Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel

<b>Background:</b> Typhoid fever remains a significant public health challenge, especially in developing countries where diagnostic resources are limited. Accurate and timely diagnosis is crucial for effective treatment and disease containment. Traditional diagnostic methods, while effec...

Full description

Saved in:
Bibliographic Details
Main Authors: Fariha Ahmed Nishat, M. F. Mridha, Istiak Mahmud, Meshal Alfarhood, Mejdl Safran, Dunren Che
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/5/562
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850052709435047936
author Fariha Ahmed Nishat
M. F. Mridha
Istiak Mahmud
Meshal Alfarhood
Mejdl Safran
Dunren Che
author_facet Fariha Ahmed Nishat
M. F. Mridha
Istiak Mahmud
Meshal Alfarhood
Mejdl Safran
Dunren Che
author_sort Fariha Ahmed Nishat
collection DOAJ
description <b>Background:</b> Typhoid fever remains a significant public health challenge, especially in developing countries where diagnostic resources are limited. Accurate and timely diagnosis is crucial for effective treatment and disease containment. Traditional diagnostic methods, while effective, can be time-consuming and resource-intensive. This study aims to develop a lightweight machine learning-based diagnostic tool for the early and efficient detection of typhoid fever using clinical data. <b>Methods:</b> A custom dataset comprising 14 clinical and demographic parameters—including age, gender, headache, muscle pain, nausea, diarrhea, cough, fever range (°F), hemoglobin (g/dL), platelet count, urine culture bacteria, calcium (mg/dL), and potassium (mg/dL)—was analyzed. A machine learning metamodel, integrating Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), and Decision Tree classifiers with a Light Gradient Boosting Machine (LGBM), was trained and evaluated using k-fold cross-validation. Performance was assessed using precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). <b>Results:</b> The proposed metamodel demonstrated superior diagnostic performance, achieving a precision of 99%, recall of 100%, and an AUC of 1.00. It outperformed traditional diagnostic methods and other standalone machine learning algorithms, offering high accuracy and generalizability. <b>Conclusions:</b> The lightweight machine learning metamodel provides a cost-effective, non-invasive, and rapid diagnostic alternative for typhoid fever, particularly suited for resource-limited settings. Its reliance on accessible clinical parameters ensures practical applicability and scalability, potentially improving patient outcomes and aiding in disease control. Future work will focus on broader validation and integration into clinical workflows to further enhance its utility.
format Article
id doaj-art-07ba7f9fdcf94f6e92e5a5a317cd844b
institution DOAJ
issn 2075-4418
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-07ba7f9fdcf94f6e92e5a5a317cd844b2025-08-20T02:52:45ZengMDPI AGDiagnostics2075-44182025-02-0115556210.3390/diagnostics15050562Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning MetamodelFariha Ahmed Nishat0M. F. Mridha1Istiak Mahmud2Meshal Alfarhood3Mejdl Safran4Dunren Che5Dhaka National Medical College, Dhaka 1100, BangladeshDepartment of Computer Science, American International University-Bangladesh, Dhaka 1229, BangladeshDepartment of Electrical and Electronic Engineering, Ahsanullah University of Science and Technology, Dhaka 1208, BangladeshDepartment of Computer Science, College of Computer and Information Sciences, King Saud University, P.O. Box 51178, Riyadh 11543, Saudi ArabiaDepartment of Computer Science, College of Computer and Information Sciences, King Saud University, P.O. Box 51178, Riyadh 11543, Saudi ArabiaDepartment of Electrical Engineering and Computer Science, Texas A & M University-Kingsville, Kingsville, TX 78363, USA<b>Background:</b> Typhoid fever remains a significant public health challenge, especially in developing countries where diagnostic resources are limited. Accurate and timely diagnosis is crucial for effective treatment and disease containment. Traditional diagnostic methods, while effective, can be time-consuming and resource-intensive. This study aims to develop a lightweight machine learning-based diagnostic tool for the early and efficient detection of typhoid fever using clinical data. <b>Methods:</b> A custom dataset comprising 14 clinical and demographic parameters—including age, gender, headache, muscle pain, nausea, diarrhea, cough, fever range (°F), hemoglobin (g/dL), platelet count, urine culture bacteria, calcium (mg/dL), and potassium (mg/dL)—was analyzed. A machine learning metamodel, integrating Support Vector Machine (SVM), Gaussian Naive Bayes (GNB), and Decision Tree classifiers with a Light Gradient Boosting Machine (LGBM), was trained and evaluated using k-fold cross-validation. Performance was assessed using precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). <b>Results:</b> The proposed metamodel demonstrated superior diagnostic performance, achieving a precision of 99%, recall of 100%, and an AUC of 1.00. It outperformed traditional diagnostic methods and other standalone machine learning algorithms, offering high accuracy and generalizability. <b>Conclusions:</b> The lightweight machine learning metamodel provides a cost-effective, non-invasive, and rapid diagnostic alternative for typhoid fever, particularly suited for resource-limited settings. Its reliance on accessible clinical parameters ensures practical applicability and scalability, potentially improving patient outcomes and aiding in disease control. Future work will focus on broader validation and integration into clinical workflows to further enhance its utility.https://www.mdpi.com/2075-4418/15/5/562typhoid fever diagnosismachine learning metamodelclinical data analysisensemble learningnon-invasive diagnosticspredictive modeling
spellingShingle Fariha Ahmed Nishat
M. F. Mridha
Istiak Mahmud
Meshal Alfarhood
Mejdl Safran
Dunren Che
Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
Diagnostics
typhoid fever diagnosis
machine learning metamodel
clinical data analysis
ensemble learning
non-invasive diagnostics
predictive modeling
title Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
title_full Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
title_fullStr Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
title_full_unstemmed Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
title_short Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel
title_sort enhancing typhoid fever diagnosis based on clinical data using a lightweight machine learning metamodel
topic typhoid fever diagnosis
machine learning metamodel
clinical data analysis
ensemble learning
non-invasive diagnostics
predictive modeling
url https://www.mdpi.com/2075-4418/15/5/562
work_keys_str_mv AT farihaahmednishat enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel
AT mfmridha enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel
AT istiakmahmud enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel
AT meshalalfarhood enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel
AT mejdlsafran enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel
AT dunrenche enhancingtyphoidfeverdiagnosisbasedonclinicaldatausingalightweightmachinelearningmetamodel