A machine learning model for early detection of sexually transmitted infections

Sexually transmitted infections (STIs) are diseases transmitted mostly through unprotected sex with an infected partner. STIs can be transmitted to an infant before or during childbirth. More than one million sexually transmitted infections (STIs) are acquired every day worldwide. In the most recent...

Full description

Saved in:
Bibliographic Details
Main Authors: Juma Shija, Judith Leo, Elizabeth Mkoba
Format: Article
Language:English
Published: International Academy of Ecology and Environmental Sciences 2025-06-01
Series:Computational Ecology and Software
Subjects:
Online Access:http://www.iaees.org/publications/journals/ces/articles/2025-15(2)/machine-learning-model.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849322649946161152
author Juma Shija
Judith Leo
Elizabeth Mkoba
author_facet Juma Shija
Judith Leo
Elizabeth Mkoba
author_sort Juma Shija
collection DOAJ
description Sexually transmitted infections (STIs) are diseases transmitted mostly through unprotected sex with an infected partner. STIs can be transmitted to an infant before or during childbirth. More than one million sexually transmitted infections (STIs) are acquired every day worldwide. In the most recent years, the prevalence of STIs reached approximately 20% among Tanzanian older adults living in metropolitan areas. If not treated properly and on time, STIs can have severe consequences, including infertility, sterility, increased susceptibility to more serious diseases such as the Human Immunodeficiency Virus (HIV), and even death. However, stigma and shame associated with STIs remain significant barriers to proper diagnosis and timely treatment, leading many patients to face increased risks. The purpose of this paper is to present a machine-learning model for early detection of sexually transmitted infections that was developed. The developed model can be deployed into health systems for self-diagnosis to remove communication barriers between sexual health clinics and STI patients. The study used a quantitative research method and got its dataset of 13,335 records from the Government of Tanzania Health Operations Management Information System (GoT-HoMIS) in areas with many STI cases. This was done by using surveys and questionnaires to get the data. The dataset was split into a 70%:15%:15% ratio for training, testing, and validation, respectively, and five machine learning algorithms were evaluated: AdaBoost, Support Vector Machine, Random Forest, Decision Tree, and Stochastic Gradient Descent. Based on evaluation metrics, the AdaBoost model was identified as the best-performing model, achieving an accuracy of 97.45%, an F1 score of 97.7%, and the Receiver Operating Characteristics Area Under the Curve (ROC-AUC) with a higher true positive rate and a lower false positive rate. The study recommends integrating a machine learning model into healthcare systems to detect STIs early, improve medical care, reduce disease progression, and remove stigmatisation barriers. Also, it can provide insights into infection patterns, allowing practitioners to adapt their responses. Machine learning-based solutions in mobile apps and telemedicine systems promote early testing and treatment.
format Article
id doaj-art-f08ca51202e2435d9486561b8a9f9d7c
institution Kabale University
issn 2220-721X
language English
publishDate 2025-06-01
publisher International Academy of Ecology and Environmental Sciences
record_format Article
series Computational Ecology and Software
spelling doaj-art-f08ca51202e2435d9486561b8a9f9d7c2025-08-20T03:49:17ZengInternational Academy of Ecology and Environmental SciencesComputational Ecology and Software2220-721X2025-06-011523044A machine learning model for early detection of sexually transmitted infectionsJuma Shija0Judith Leo1Elizabeth Mkoba2The School of Computational and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, Arusha, TanzaniaThe School of Computational and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, Arusha, TanzaniaThe School of Computational and Communication Science and Engineering, The Nelson Mandela African Institution of Science and Technology, Arusha, TanzaniaSexually transmitted infections (STIs) are diseases transmitted mostly through unprotected sex with an infected partner. STIs can be transmitted to an infant before or during childbirth. More than one million sexually transmitted infections (STIs) are acquired every day worldwide. In the most recent years, the prevalence of STIs reached approximately 20% among Tanzanian older adults living in metropolitan areas. If not treated properly and on time, STIs can have severe consequences, including infertility, sterility, increased susceptibility to more serious diseases such as the Human Immunodeficiency Virus (HIV), and even death. However, stigma and shame associated with STIs remain significant barriers to proper diagnosis and timely treatment, leading many patients to face increased risks. The purpose of this paper is to present a machine-learning model for early detection of sexually transmitted infections that was developed. The developed model can be deployed into health systems for self-diagnosis to remove communication barriers between sexual health clinics and STI patients. The study used a quantitative research method and got its dataset of 13,335 records from the Government of Tanzania Health Operations Management Information System (GoT-HoMIS) in areas with many STI cases. This was done by using surveys and questionnaires to get the data. The dataset was split into a 70%:15%:15% ratio for training, testing, and validation, respectively, and five machine learning algorithms were evaluated: AdaBoost, Support Vector Machine, Random Forest, Decision Tree, and Stochastic Gradient Descent. Based on evaluation metrics, the AdaBoost model was identified as the best-performing model, achieving an accuracy of 97.45%, an F1 score of 97.7%, and the Receiver Operating Characteristics Area Under the Curve (ROC-AUC) with a higher true positive rate and a lower false positive rate. The study recommends integrating a machine learning model into healthcare systems to detect STIs early, improve medical care, reduce disease progression, and remove stigmatisation barriers. Also, it can provide insights into infection patterns, allowing practitioners to adapt their responses. Machine learning-based solutions in mobile apps and telemedicine systems promote early testing and treatment.http://www.iaees.org/publications/journals/ces/articles/2025-15(2)/machine-learning-model.pdfmachine learningsexually transmitted infectionsartificial intelligencestigmatisation
spellingShingle Juma Shija
Judith Leo
Elizabeth Mkoba
A machine learning model for early detection of sexually transmitted infections
Computational Ecology and Software
machine learning
sexually transmitted infections
artificial intelligence
stigmatisation
title A machine learning model for early detection of sexually transmitted infections
title_full A machine learning model for early detection of sexually transmitted infections
title_fullStr A machine learning model for early detection of sexually transmitted infections
title_full_unstemmed A machine learning model for early detection of sexually transmitted infections
title_short A machine learning model for early detection of sexually transmitted infections
title_sort machine learning model for early detection of sexually transmitted infections
topic machine learning
sexually transmitted infections
artificial intelligence
stigmatisation
url http://www.iaees.org/publications/journals/ces/articles/2025-15(2)/machine-learning-model.pdf
work_keys_str_mv AT jumashija amachinelearningmodelforearlydetectionofsexuallytransmittedinfections
AT judithleo amachinelearningmodelforearlydetectionofsexuallytransmittedinfections
AT elizabethmkoba amachinelearningmodelforearlydetectionofsexuallytransmittedinfections
AT jumashija machinelearningmodelforearlydetectionofsexuallytransmittedinfections
AT judithleo machinelearningmodelforearlydetectionofsexuallytransmittedinfections
AT elizabethmkoba machinelearningmodelforearlydetectionofsexuallytransmittedinfections