MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes
Existing AI-based medicine prediction systems require substantial training time, computing resources, and extensive labeled data, yet they often lack scalability. To bridge these gaps, this study introduces a novel MED-Prompt framework that employs pretrained models such as BERT, BioBERT, and Clinic...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2024-02-01
|
Series: | Journal of King Saud University: Computer and Information Sciences |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S1319157824000223 |
_version_ | 1797272534410854400 |
---|---|
author | Awais Ahmed Xiaoyang Zeng Rui Xi Mengshu Hou Syed Attique Shah |
author_facet | Awais Ahmed Xiaoyang Zeng Rui Xi Mengshu Hou Syed Attique Shah |
author_sort | Awais Ahmed |
collection | DOAJ |
description | Existing AI-based medicine prediction systems require substantial training time, computing resources, and extensive labeled data, yet they often lack scalability. To bridge these gaps, this study introduces a novel MED-Prompt framework that employs pretrained models such as BERT, BioBERT, and ClinicalBERT. The core of our framework lies in developing specialized prompts, which act as guiding instructions for the models during the prediction process. MED-Prompt develops prompts that help models interpret and extract medical information from clinical corpus. The clinical text was derived from the widely known MIMIC-III 11 https://physionet.org/content/mimiciii/1.4/. dataset. The study performs a comparative analysis and evaluates the performance of Manual-Prompt and GPT-Prompts. Further, a fine-tuned approach is developed within MED-Prompt, leveraging transfer learning to achieve prompt-guided medicine predictions. The proposed method achieved a maximum F1-score of 96.8%, which is more than 40% F1-score higher than the pretrained model. In addition, the fine-tuned also showed an average of 2.38 times better processing performance. These results revealed that MED-Prompt is scalable regarding the number of training records and input prompts. These results not only demonstrate the proficiency and effectiveness of the framework but also significantly reduce computational requirements. This also indicates that the proposed approach has the potential to significantly improve patient care, reduce resource requirements, and increase the overall effectiveness of AI-driven medical prediction systems. |
first_indexed | 2024-03-07T14:29:46Z |
format | Article |
id | doaj.art-1c113d63eae34af0ba42f574c0a595a0 |
institution | Directory Open Access Journal |
issn | 1319-1578 |
language | English |
last_indexed | 2024-03-07T14:29:46Z |
publishDate | 2024-02-01 |
publisher | Elsevier |
record_format | Article |
series | Journal of King Saud University: Computer and Information Sciences |
spelling | doaj.art-1c113d63eae34af0ba42f574c0a595a02024-03-06T05:25:39ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782024-02-01362101933MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notesAwais Ahmed0Xiaoyang Zeng1Rui Xi2Mengshu Hou3Syed Attique Shah4School of Computer Science and Engineering, University of Electronic Science and Technology of China - UESTC, Sichuan, 611731, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China - UESTC, Sichuan, 611731, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China - UESTC, Sichuan, 611731, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China - UESTC, Sichuan, 611731, China; School of Big Data and Artificial Intelligence, Chengdu Technological University, Sichuan, 611730, China; Corresponding author at: School of Computer Science and Engineering, University of Electronic Science and Technology of China - UESTC, Sichuan, 611731, China.School of Computing and Digital Technology, Birmingham City University, STEAMhouse, B4 7RQ, Birmingham, United KingdomExisting AI-based medicine prediction systems require substantial training time, computing resources, and extensive labeled data, yet they often lack scalability. To bridge these gaps, this study introduces a novel MED-Prompt framework that employs pretrained models such as BERT, BioBERT, and ClinicalBERT. The core of our framework lies in developing specialized prompts, which act as guiding instructions for the models during the prediction process. MED-Prompt develops prompts that help models interpret and extract medical information from clinical corpus. The clinical text was derived from the widely known MIMIC-III 11 https://physionet.org/content/mimiciii/1.4/. dataset. The study performs a comparative analysis and evaluates the performance of Manual-Prompt and GPT-Prompts. Further, a fine-tuned approach is developed within MED-Prompt, leveraging transfer learning to achieve prompt-guided medicine predictions. The proposed method achieved a maximum F1-score of 96.8%, which is more than 40% F1-score higher than the pretrained model. In addition, the fine-tuned also showed an average of 2.38 times better processing performance. These results revealed that MED-Prompt is scalable regarding the number of training records and input prompts. These results not only demonstrate the proficiency and effectiveness of the framework but also significantly reduce computational requirements. This also indicates that the proposed approach has the potential to significantly improve patient care, reduce resource requirements, and increase the overall effectiveness of AI-driven medical prediction systems.http://www.sciencedirect.com/science/article/pii/S1319157824000223AI-enabled healthcare decisionsMedical promptsPretrained modelsFree-text clinical notesAnd natural language processing |
spellingShingle | Awais Ahmed Xiaoyang Zeng Rui Xi Mengshu Hou Syed Attique Shah MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes Journal of King Saud University: Computer and Information Sciences AI-enabled healthcare decisions Medical prompts Pretrained models Free-text clinical notes And natural language processing |
title | MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes |
title_full | MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes |
title_fullStr | MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes |
title_full_unstemmed | MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes |
title_short | MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes |
title_sort | med prompt a novel prompt engineering framework for medicine prediction on free text clinical notes |
topic | AI-enabled healthcare decisions Medical prompts Pretrained models Free-text clinical notes And natural language processing |
url | http://www.sciencedirect.com/science/article/pii/S1319157824000223 |
work_keys_str_mv | AT awaisahmed medpromptanovelpromptengineeringframeworkformedicinepredictiononfreetextclinicalnotes AT xiaoyangzeng medpromptanovelpromptengineeringframeworkformedicinepredictiononfreetextclinicalnotes AT ruixi medpromptanovelpromptengineeringframeworkformedicinepredictiononfreetextclinicalnotes AT mengshuhou medpromptanovelpromptengineeringframeworkformedicinepredictiononfreetextclinicalnotes AT syedattiqueshah medpromptanovelpromptengineeringframeworkformedicinepredictiononfreetextclinicalnotes |