Startseite Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid
Artikel
Lizenziert
Nicht lizenziert Erfordert eine Authentifizierung

Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid

  • Xianfei Ye , Xinfeng Zhao , Yinyu Lou , Hanqi Pan und Yunying Chen EMAIL logo
Veröffentlicht/Copyright: 28. Mai 2025
Veröffentlichen auch Sie bei De Gruyter Brill
Manuskript einreichen Informationen für Autor*innen
Clinical Chemistry and Laboratory Medicine (CCLM)
Aus der Zeitschrift Clinical Chemistry and Laboratory Medicine (CCLM) Band 63 Heft 10

Abstract

Objectives

This study aimed to develop and validate a machine learning (ML) model utilizing cerebrospinal fluid (CSF) body fluid parameters from hematology analyzers to screen for malignant cells.

Methods

We analyzed 643 consecutive CSF samples from patients with central nervous system symptoms, with 191 samples classified as positive for malignant cells based on cytological examination, for model derivation. Body fluid parameters were measured using the body fluid mode of a hematology analyzer. Least Absolute Shrinkage and Selection Operator (LASSO) regression was applied to identify predictive biomarkers, followed by performance evaluations of six ML algorithms. Model interpretability was assessed using SHapley Additive exPlanations (SHAP). The selected model was also externally validated with an additional 136 CSF samples.

Results

The median leukocyte (WBC) and total nucleated cell (TNC) counts in the cytology-positive samples were significantly lower than those in the cytology-negative samples (5.4 vs. 31.8 and 7.4 vs. 32.6, respectively, p<0.001). The support vector machine (SVM) model achieved the highest area under the curve (AUC) of 0.899 (SD: 0.035) and the highest sensitivity of 0.827 (SD: 0.059) in internal validation. SHAP analysis identified the percentage of high fluorescence cells and monocytes as the two most significant predictors, both positively correlated with malignant cell outcomes. External validation demonstrated a comparable AUC and sensitivity, confirming the model’s generalizability.

Conclusions

We developed an ML model that predicts cytological outcomes in CSF using routinely available body fluid parameters. The model demonstrated consistent performance during external validation.

Keywords: machine learning; body fluid; cerebrospinal fluid; high fluorescence cells; malignant cells
Corresponding author: Yunying Chen, Department of Laboratory Medicine, Hangzhou Children’s Hospital, No.195 Wenhui Road, Hangzhou, Zhejiang, 310014, P.R. China, E-mail:

Acknowledgments

We would like to thank the Extreme Smart Analysis platform for its analysis assistance.

  1. Research ethics: This study was approved by the Ethics Committee of the First Affiliated Hospital of Zhejiang University, ethics approval number: (2024) IIT consent letter No. (0811).

  2. Informed consent: The requirement for informed consent was waived for this study.

  3. Author contributions: Xianfei Ye collected the samples, analyzed the data, and wrote the draft. Xinfeng Zhao contributed to data analysis and figure creation. Yinyu Lou assisted with data collection, sample analysis, and morphological evaluation. Hanqi Pan was responsible for the cellular pathological diagnosis and reviewed the results of the discordant samples. Yunying Chen developed the research idea, designed the study, and established the machine learning models. All authors reviewed and approved the final manuscript.

  4. Use of Large Language Models, AI and Machine Learning Tool: None declared.

  5. Conflict of interest: The authors state no conflict of interest.

  6. Research funding: This work was supported by the Hangzhou Municipal Health Commission Project (20241029Y063).

  7. Data availability: The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

1. Clinical and Laboratory Standards Institution (CLSI). Body fluid analysis for cellular composition. Wayne, PA: Approved Guideline: CLSI Document H56-A; 2006.Suche in Google Scholar

2. Chamberlain, M, Soffietti, R, Raizer, J, Rudà, R, Brandsma, D, Boogerd, W, et al.. Leptomeningeal metastasis: a Response Assessment in Neuro-Oncology critical review of endpoints and response criteria of published randomized clinical trials. Neuro Oncol 2014;16:1176–85. https://doi.org/10.1093/neuonc/nou089.Suche in Google Scholar PubMed PubMed Central

