EUVIMEDObjevte EUVIMED
Systematische ÜbersichtOpen access

Diagnostic Performance of Artificial Intelligence–Based Methods for Tuberculosis Detection: Systematic Review

Diagnostische Leistung KI-basierter Methoden zur Tuberkuloseerkennung: systematische Übersichtsarbeit

Seng Hansun iD, Ahmadreza Argha iD, Ivan Bakhshayeshi iD, Arya Wicaksana iD, Hamid Alinejad-Rokny iD, Greg J Fox iD, Siaw-Teng Liaw iD, Branko G Celler iD, Guy B Marks iD

Journal of Medical Internet Research · 2025-03-07 · 27 · e69068

DOI: 10.2196/69068 ↗

Auf einen Blick

Die Übersichtsarbeit mit 152 Studien zeigt gute diagnostische Leistungen KI-basierter TB-Methoden. Die mittlere Genauigkeit lag bei 91,93 % und die mittlere Sensitivität bei 92,77 %.

Fragestellung

Wie gut erkennen KI-basierte Algorithmen Tuberkulose über verschiedene Datenmodalitäten hinweg?

Methode

Systematische Übersicht nach PRISMA 2020; Suche in Scopus, PubMed und ACM Digital Library.

Ergebnisse

152 Studien aus 1146 Treffern. Mittlere Genauigkeit 91,93 % (95 %-KI 90,52–93,33 %), AUC 93,48 % (95 %-KI 91,90–95,06 %), Sensitivität 92,77 % (95 %-KI 91,38–94,15 %), Spezifität 92,39 % (95 %-KI 90,30–94,49 %).

Einschränkungen

Nur 1 Studie führte eine Domain-Shift-Analyse durch.

Schlussfolgerung der Autor:innen

Die Autor:innen sehen ein erhebliches Potenzial KI-basierter Methoden zur TB-Erkennung und empfehlen mehr Domain-Shift-Analysen für realistischere Szenarien.

Kernaussagen

  • 152 Studien wurden eingeschlossen
  • Mittlere Genauigkeit 91,93 %; AUC 93,48 %
  • Nur 1 Studie mit Domain-Shift-Analyse

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Systematische Übersicht in Journal of Medical Internet Research, published online on 2025-03-07. Written by Seng Hansun and 8 co-authors. Institution: School of Clinical Medicine, South West Sydney, UNSW Medicine & Health, UNSW Sydney, Sydney, Australia. Major MEDLINE topics: Artificial Intelligence, Tuberculosis. Cited 41 times so far (Crossref); 198 references. The full text is freely available in PubMed Central.

Auf einen BlickAbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Background

Tuberculosis (TB) remains a significant health concern, contributing to the highest mortality among infectious diseases worldwide. However, none of the various TB diagnostic tools introduced is deemed sufficient on its own for the diagnostic pathway, so various artificial intelligence (AI)–based methods have been developed to address this issue.

Objective

We aimed to provide a comprehensive evaluation of AI-based algorithms for TB detection across various data modalities.

Methods

Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) 2020 guidelines, we conducted a systematic review to synthesize current knowledge on this topic. Our search across 3 major databases (Scopus, PubMed, Association for Computing Machinery [ACM] Digital Library) yielded 1146 records, of which we included 152 (13.3%) studies in our analysis. QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies version 2) was performed for the risk-of-bias assessment of all included studies.

