Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis

Background: Machine learning (ML) shows promise for sexually transmitted infection (STI) risk prediction, but systematic evidence of its effectiveness remains fragmented. Methods: We systematically searched six electronic databases, three preprint archives and conference proceedings (January 2010-Ap...

Full description

Saved in:
Bibliographic Details
Main Authors: Phyu M. Latt, Nyi N. Soe, Christopher K. Fairley, Eric P.  F. Chow, Cheryl C. Johnson, Purvi Shah, Ismail Maatouk, Lei Zhang, Jason J. Ong
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:International Journal of Infectious Diseases
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1201971225001456
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849687173678235648
author Phyu M. Latt
Nyi N. Soe
Christopher K. Fairley
Eric P.  F. Chow
Cheryl C. Johnson
Purvi Shah
Ismail Maatouk
Lei Zhang
Jason J. Ong
author_facet Phyu M. Latt
Nyi N. Soe
Christopher K. Fairley
Eric P.  F. Chow
Cheryl C. Johnson
Purvi Shah
Ismail Maatouk
Lei Zhang
Jason J. Ong
author_sort Phyu M. Latt
collection DOAJ
description Background: Machine learning (ML) shows promise for sexually transmitted infection (STI) risk prediction, but systematic evidence of its effectiveness remains fragmented. Methods: We systematically searched six electronic databases, three preprint archives and conference proceedings (January 2010-April 2024). Studies reporting quantitative performance metrics for supervised ML-based STI risk prediction models were included. We used a bivariate random-effects model to estimate pooled sensitivity, specificity and area under the curve (AUC). The risk of bias was assessed using the Prediction model Risk of Bias Assessment Tool. We conducted sequential analyses of studies with complete and reconstructed confusion matrices. Subgroup analyses and meta-regression explored potential sources of heterogeneity. Results: Among 3877 records screened, 25 studies comprising 45 unique models met inclusion criteria. For HIV, analysis of studies with complete confusion matrices (7 studies, 9 contingency tables) demonstrated summary AUC of 0.91 (95% CI: 0.88-0.93), pooled sensitivity 0.84 (0.76-0.90) and specificity 0.84 (0.70-0.93). Substantial heterogeneity persisted across subgroups (I² > 98%). For other STIs, individual studies reported AUCs ranging from 0.75-0.87 for syphilis (n = 5), 0.73-1.00 for gonorrhoea (n = 6) and 0.67-1.00 for chlamydia (n = 6). Discussion: While ML models show promising performance, particularly for HIV, significant heterogeneity complicates interpretation. Future research should prioritize external validation, standardized guidelines and multi-centred robust implementation studies to evaluate clinical impact.
format Article
id doaj-art-2d9a36c897794a1c901fea438fa79c19
institution DOAJ
issn 1201-9712
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series International Journal of Infectious Diseases
spelling doaj-art-2d9a36c897794a1c901fea438fa79c192025-08-20T03:22:23ZengElsevierInternational Journal of Infectious Diseases1201-97122025-08-0115710792210.1016/j.ijid.2025.107922Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysisPhyu M. Latt0Nyi N. Soe1Christopher K. Fairley2Eric P.  F. Chow3Cheryl C. Johnson4Purvi Shah5Ismail Maatouk6Lei Zhang7Jason J. Ong8Artificial Intelligence and Modelling in Epidemiology Program, Melbourne Sexual Health Centre, Alfred Health, Melbourne, Australia; School of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; Corresponding author: Phyu M. Latt, Artificial Intelligence and Modelling in Epidemiology Program, Melbourne Sexual Health Centre, Alfred Health, 580 Swanston St, Carlton VIC 3053, Melbourne, Australia.Artificial Intelligence and Modelling in Epidemiology Program, Melbourne Sexual Health Centre, Alfred Health, Melbourne, Australia; School of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, AustraliaSchool of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; Melbourne Sexual Health Centre, Alfred Health, Melbourne, AustraliaSchool of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; Melbourne Sexual Health Centre, Alfred Health, Melbourne, Australia; Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, AustraliaGlobal HIV, Hepatitis and STIs Programmes, World Health Organization, Geneva, SwitzerlandGlobal