Facilitating feature selection and extraction in clinical trials with large language models

Jiaji Guo; Wen Sun; Shiting Wen; Di Wu; Yipeng Zhou

doi:10.1007/978-981-96-0840-9_16

Facilitating feature selection and extraction in clinical trials with large language models

Jiaji Guo, Wen Sun, Shiting Wen^*, Di Wu, Yipeng Zhou

^*Corresponding author for this work

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

Abstract

Research on clinical trials requires substantial background and technical knowledge. Large language models (LLMs) have already made a significant impact in various fields. We attempted to use general-purpose LLMs to assist newcomers in clinical trials, enabling them to quickly begin their work. In our work, we demonstrated that in the domain of clinical trials, the current cutting-edge LLMs can provide excellent recommendations for feature selection. By utilizing the features suggested by LLMs, we achieved a 2.5% improvement in AUC compared to complex neural network models when using simpler algorithms. We have also demonstrated that by adjusting the prompts, LLM can play a significant role in the feature extraction process. By adjusting the prompts for certain features suggested by LLM, LLM-assisted feature extraction achieved 100% accuracy in a random sample covering approximately 10% of the entire dataset.

Original language	English
Title of host publication	Advanced Data Mining and Applications
Subtitle of host publication	20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV
Editors	Quan Z. Sheng, Gill Dobbie, Jing Jiang, Xuyun Zhang, Wei Emma Zhang, Yannis Manolopoulos, Jia Wu, Wathiq Mansoor, Congbo Ma
Place of Publication	Singapore
Publisher	Springer, Springer Nature
Pages	230-240
Number of pages	11
ISBN (Electronic)	9789819608409
ISBN (Print)	9789819608393
DOIs	https://doi.org/10.1007/978-981-96-0840-9_16
Publication status	Published - 2025
Event	20th International Conference on Advanced Data Mining Applications, ADMA 2024 - Sydney, Australia Duration: 3 Dec 2024 → 5 Dec 2024

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	15390
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	20th International Conference on Advanced Data Mining Applications, ADMA 2024
Country/Territory	Australia
City	Sydney
Period	3/12/24 → 5/12/24

Keywords

Clinical Trials
Application of LLMs
Healthcare

Access to Document

10.1007/978-981-96-0840-9_16

Cite this

Guo, J., Sun, W., Wen, S., Wu, D., & Zhou, Y. (2025). Facilitating feature selection and extraction in clinical trials with large language models. In Q. Z. Sheng, G. Dobbie, J. Jiang, X. Zhang, W. E. Zhang, Y. Manolopoulos, J. Wu, W. Mansoor, & C. Ma (Eds.), Advanced Data Mining and Applications: 20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV (pp. 230-240). (Lecture Notes in Computer Science; Vol. 15390). Springer, Springer Nature. https://doi.org/10.1007/978-981-96-0840-9_16

Guo, Jiaji ; Sun, Wen ; Wen, Shiting et al. / Facilitating feature selection and extraction in clinical trials with large language models. Advanced Data Mining and Applications: 20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV. editor / Quan Z. Sheng ; Gill Dobbie ; Jing Jiang ; Xuyun Zhang ; Wei Emma Zhang ; Yannis Manolopoulos ; Jia Wu ; Wathiq Mansoor ; Congbo Ma. Singapore : Springer, Springer Nature, 2025. pp. 230-240 (Lecture Notes in Computer Science).

@inproceedings{7d47ae93cbdf429584924d41a8b5d842,

title = "Facilitating feature selection and extraction in clinical trials with large language models",

abstract = "Research on clinical trials requires substantial background and technical knowledge. Large language models (LLMs) have already made a significant impact in various fields. We attempted to use general-purpose LLMs to assist newcomers in clinical trials, enabling them to quickly begin their work. In our work, we demonstrated that in the domain of clinical trials, the current cutting-edge LLMs can provide excellent recommendations for feature selection. By utilizing the features suggested by LLMs, we achieved a 2.5% improvement in AUC compared to complex neural network models when using simpler algorithms. We have also demonstrated that by adjusting the prompts, LLM can play a significant role in the feature extraction process. By adjusting the prompts for certain features suggested by LLM, LLM-assisted feature extraction achieved 100% accuracy in a random sample covering approximately 10% of the entire dataset.",

keywords = "Clinical Trials, Application of LLMs, Healthcare",

author = "Jiaji Guo and Wen Sun and Shiting Wen and Di Wu and Yipeng Zhou",

year = "2025",

doi = "10.1007/978-981-96-0840-9_16",

language = "English",

isbn = "9789819608393",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Springer Nature",

pages = "230--240",

editor = "Sheng, {Quan Z.} and Gill Dobbie and Jing Jiang and Xuyun Zhang and Zhang, {Wei Emma} and Yannis Manolopoulos and Jia Wu and Wathiq Mansoor and Congbo Ma",

booktitle = "Advanced Data Mining and Applications",

address = "United States",

note = "20th International Conference on Advanced Data Mining Applications, ADMA 2024 ; Conference date: 03-12-2024 Through 05-12-2024",

