BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records

Alvaro D. Orjuela-Cañón

doi:10.1109/LA-CCI66231.2025.11270453

BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records

Alvaro D. Orjuela-Cañón

School of Medicine and Health Sciences

Research output: Chapter in Book/Inform › Conference contribution

Abstract

Tuberculosis (TB) is an infectious disease and remains one of the leading causes of death worldwide. Its diagnosis poses significant challenges, often depending on the resources and capacities available within healthcare institutions. The analysis of data from various sources has enabled the development of models based on artificial and computational intelligence. In this study, Bidirectional Encoder Representations from Transformers (BERT) models were applied to classify electronic medical reports for TB detection. Their performance was compared with that of traditional natural language processing (NLP) approaches to evaluate improvements in classification accuracy. Results indicate that BERT is a promising strategy for this specific task, achieving an area under the ROC curve (AUC) of 71.81% in the best scenario, which depends on computational resources and data availability.

Original language	English (US)
Title of host publication	2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings
Editors	Ilse Cervantes, Amadeo Arguelles
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331554323
DOIs	https://doi.org/10.1109/LA-CCI66231.2025.11270453
State	Published - 2025
Event	2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Mexico City, Mexico Duration: Nov 5 2025 → Nov 7 2025

Publication series

Name	2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings

Conference

Conference	2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025
Country/Territory	Mexico
City	Mexico City
Period	11/5/25 → 11/7/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Computer Science Applications
Computer Vision and Pattern Recognition
Safety, Risk, Reliability and Quality

Access to Document

10.1109/LA-CCI66231.2025.11270453

Cite this

Orjuela-Cañón, A. D. (2025). BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records. In I. Cervantes, & A. Arguelles (Eds.), 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings (2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LA-CCI66231.2025.11270453

Orjuela-Cañón, Alvaro D. / BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records. 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings. editor / Ilse Cervantes ; Amadeo Arguelles. Institute of Electrical and Electronics Engineers Inc., 2025. (2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings).

@inproceedings{d7ec2a8376014cb794e3a0e673a3a458,

title = "BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records",

abstract = "Tuberculosis (TB) is an infectious disease and remains one of the leading causes of death worldwide. Its diagnosis poses significant challenges, often depending on the resources and capacities available within healthcare institutions. The analysis of data from various sources has enabled the development of models based on artificial and computational intelligence. In this study, Bidirectional Encoder Representations from Transformers (BERT) models were applied to classify electronic medical reports for TB detection. Their performance was compared with that of traditional natural language processing (NLP) approaches to evaluate improvements in classification accuracy. Results indicate that BERT is a promising strategy for this specific task, achieving an area under the ROC curve (AUC) of 71.81\% in the best scenario, which depends on computational resources and data availability.",

author = "Orjuela-Ca{\~n}{\'o}n, \{Alvaro D.\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 ; Conference date: 05-11-2025 Through 07-11-2025",

year = "2025",

doi = "10.1109/LA-CCI66231.2025.11270453",

language = "English (US)",

series = "2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Ilse Cervantes and Amadeo Arguelles",

booktitle = "2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings",

address = "United States",

}

Orjuela-Cañón, AD 2025, BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records. in I Cervantes & A Arguelles (eds), 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings. 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings, Institute of Electrical and Electronics Engineers Inc., 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025, Mexico City, Mexico, 11/5/25. https://doi.org/10.1109/LA-CCI66231.2025.11270453

BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records. / Orjuela-Cañón, Alvaro D.
2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings. ed. / Ilse Cervantes; Amadeo Arguelles. Institute of Electrical and Electronics Engineers Inc., 2025. (2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings).

Research output: Chapter in Book/Inform › Conference contribution

TY - GEN

T1 - BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records

AU - Orjuela-Cañón, Alvaro D.

PY - 2025

Y1 - 2025

N2 - Tuberculosis (TB) is an infectious disease and remains one of the leading causes of death worldwide. Its diagnosis poses significant challenges, often depending on the resources and capacities available within healthcare institutions. The analysis of data from various sources has enabled the development of models based on artificial and computational intelligence. In this study, Bidirectional Encoder Representations from Transformers (BERT) models were applied to classify electronic medical reports for TB detection. Their performance was compared with that of traditional natural language processing (NLP) approaches to evaluate improvements in classification accuracy. Results indicate that BERT is a promising strategy for this specific task, achieving an area under the ROC curve (AUC) of 71.81% in the best scenario, which depends on computational resources and data availability.

AB - Tuberculosis (TB) is an infectious disease and remains one of the leading causes of death worldwide. Its diagnosis poses significant challenges, often depending on the resources and capacities available within healthcare institutions. The analysis of data from various sources has enabled the development of models based on artificial and computational intelligence. In this study, Bidirectional Encoder Representations from Transformers (BERT) models were applied to classify electronic medical reports for TB detection. Their performance was compared with that of traditional natural language processing (NLP) approaches to evaluate improvements in classification accuracy. Results indicate that BERT is a promising strategy for this specific task, achieving an area under the ROC curve (AUC) of 71.81% in the best scenario, which depends on computational resources and data availability.

UR - https://www.scopus.com/pages/publications/105031379175

UR - https://www.scopus.com/pages/publications/105031379175#tab=citedBy

U2 - 10.1109/LA-CCI66231.2025.11270453

DO - 10.1109/LA-CCI66231.2025.11270453

M3 - Conference contribution

AN - SCOPUS:105031379175

T3 - 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings

BT - 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings

A2 - Cervantes, Ilse

A2 - Arguelles, Amadeo

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025

Y2 - 5 November 2025 through 7 November 2025

ER -

Orjuela-Cañón AD. BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records. In Cervantes I, Arguelles A, editors, 2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc. 2025. (2025 IEEE Latin American Conference on Computational Intelligence, LA-CCI 2025 - Conference Proceedings). doi: 10.1109/LA-CCI66231.2025.11270453

BERT-Based Approach for Tuberculosis Diagnosis from Electronic Medical Records

Abstract

Publication series

Conference

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this