3. Bourner, G, De la Salle, B, George, T, Tabe, Y, Baum, H, Culp, N, et al.. ICSH guidelines for the verification and performance of automated cell counters for body fluids. Int J Lab Hematol 2014;36:598–612. https://doi.org/10.1111/ijlh.12196.Suche in Google Scholar PubMed

4. Fleming, C, Brouwer, R, Lindemans, J, de Jonge, R. Validation of the body fluid module on the new Sysmex XN-1000 for counting blood cells in cerebrospinal fluid and other body fluids. Clin Chem Lab Med 2012;50:1791–8. https://doi.org/10.1515/cclm-2011-0927.Suche in Google Scholar PubMed

5. de Jonge, R, Brouwer, R, de Graaf, MT, Luitwieler, RL, Fleming, C, de Frankrijker-Merkestijn, M, et al.. Evaluation of the new body fluid mode on the Sysmex XE-5000 for counting leukocytes and erythrocytes in cerebrospinal fluid and other body fluids. Clin Chem Lab Med 2010;48:665–75. https://doi.org/10.1515/cclm.2010.108.Suche in Google Scholar PubMed

6. Sandhaus, LM. Is the hemocytometer obsolete for body fluid cell counting? Am J Clin Pathol 2016;145:294–5. https://doi.org/10.1093/ajcp/aqw014.Suche in Google Scholar PubMed

7. Buoro, S, Peruzzi, B, Fanelli, A, Seghezzi, M, Manenti, B, Lorubbio, M, et al.. Two-site evaluation of the diagnostic performance of the Sysmex XN Body Fluid (BF) module for cell count and differential in Cerebrospinal Fluid. Int J Lab Hematol 2018;40:26–33. https://doi.org/10.1111/ijlh.12723.Suche in Google Scholar PubMed

8. Waldrop, GE, Cocuzzo, K, Schneider, CL, Kim, CY, Goetz, TG, Chomba, MS, et al.. Accuracy of automated analyzers for the estimation of CSF cell counts: a systematic review and meta-analysis. Int J Lab Hematol 2024;46:234–42. https://doi.org/10.1111/ijlh.14236.Suche in Google Scholar PubMed

9. Cho, YU, Chi, HS, Park, SH, Jang, S, Kim, YJ, Park, CJ. Body fluid cellular analysis using the Sysmex XN-2000 automatic hematology analyzer: focusing on malignant samples. Int J Lab Hematol 2015;37:346–56. https://doi.org/10.1111/ijlh.12292.Suche in Google Scholar PubMed

10. Cho, HE, Kim, YJ, Cho, SY, Park, TS, Park, KS. Clinical application of an algorithm to screen for malignant cells in body fluids using an automated hematology analyzer. Int J Lab Hematol 2022;44:483–9. https://doi.org/10.1111/ijlh.13813.Suche in Google Scholar PubMed

11. Saini, A, Sareen, R, Gupta, GN. High fluorescent cells on automated body fluid analysis as discriminator for malignant cell detection. S Asian J Cancer 2023;00:00. https://doi.org/10.1055/s-0043-1776287.Suche in Google Scholar

12. Alcaide Martín, MJ, Altimira Queral, L, Sahuquillo Frías, L, Valiña Amado, L, Merino, A, García de Guadiana-Romualdo, L. Automated cell count in body fluids: a review. Adv Lab Med 2021;2:149–77. https://doi.org/10.1515/almed-2021-0011.Suche in Google Scholar PubMed PubMed Central

13. Rahimi, J, Woehrer, A. Overview of cerebrospinal fluid cytology. Handb Clin Neurol 2017;145:563–71. https://doi.org/10.1016/b978-0-12-802395-2.00035-3.Suche in Google Scholar

