What question did this study set out to answer?

This study aimed to evaluate how accurately AI models can determine adherence to clinical trial reporting guidelines.

December 15, 2023Open Access

GPT for RCTs?: Using AI to determine adherence to reporting guidelines

Puntos clave

This study aimed to evaluate how accurately AI models can determine adherence to clinical trial reporting guidelines.
Exploratory retrospective data analysis of 113 sports medicine clinical trial reports.
Evaluated GPT-4 and Llama2 AI models for compliance with nine reporting guideline questions.
Performance assessed using F1-score and classification accuracy on a TEST dataset split from the original data.
GPT-4-Turbo achieved an F1-score of 0.89 and accuracy of 90% (95% CI 85-94%).
Llama2's initial F1-score was 0.63, improving to 0.84 after fine-tuning, with an accuracy of 83% (95% CI 77-88%).
GPT-4-Vision accurately identified all participant flow diagrams with 100% accuracy, but its accuracy for identifying missing details was only 57% (95% CI 39-73%).

Resumen

Abstract Background Adherence to established reporting guidelines can improve clinical trial reporting standards, but attempts to improve adherence have produced mixed results. This exploratory study aimed to determine how accurate a Large Language Model generative AI system (AI-LLM) was for determining reporting guideline compliance in a sample of sports medicine clinical trial reports. Design and Methods This study was an exploratory retrospective data analysis. The OpenAI GPT-4 and Meta LLama2 AI-LLMa were evaluated for their ability to determine reporting guideline adherence in a sample of 113 published sports medicine and exercise science clinical trial reports. For each paper, the GPT-4-Turbo and Llama 2 70B models were prompted to answer a series of nine reporting guideline questions about the text of the article. The GPT-4-Vision model was prompted to answer two additional reporting guideline questions about the participant flow diagram in a subset of articles. The dataset was randomly split (80/20) into a TRAIN and TEST dataset. Hyperparameter and fine-tuning were performed using the TRAIN dataset. The Llama2 model was fine-tuned using the data from the GPT-4-Turbo analysis of the TRAIN dataset. Primary outcome measure: Model performance (F1-score, classification accuracy) was assessed using the TEST dataset. Results Across all questions about the article text, the GPT-4-Turbo AI-LLM demonstrated acceptable performance (F1-score = 0.89, accuracy95% CI = 90%85-94%). Accuracy for all reporting guidelines was > 80%. The Llama2 model accuracy was initially poor (F1-score = 0.63, accuracy95%CI = 64%57-71%), and improved with fine-tuning (F1-score = 0.84, accuracy95%CI = 83%77-88%). The GPT-4-Vision model accurately identified all participant flow diagrams (accuracy95% CI = 100%89-100%) but was less accurate at identifying when details were missing from the flow diagram (accuracy95% CI = 57%39-73%). Conclusions Both the GPT-4 and fine-tuned Llama2 AI-LLMs showed promise as tools for assessing reporting guideline compliance. Next steps should include developing an efficent, open-source AI-LLM and exploring methods to improve model accuracy.

Preguntar a la IA

Me gusta

Guardar

Ver artículo completo