1887
Volume 2026, Issue 1
  • ISSN: 0253-8253
  • EISSN: 2227-0426

Abstract

The conventional diagnostic pathway for pleural malignancy involves pleural aspiration, often followed by tissue biopsy. However, pleural aspiration has limited sensitivity, which can potentially delay a definitive diagnosis. The objective of this study was to develop and externally validate a predictive model using clinical data to identify patients who could safely bypass pleural aspiration and proceed directly to tissue biopsy.

This was a retrospective cohort study of patients presenting to an acute hospital in the UK with suspected pleural malignancy between 2016 and 2025 (n = 646). A Random Forest classifier was trained on this dataset, and its performance was evaluated using an independent external validation cohort from another acute hospital in the UK ( = 32). Model performance was assessed using standard metrics, including accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (ROC-AUC).

The model demonstrated robust performance in the internal dataset and maintained its predictive strength during external validation (AUC: 0.86 vs. 0.83; precision 87% vs. 85%; recall 79% vs. 75%). Among the variables, computed tomography findings were the most influential predictor, as quantified by Gini importance. A simplified scoring system was subsequently derived for potential bedside clinical application.

This externally validated model provides a valuable decision-support tool for clinicians, facilitating earlier tissue diagnosis in patients with suspected pleural malignancy. By potentially avoiding non-diagnostic aspiration, the model can streamline the diagnostic pathway and expedite patient care. Successful external validation enhances confidence in the model’s generalizability and supports its potential for implementation into routine clinical practice.

© 2026 Omar, Gonzales, Ionescu, licensee HBKU Press.
Loading

Article metrics loading...

/content/journals/10.5339/qmj.2026.14
2026-03-19
2026-03-24

Metrics

Loading full text...

Full text loading...

/deliver/fulltext/qmj/2026/1/qmj.2026.14.html?itemId=/content/journals/10.5339/qmj.2026.14&mimeType=html&fmt=ahah