14. Fleming, C, Russcher, H, Lindemans, J, de Jonge, R. Clinical relevance and contemporary methods for counting blood cells in body fluids suspected of inflammatory disease. Clin Chem Lab Med 2015;53:1689–706. https://doi.org/10.1515/cclm-2014-1247.Suche in Google Scholar PubMed

15. Schwarz, B, Hardt, C, Friedrich, K, Prpic, M, Osterloh, A, Heppner, FL, et al.. Sysmex XN-based evaluation of the diagnostic performance of high-fluorescent cells from CSF as a supportive diagnostic criterion in neurological diseases. Int J Lab Hematol 2025. https://doi.org/10.1111/ijlh.14466. [Epub ahead of print].Suche in Google Scholar PubMed PubMed Central

16. Wong-Arteta, J, Merino, A, Torrente, S, Banales, JM, Bujanda, L. High fluorescence cell count in ascitic body fluids for carcinomatosis screening. Clin Chem Lab Med 2018;56:272–4. https://doi.org/10.1515/cclm-2018-0359.Suche in Google Scholar PubMed

17. Larruzea, A, Aguadero, V, Orellana, R, Berlanga, E. High-fluorescent cells: a marker of malignancy in the analysis of body fluid samples. Int J Lab Hematol 2018;40:e43–5. https://doi.org/10.1111/ijlh.12793.Suche in Google Scholar PubMed

18. Wong-Arteta, J, Gil-Rodríguez, E, Cabezon-Vicente, R, Bereciartua-Urbieta, E, Bujanda, L. High fluorescence cell count in pleural fluids for malignant effusion screening. Clin Chim Acta 2019;499:115–7. https://doi.org/10.1016/j.cca.2019.09.008.Suche in Google Scholar PubMed

19. Mishra, S, Parikh, BP, Singh, J. Diagnostic utility of high fluorescence cells in detecting malignant effusions. J Cytol 2024;41:176–80. https://doi.org/10.4103/joc.joc_122_23.Suche in Google Scholar PubMed PubMed Central

20. Rastogi, L, Dass, J, Arya, V, Kotwal, J. Evaluation of high-fluorescence body fluid (HF-BF) parameter as a screening tool of malignancy in body fluids. Indian J Pathol Microbiol 2019;62:572–7. https://doi.org/10.4103/ijpm.ijpm_802_18.Suche in Google Scholar PubMed

21. Zimmermann, M, Otto, C, Gonzalez, JB, Prokop, S, Ruprecht, K. Cellular origin and diagnostic significance of high-fluorescent cells in cerebrospinal fluid detected by the XE-5000 hematology analyzer. Int J Lab Hematol 2013;35:580–8. https://doi.org/10.1111/ijlh.12090.Suche in Google Scholar PubMed

22. Buoro, S, Mecca, T, Azzarà, G, Seghezzi, M, Dominoni, P, Crippa, A, et al.. Cell Population Data and reflex testing rules of cell analysis in pleural and ascitic fluids using body fluid mode on Sysmex XN-9000. Clin Chim Acta 2016;452:92–8. https://doi.org/10.1016/j.cca.2015.11.005.Suche in Google Scholar PubMed

23. Cho, YU, You, E, Jang, S, Park, CJ. Validation of reflex testing rules and establishment of a new workflow for body fluid cell analysis using a Sysmex XN-550 automatic hematology analyzer. Int J Lab Hematol 2018;40:258–67. https://doi.org/10.1111/ijlh.12774.Suche in Google Scholar PubMed

24. Cancella, DAM, Brumpt, C, Sala, T, Oueidat, N, Larsen, M, Hausfater, P. Cell population data for early detection of sepsis in patients with suspected infection in the emergency department. Clin Chem Lab Med 2025;63:1654–62. https://doi.org/10.1515/cclm-2025-0180.Suche in Google Scholar PubMed

25. Roccaforte, V, Sabbatini, G, Panella, R, Daves, M, Formenti, P, Gotti, M, et al.. The potential role of leukocytes cell population data (CPD) for diagnosing sepsis in adult patients admitted to the intensive care unit. Clin Chem Lab Med 2025;63:1031–42. https://doi.org/10.1515/cclm-2024-1202.Suche in Google Scholar PubMed

26. Davis, SE, Greevy, RA, Fonnesbeck, C, Lasko, TA, Walsh, CG, Matheny, ME. A nonparametric updating method to correct clinical prediction model drift. J Am Med Inf Assoc 2019;26:1448–57. https://doi.org/10.1093/jamia/ocz127.Suche in Google Scholar PubMed PubMed Central

Supplementary Material

This article contains supplementary material (https://doi.org/10.1515/cclm-2025-0302).