}

Guo, J, Sun, W, Wen, S, Wu, D & Zhou, Y 2025, Facilitating feature selection and extraction in clinical trials with large language models. in QZ Sheng, G Dobbie, J Jiang, X Zhang, WE Zhang, Y Manolopoulos, J Wu, W Mansoor & C Ma (eds), Advanced Data Mining and Applications: 20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV. Lecture Notes in Computer Science, vol. 15390, Springer, Springer Nature, Singapore, pp. 230-240, 20th International Conference on Advanced Data Mining Applications, ADMA 2024, Sydney, Australia, 3/12/24. https://doi.org/10.1007/978-981-96-0840-9_16

Facilitating feature selection and extraction in clinical trials with large language models. / Guo, Jiaji; Sun, Wen; Wen, Shiting et al.
Advanced Data Mining and Applications: 20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV. ed. / Quan Z. Sheng; Gill Dobbie; Jing Jiang; Xuyun Zhang; Wei Emma Zhang; Yannis Manolopoulos; Jia Wu; Wathiq Mansoor; Congbo Ma. Singapore: Springer, Springer Nature, 2025. p. 230-240 (Lecture Notes in Computer Science; Vol. 15390).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Facilitating feature selection and extraction in clinical trials with large language models

AU - Guo, Jiaji

AU - Sun, Wen

AU - Wen, Shiting

AU - Wu, Di

AU - Zhou, Yipeng

PY - 2025

Y1 - 2025

N2 - Research on clinical trials requires substantial background and technical knowledge. Large language models (LLMs) have already made a significant impact in various fields. We attempted to use general-purpose LLMs to assist newcomers in clinical trials, enabling them to quickly begin their work. In our work, we demonstrated that in the domain of clinical trials, the current cutting-edge LLMs can provide excellent recommendations for feature selection. By utilizing the features suggested by LLMs, we achieved a 2.5% improvement in AUC compared to complex neural network models when using simpler algorithms. We have also demonstrated that by adjusting the prompts, LLM can play a significant role in the feature extraction process. By adjusting the prompts for certain features suggested by LLM, LLM-assisted feature extraction achieved 100% accuracy in a random sample covering approximately 10% of the entire dataset.

AB - Research on clinical trials requires substantial background and technical knowledge. Large language models (LLMs) have already made a significant impact in various fields. We attempted to use general-purpose LLMs to assist newcomers in clinical trials, enabling them to quickly begin their work. In our work, we demonstrated that in the domain of clinical trials, the current cutting-edge LLMs can provide excellent recommendations for feature selection. By utilizing the features suggested by LLMs, we achieved a 2.5% improvement in AUC compared to complex neural network models when using simpler algorithms. We have also demonstrated that by adjusting the prompts, LLM can play a significant role in the feature extraction process. By adjusting the prompts for certain features suggested by LLM, LLM-assisted feature extraction achieved 100% accuracy in a random sample covering approximately 10% of the entire dataset.

KW - Clinical Trials

KW - Application of LLMs

KW - Healthcare

UR - http://www.scopus.com/inward/record.url?scp=85213296398&partnerID=8YFLogxK

U2 - 10.1007/978-981-96-0840-9_16

DO - 10.1007/978-981-96-0840-9_16

M3 - Conference proceeding contribution

AN - SCOPUS:85213296398

SN - 9789819608393

T3 - Lecture Notes in Computer Science

SP - 230

EP - 240

BT - Advanced Data Mining and Applications

A2 - Sheng, Quan Z.

A2 - Dobbie, Gill

A2 - Jiang, Jing

A2 - Zhang, Xuyun

A2 - Zhang, Wei Emma

A2 - Manolopoulos, Yannis

A2 - Wu, Jia

A2 - Mansoor, Wathiq

A2 - Ma, Congbo

PB - Springer, Springer Nature

CY - Singapore

T2 - 20th International Conference on Advanced Data Mining Applications, ADMA 2024

Y2 - 3 December 2024 through 5 December 2024

ER -

Guo J, Sun W, Wen S, Wu D, Zhou Y. Facilitating feature selection and extraction in clinical trials with large language models. In Sheng QZ, Dobbie G, Jiang J, Zhang X, Zhang WE, Manolopoulos Y, Wu J, Mansoor W, Ma C, editors, Advanced Data Mining and Applications: 20th International Conference, ADMA 2024, Sydney, NSW, Australia, December 3–5, 2024, proceedings, part IV. Singapore: Springer, Springer Nature. 2025. p. 230-240. (Lecture Notes in Computer Science). doi: 10.1007/978-981-96-0840-9_16

Facilitating feature selection and extraction in clinical trials with large language models

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this