References

  1. Avasarala SK, Lentz RJ, Maldonado F. Medical thoracoscopy. Clin Chest Med. 2021 Dec; 42:(4):751–66. https://doi.org/10.1016/j.ccm.2021.08.010
     [Google Scholar]
  2. Roberts ME, Rahman NM, Maskell NA, Bibby AC, Blyth KG, Corcoran JP et al.. British Thoracic Society guideline for pleural disease. Thorax. 2023 Jul; 78:(Suppl 3):s1–42. https://doi.org/10.1136/thorax-2022-219784
     [Google Scholar]
  3. Mercer RM, Varatharajah R, Shepherd G, Lu Q, Castro-Añón O, McCracken DJ et al.. Critical analysis of the utility of initial pleural aspiration in the diagnosis and management of suspected malignant pleural effusion. BMJ Open Respir Res. 2020 Sep; 7:(1):e000701. https://doi.org/10.1136/bmjresp-2020-000701
     [Google Scholar]
  4. Ferreiro L, Suárez-Antelo J, Álvarez-Dobaño JM, Toubes ME, Riveiro V, Valdés L. Malignant pleural effusion: Diagnosis and management. Can Respir J. 2020 Sep 23:(1):2950751. https://doi.org/10.1155/2020/2950751
     [Google Scholar]
  5. Addala DN, Rahman NM. Man versus machine in pleural diagnostics: Does artificial intelligence provide the solution?. Ann Am Thorac Soc. 2024 Feb; 21:(2):202–3. https://doi.org/10.1513/AnnalsATS.202311-960ED
     [Google Scholar]
  6. Garcia-Zamalloa A, Arnay R, Castilla-Rodriguez I, Mar J, Gonzalez-Cava JM, Ibarrondo O et al.. Machine learning for predicting the diagnosis of tuberculous versus malignant pleural effusion: External validation and accuracy in two different settings. PLoS One. 2025 Sep 5; 20:(9):e0329668. https://doi.org/10.1371/journal.pone.0329668
     [Google Scholar]
  7. Hao J, Ho TK. Machine learning made easy: a review of scikit-learn package in python programming language. J Educ Behav Stat. 2019 June; 44:(3):348–61. https://doi.org/10.3102/1076998619832248
     [Google Scholar]
  8. Bradley AA, Schwartz SS, Hashino T. Sampling uncertainty and confidence intervals for the Brier score and Brier skill score. Weather Forecast. 2008 Oct 1; 23:(5):992–1006. https://doi.org/10.1175/2007WAF2007049.1
     [Google Scholar]
  9. Parmar A, Katariya R, Patel V. A review on random forest: An ensemble classifier. In: International conference on intelligent data communication technologies and internet of things. Springer; 2018 Aug 7. p. 758–63. https://doi.org/10.1007/978-3-030-03146-6_86
     [Google Scholar]
  10. Menze BH, Kelm BM, Masuch R, Himmelreich U, Bachert P, Petrich W et al.. A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data. BMC Bioinform. 2009 Jul 10; 10:(1):213. https://doi.org/10.1186/1471-2105-10-213
     [Google Scholar]
  11. Clive AO, Kahan BC, Hooper CE, Bhatnagar R, Morley AJ, Zahan-Evans N et al.. Predicting survival in malignant pleural effusion: development and validation of the LENT prognostic score. Thorax. 2014 Dec; 69:(12):1098–104. https://doi.org/10.1136/thoraxjnl-2014-205285
     [Google Scholar]
  12. Arnold DT, De Fonseka D, Perry S, Morley A, Harvey JE, Medford A et al.. Investigating unilateral pleural effusions: The role of cytology. Eur Respir J. 2018 Nov 8; 52:(5). https://doi.org/10.1183/13993003.01254-2018
     [Google Scholar]
  13. Tursz T, Bernards R. Hurdles on the road to personalized medicine. Mol Oncol. 2015 May; 9:(5):935–9. https://doi.org/10.1016/j.molonc.2014.08.009
     [Google Scholar]
  14. Hirsch FR, Wynes MW, Gandara DR, Bunn Jr PA. The tissue is the issue: personalized medicine for non-small cell lung cancer. Clin Cancer Res. 2010 Oct 15; 16:(20):4909–11. https://doi.org/10.1158/1078-0432.CCR-10-2005
     [Google Scholar]
  15. Tsim S, Paterson S, Cartwright D, Fong CJ, Alexander L, Kelly C et al.. Baseline predictors of negative and incomplete pleural cytology in patients with suspected pleural malignancy–data supporting ‘Direct to LAT’in selected groups. Lung Cancer. 2019 Jul; 133:123–9. https://doi.org/10.1016/j.lungcan.2019.05.017
     [Google Scholar]
  16. Kim NY, Jang BGu K-MPark YSKim Y-GCho J. Differential diagnosis of pleural effusion using machine learning. Ann Am Thorac Soc. 2024 Feb; 21:(2):211–7. https://doi.org/10.1513/AnnalsATS.202305-410OC
     [Google Scholar]
  17. Henderson DW, Reid G, Kao SC, van Zandwijk N, Klebe S. Challenges and controversies in the diagnosis of mesothelioma: Part 1. J Clin Pathol. 2013 Oct; 66:(10):847–53. https://doi.org/10.1136/jclinpath-2012-201303
     [Google Scholar]
  18. Rakha EA, Patil S, Abdulla K, Abdulkader M, Chaudry Z, Soomro IN. The sensitivity of cytologic evaluation of pleural fluid in the diagnosis of malignant mesothelioma. Diagn Cytopathol. 2010 Dec; 38:(12):874–9. https://doi.org/10.1002/dc.21303
     [Google Scholar]
  19. Navani N, Nankivell M, Lawrence DR, Lock S, Makker H, Baldwin DR et al.. Lung cancer diagnosis and staging with endobronchial ultrasound-guided transbronchial needle aspiration compared with conventional approaches: An open-label, pragmatic, randomised controlled trial. Lancet Respir Med. 2015 Apr; 3:(4):282–9. https://doi.org/10.1016/S2213-2600(15)00029-6
     [Google Scholar]
  20. Tsai W-C, Kung P-T, Wang Y-H, Kuo W-Y, Li Y-H. Influence of the time interval from diagnosis to treatment on survival for early-stage liver cancer. PLoS One. 2018 Jun 22; 13:(6):e0199532. https://doi.org/10.1371/journal.pone.0199532
     [Google Scholar]
  21. Segal A, Sterrett GF, Frost FA, Shilkin KB, Olsen NJ, Musk AW et al.. A diagnosis of malignant pleural mesothelioma can be made by effusion cytology: Results of a 20 year audit. Pathology. 2013 Jan; 45:(1):44–8. https://doi.org/10.1097/PAT.0b013e32835bc848
     [Google Scholar]
  22. Walters J, Maskell NA. Biopsy techniques for the diagnosis of mesothelioma. Recent Results Cancer Res. 2011: 189:;45–55. https://doi.org/10.1007/978-3-642-10862-4_4
     [Google Scholar]
  23. Doshi-Velez F, Kim B. Towards a rigorous science of interpretable machine learning [online]. arXiv Prepr arXiv170208608. 2017 Mar 2; Available from: https://arxiv.org/pdf/1702.08608v2
     [Google Scholar]
  24. Mediouni M, Makarenkov V, Diallo AB. Towards an interpretable machine learning model for predicting antimicrobial resistance. J Glob Antimicrob Resist. 2025 Dec; 45:47–51. https://doi.org/10.1016/j.jgar.2025.08.011
     [Google Scholar]
  25. Lipton ZC. The mythos of model interpretability: In machine learning, the concept of interpretability is both important and slippery. Queue [online]. 2018 Jun 1; 16:(3):31–57. Available from: https://spawn-queue.acm.org/doi/
     [Google Scholar]
  26. Orlandi R, Cara A, Cassina EM, Degiovanni S, Libretti L, Pirondini E et al.. Malignant pleural effusion: Diagnosis and treatment-up-to-date perspective. Curr Oncol. 2024 Nov 2; 31:(11):6867–78. https://doi.org/10.3390/curroncol31110507
     [Google Scholar]
  27. Ren W, Zhu Y, Wang Q, Jin H, Guo Y, Lin D. Deep learning-based classification and targeted gene alteration prediction from pleural effusion cell block whole-slide images. Cancers (Basel). 2023 Jan 25; 15:(3):752. https://doi.org/10.3390/cancers15030752
     [Google Scholar]
/content/journals/10.5339/qmj.2026.14
Loading
Bypassing pleural aspiration for early tissue diagnosis in suspected pleural malignancy: Development and external validation of a predictive model using supervised machine learning
Qatar Medical Journal 2026, 14 (2026); https://doi.org/10.5339/qmj.2026.14
/content/journals/10.5339/qmj.2026.14
/content/journals/10.5339/qmj.2026.14
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): machine learning, Pleural neoplasm, predictive value of tests, tissue biopsy and validation studies

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error