Received: 2025-03-13
Accepted: 2025-05-13
Published Online: 2025-05-28
Published in Print: 2025-09-25

© 2025 Walter de Gruyter GmbH, Berlin/Boston

Artikel in diesem Heft

  1. Frontmatter
  2. Editorial
  3. Quality indicators: an evolving target for laboratory medicine
  4. Reviews
  5. Regulating the future of laboratory medicine: European regulatory landscape of AI-driven medical device software in laboratory medicine
  6. The spectrum of nuclear patterns with stained metaphase chromosome plate: morphology nuances, immunological associations, and clinical relevance
  7. Opinion Papers
  8. Comprehensive assessment of medical laboratory performance: a 4D model of quality, economics, velocity, and productivity indicators
  9. Detecting cardiac injury: the next generation of high-sensitivity cardiac troponins improving diagnostic outcomes
  10. Perspectives
  11. Can Theranos resurrect from its ashes?
  12. Guidelines and Recommendations
  13. Australasian guideline for the performance of sweat chloride testing 3rd edition: to support cystic fibrosis screening, diagnosis and monitoring
  14. General Clinical Chemistry and Laboratory Medicine
  15. Recommendations for the integration of standardized quality indicators for glucose point-of-care testing
  16. A cost-effective assessment for the combination of indirect immunofluorescence and solid-phase assay in ANA-screening
  17. Assessment of measurement uncertainty of immunoassays and LC-MS/MS methods for serum 25-hydroxyvitamin D
  18. A novel immunoprecipitation-based targeted liquid chromatography-tandem mass spectrometry analysis for accurate determination for copeptin in human serum
  19. Histamine metabolite to basal serum tryptase ratios in systemic mastocytosis and hereditary alpha tryptasemia using a validated LC-MS/MS approach
  20. Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid
  21. Impact of analytical bias on machine learning models for sepsis prediction using laboratory data
  22. Immunochemical measurement of urinary free light chains and Bence Jones proteinuria
  23. Serum biomarkers as early indicators of outcomes in spontaneous subarachnoid hemorrhage
  24. High myoglobin plasma samples risk being reported as falsely low due to antigen excess – follow up after a 2-year period of using a mitigating procedure
  25. Candidate Reference Measurement Procedures and Materials
  26. Commutability evaluation of glycated albumin candidate EQA materials
  27. Reference Values and Biological Variations
  28. Health-related reference intervals for heavy metals in non-exposed young adults
  29. Hematology and Coagulation
  30. Practical handling of hemolytic, icteric and lipemic samples for coagulation testing in European laboratories. A collaborative survey from the European Organisation for External Quality Assurance Providers in Laboratory Medicine (EQALM)
  31. Cancer Diagnostics
  32. Assessment of atypical cells in detecting bladder cancer in female patients
  33. Cardiovascular Diseases
  34. False-positive cardiac troponin I values due to macrotroponin in healthy athletes after COVID-19
  35. Diabetes
  36. A comparison of current methods to measure antibodies in type 1 diabetes
  37. Letters to the Editor
  38. The neglected issue of pyridoxal- 5′ phosphate
  39. Error in prostate-specific antigen levels after prostate cancer treatment with radical prostatectomy
  40. Arivale is dead ‒ Hooke is alive
  41. A single dose of 20-mg of ostarine is detectable in hair
  42. Growing importance of vocabularies in medical laboratories
  43. Congress Abstracts
  44. 62nd National Congress of the Hungarian Society of Laboratory Medicine Szeged, Hungary, August 28–30, 2025
https://doi.org/10.1515/cclm-2025-0302
Schlagwörter für diesen Artikel
machine learning; body fluid; cerebrospinal fluid; high fluorescence cells; malignant cells

