MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes

Ahmed, Awais; Zeng, Xiaoyang; Xi, Rui; Hou, Mengshu; Shah, Syed Attique

doi:10.1016/j.jksuci.2024.101933

Cited by 2 publications

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Mitigating Grand Challenges in Life Cycle Inventory Modeling through the Applications of Large Language Models

Tu,

Guo,

et al. 2024

Environ. Sci. Technol.

View full text Add to dashboard Cite

The accuracy of life cycle assessment (LCA) studies is often questioned due to the two grand challenges of life cycle inventory (LCI) modeling: (1) missing foreground flow data and (2) inconsistency in background data matching. Traditional mechanistic methods (e.g., process simulation) and existing machine learning (ML) methods (e.g., similarity-based selection methods) are inadequate due to their limitations in scalability and generalizability. The large language models (LLMs) are well-positioned to address these challenges, given the massive and diverse knowledge learned through the pretraining step. Incorporating LLMs into LCI modeling can lead to the automation of inventory data curation from diverse data sources and to the implementation of a multimodal analytical capacity. In this article, we delineated the mechanisms and advantages of LLMs to addressing these two grand challenges. We also discussed the future research to enhance the use of LLMs for LCI modeling, which includes the key areas such as improving retrieval augmented generation (RAG), integration with knowledge graphs, developing prompt engineering strategies, and fine-tuning pretrained LLMs for LCI-specific tasks. The findings from our study serve as a foundation for future research on scalable and automated LCI modeling methods that can provide more appropriate data for LCA calculations.

show abstract

Mitigating Grand Challenges in Life Cycle Inventory Modeling through the Applications of Large Language Models

Tu,

Guo,

et al. 2024

Environ. Sci. Technol.

View full text Add to dashboard Cite

show abstract

Toward Digital Twins in the Intensive Care Unit: A Medication Management Case Study

Eslami,

Afshar,

Tootooni

et al. 2024

Preprint

View full text Add to dashboard Cite

Digital twins, computational representations of individuals or systems, offer promising applications in the intensive care unit (ICU) by enhancing decision-making and reducing cognitive load. We developed digital twins using a large language model (LLM), LLaMA-3, fine-tuned with Low-Rank Adapters (LoRA) on physician notes from different ICU specialties in the MIMIC-III dataset. This study hypothesizes that specialty-specific training improves treatment recommendation accuracy compared to training on other ICU specialties. Additionally, we evaluated a zero-shot baseline model, which relied solely on contextual instructions without training. Discharge summaries were analyzed, and medications were masked to create datasets for model training and testing. The medical ICU dataset (1,000 notes) was used for evaluation, and performance was measured using BERTScore and ROUGE-L. LLMs trained on medical ICU notes achieved the highest BERTScore (0.842), outperforming models trained on other specialties or mixed datasets, while untrained zero-shot models showed the lowest performance. These results underscore the value of context-specific training for digital twins, offering foundational insights into LLMs for personalized clinical decision support.

show abstract

MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes

Cited by 2 publications

References 27 publications

Mitigating Grand Challenges in Life Cycle Inventory Modeling through the Applications of Large Language Models

Mitigating Grand Challenges in Life Cycle Inventory Modeling through the Applications of Large Language Models

Toward Digital Twins in the Intensive Care Unit: A Medication Management Case Study

Contact Info

Product

Resources

About