Results

Radiographic biomarkers (n=129, 84.9%) and deep learning (DL; n=122, 80.3%) approaches were predominantly used, with convolutional neural networks (CNNs) using Visual Geometry Group (VGG)-16 (n=37, 24.3%), ResNet-50 (n=33, 21.7%), and DenseNet-121 (n=19, 12.5%) architectures being the most common DL approach. The majority of studies focused on model development (n=143, 94.1%) and used a single modality approach (n=141, 92.8%). AI methods demonstrated good performance in all studies: mean accuracy=91.93% (SD 8.10%, 95% CI 90.52%-93.33%; median 93.59%, IQR 88.33%-98.32%), mean area under the curve (AUC)=93.48% (SD 7.51%, 95% CI 91.90%-95.06%; median 95.28%, IQR 91%-99%), mean sensitivity=92.77% (SD 7.48%, 95% CI 91.38%-94.15%; median 94.05% IQR 89%-98.87%), and mean specificity=92.39% (SD 9.4%, 95% CI 90.30%-94.49%; median 95.38%, IQR 89.42%-99.19%). AI performance across different biomarker types showed mean accuracies of 92.45% (SD 7.83%), 89.03% (SD 8.49%), and 84.21% (SD 0%); mean AUCs of 94.47% (SD 7.32%), 88.45% (SD 8.33%), and 88.61% (SD 5.9%); mean sensitivities of 93.8% (SD 6.27%), 88.41% (SD 10.24%), and 93% (SD 0%); and mean specificities of 94.2% (SD 6.63%), 85.89% (SD 14.66%), and 95% (SD 0%) for radiographic, molecular/biochemical, and physiological types, respectively. AI performance across various reference standards showed mean accuracies of 91.44% (SD 7.3%), 93.16% (SD 6.44%), and 88.98% (SD 9.77%); mean AUCs of 90.95% (SD 7.58%), 94.89% (SD 5.18%), and 92.61% (SD 6.01%); mean sensitivities of 91.76% (SD 7.02%), 93.73% (SD 6.67%), and 91.34% (SD 7.71%); and mean specificities of 86.56% (SD 12.8%), 93.69% (SD 8.45%), and 92.7% (SD 6.54%) for bacteriological, human reader, and combined reference standards, respectively. The transfer learning (TL) approach showed increasing popularity (n=89, 58.6%). Notably, only 1 (0.7%) study conducted domain-shift analysis for TB detection.

Conclusions

Findings from this review underscore the considerable promise of AI-based methods in the realm of TB detection. Future research endeavors should prioritize conducting domain-shift analyses to better simulate real-world scenarios in TB detection. Trial Registration PROSPERO CRD42023453611; https://www.crd.york.ac.uk/PROSPERO/view/CRD42023453611

Subjects

MeSH terms (MEDLINE)

Artificial IntelligenceTuberculosisHumansNeural Networks, ComputerDeep LearningAlgorithms

Author keywords

AIPRISMAPreferred Reporting Items for Systematic Reviews and Meta-AnalysisQUADAS-2Quality Assessment of Diagnostic Accuracy Studies version 2artificial intelligencedeep learningdiagnostic performancemachine learningsystematic literature reviewtuberculosis detection

Authors and institutions

  • School of Clinical Medicine, South West Sydney, UNSW Medicine & Health, UNSW Sydney, Sydney, Australia.
  • Woolcock Vietnam Research Group, Woolcock Institute of Medical Research, Sydney, Australia.
  • Graduate School of Biomedical Engineering, UNSW Sydney, Sydney, Australia.
  • Tyree Institute of Health Engineering, UNSW Sydney, Sydney, Australia.
  • Ageing Future Institute, UNSW Sydney, Sydney, Australia.
  • BioMedical Machine Learning Lab, Graduate School of Biomedical Engineering, UNSW Sydney, Sydney, Australia.

ORCID profiles: Seng Hansun, Ahmadreza Argha, Ivan Bakhshayeshi, Arya Wicaksana, Hamid Alinejad-Rokny, Greg J Fox, Siaw-Teng Liaw, Branko G Celler, Guy B Marks

Cite

APA

Hansun, S., Argha, A., Bakhshayeshi, I., Wicaksana, A., Alinejad-Rokny, H., Fox, G. J., Liaw, S. T., Celler, B. G., & Marks, G. B. (2025). Diagnostic Performance of Artificial Intelligence–Based Methods for Tuberculosis Detection: Systematic Review. Journal of Medical Internet Research, 27, e69068. https://doi.org/10.2196/69068

Vancouver

Hansun S, Argha A, Bakhshayeshi I, Wicaksana A, Alinejad-Rokny H, Fox GJ, et al. Diagnostic Performance of Artificial Intelligence–Based Methods for Tuberculosis Detection: Systematic Review. Journal of Medical Internet Research. 2025;27:e69068. doi:10.2196/69068

BibTeX
@article{Hansun2025,
  author = {Hansun, Seng and Argha, Ahmadreza and Bakhshayeshi, Ivan and Wicaksana, Arya and Alinejad-Rokny, Hamid and Fox, Greg J and Liaw, Siaw-Teng and Celler, Branko G and Marks, Guy B},
  title = {Diagnostic Performance of Artificial Intelligence–Based Methods for Tuberculosis Detection: Systematic Review},
  journal = {Journal of Medical Internet Research},
  year = {2025},
  volume = {27},
  pages = {e69068},
  doi = {10.2196/69068}
}
Start for free →