HIV, Hepatitis and STIs Programmes, World Health Organization, Geneva, Switzerland; Regional Support Team, Asia Pacific, UNAIDS, Bangkok, ThailandGlobal HIV, Hepatitis and STIs Programmes, World Health Organization, Geneva, SwitzerlandArtificial Intelligence and Modelling in Epidemiology Program, Melbourne Sexual Health Centre, Alfred Health, Melbourne, Australia; School of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; Clinical Medical Research Centre, Children’s Hospital of Nanjing Medical University, Nanjing, Jiangsu, ChinaSchool of Translational Medicine, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia; Melbourne Sexual Health Centre, Alfred Health, Melbourne, Australia; Department of Clinical Research, London School of Hygiene and Tropical Medicine, London, UKBackground: Machine learning (ML) shows promise for sexually transmitted infection (STI) risk prediction, but systematic evidence of its effectiveness remains fragmented. Methods: We systematically searched six electronic databases, three preprint archives and conference proceedings (January 2010-April 2024). Studies reporting quantitative performance metrics for supervised ML-based STI risk prediction models were included. We used a bivariate random-effects model to estimate pooled sensitivity, specificity and area under the curve (AUC). The risk of bias was assessed using the Prediction model Risk of Bias Assessment Tool. We conducted sequential analyses of studies with complete and reconstructed confusion matrices. Subgroup analyses and meta-regression explored potential sources of heterogeneity. Results: Among 3877 records screened, 25 studies comprising 45 unique models met inclusion criteria. For HIV, analysis of studies with complete confusion matrices (7 studies, 9 contingency tables) demonstrated summary AUC of 0.91 (95% CI: 0.88-0.93), pooled sensitivity 0.84 (0.76-0.90) and specificity 0.84 (0.70-0.93). Substantial heterogeneity persisted across subgroups (I² > 98%). For other STIs, individual studies reported AUCs ranging from 0.75-0.87 for syphilis (n = 5), 0.73-1.00 for gonorrhoea (n = 6) and 0.67-1.00 for chlamydia (n = 6). Discussion: While ML models show promising performance, particularly for HIV, significant heterogeneity complicates interpretation. Future research should prioritize external validation, standardized guidelines and multi-centred robust implementation studies to evaluate clinical impact.http://www.sciencedirect.com/science/article/pii/S1201971225001456Machine learningHIVSexually transmitted infectionsRisk assessmentSystematic reviewMeta-analysis
spellingShingle Phyu M. Latt
Nyi N. Soe
Christopher K. Fairley
Eric P.  F. Chow
Cheryl C. Johnson
Purvi Shah
Ismail Maatouk
Lei Zhang
Jason J. Ong
Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
International Journal of Infectious Diseases
Machine learning
HIV
Sexually transmitted infections
Risk assessment
Systematic review
Meta-analysis
title Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
title_full Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
title_fullStr Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
title_full_unstemmed Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
title_short Machine learning for personalized risk assessment of HIV, syphilis, gonorrhoea and chlamydia: A systematic review and meta-analysis
title_sort machine learning for personalized risk assessment of hiv syphilis gonorrhoea and chlamydia a systematic review and meta analysis
topic Machine learning
HIV
Sexually transmitted infections
Risk assessment
Systematic review
Meta-analysis
url http://www.sciencedirect.com/science/article/pii/S1201971225001456
work_keys_str_mv AT phyumlatt machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT nyinsoe machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT christopherkfairley machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT ericpfchow machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT cherylcjohnson machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT purvishah machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT ismailmaatouk machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT leizhang machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis
AT jasonjong machinelearningforpersonalizedriskassessmentofhivsyphilisgonorrhoeaandchlamydiaasystematicreviewandmetaanalysis