis, treatment positioning, application of treatment, and more. Questions were marked either right or wrong and an explanation was provided to ChatGPT regarding the answer to each question afterward. These OMM questions were written and provided by the Principal investigators of this study, who serve as part of the OMM teaching faculty at Rowan Virtua School of Osteopathic Medicine. Due to the nature of the study, the study was exempt from IRB approval. Data Analysis: OMM questions were provided in multiple-choice format to ChatGPT in the following regions: Cranial, Cervical, Thoracic, Sacral/Innominate/Pelvis, Lower Extremity, and Upper Extremity. In questioning, ChatGPT got 60% correct in the Cranial OMM section, 0% in the Cervical OMM section, 0% in the Thoracic OMM section, 40% in the Sacral/Innominate/Pelvis OMM section, 40% in the Lower Extremity OMM section, and 0% in the Upper Extremity OMM section. Conclusion: In conclusion, while the initial results indicate a limited understanding of OMM by ChatGPT, this research project represents a valuable step toward expanding the system’s knowledge in this area. As AI continues to advance, it has the potential to complement healthcare practices and provide valuable insights. By focusing on OMM and addressing the unique aspects of osteopathic medicine, AI language models such as ChatGPT can contribute to the integration of holistic beliefs and practices into clinical decision-making processes, benefiting both patients and healthcare providers.


Search results      

 
 
 






  • ImpressumLegal noticeDatenschutz


ostlib.de/data_akxgrzcdtuvqfehyjsnp



Supported by

OSTLIB recommends