Artikel in diesem Heft

  1. Frontmatter
  2. Editorial
  3. Quality indicators: an evolving target for laboratory medicine
  4. Reviews
  5. Regulating the future of laboratory medicine: European regulatory landscape of AI-driven medical device software in laboratory medicine
  6. The spectrum of nuclear patterns with stained metaphase chromosome plate: morphology nuances, immunological associations, and clinical relevance
  7. Opinion Papers
  8. Comprehensive assessment of medical laboratory performance: a 4D model of quality, economics, velocity, and productivity indicators
  9. Detecting cardiac injury: the next generation of high-sensitivity cardiac troponins improving diagnostic outcomes
  10. Perspectives
  11. Can Theranos resurrect from its ashes?
  12. Guidelines and Recommendations
  13. Australasian guideline for the performance of sweat chloride testing 3rd edition: to support cystic fibrosis screening, diagnosis and monitoring
  14. General Clinical Chemistry and Laboratory Medicine
  15. Recommendations for the integration of standardized quality indicators for glucose point-of-care testing
  16. A cost-effective assessment for the combination of indirect immunofluorescence and solid-phase assay in ANA-screening
  17. Assessment of measurement uncertainty of immunoassays and LC-MS/MS methods for serum 25-hydroxyvitamin D
  18. A novel immunoprecipitation-based targeted liquid chromatography-tandem mass spectrometry analysis for accurate determination for copeptin in human serum
  19. Histamine metabolite to basal serum tryptase ratios in systemic mastocytosis and hereditary alpha tryptasemia using a validated LC-MS/MS approach
  20. Machine learning algorithms with body fluid parameters: an interpretable framework for malignant cell screening in cerebrospinal fluid
  21. Impact of analytical bias on machine learning models for sepsis prediction using laboratory data
  22. Immunochemical measurement of urinary free light chains and Bence Jones proteinuria
  23. Serum biomarkers as early indicators of outcomes in spontaneous subarachnoid hemorrhage
  24. High myoglobin plasma samples risk being reported as falsely low due to antigen excess – follow up after a 2-year period of using a mitigating procedure
  25. Candidate Reference Measurement Procedures and Materials
  26. Commutability evaluation of glycated albumin candidate EQA materials
  27. Reference Values and Biological Variations
  28. Health-related reference intervals for heavy metals in non-exposed young adults
  29. Hematology and Coagulation
  30. Practical handling of hemolytic, icteric and lipemic samples for coagulation testing in European laboratories. A collaborative survey from the European Organisation for External Quality Assurance Providers in Laboratory Medicine (EQALM)
  31. Cancer Diagnostics
  32. Assessment of atypical cells in detecting bladder cancer in female patients
  33. Cardiovascular Diseases
  34. False-positive cardiac troponin I values due to macrotroponin in healthy athletes after COVID-19
  35. Diabetes
  36. A comparison of current methods to measure antibodies in type 1 diabetes
  37. Letters to the Editor
  38. The neglected issue of pyridoxal- 5′ phosphate
  39. Error in prostate-specific antigen levels after prostate cancer treatment with radical prostatectomy
  40. Arivale is dead ‒ Hooke is alive
  41. A single dose of 20-mg of ostarine is detectable in hair
  42. Growing importance of vocabularies in medical laboratories
  43. Congress Abstracts
  44. 62nd National Congress of the Hungarian Society of Laboratory Medicine Szeged, Hungary, August 28–30, 2025
Jetzt anmelden und 20% sparen
Institutioneller Zugang
Wie funktioniert Authentifizierung?
Haben Sie eine Idee, wie wir unsere Website verbessern können?
Bitte schreiben Sie uns.
© 2025 De Gruyter Brill
Heruntergeladen am 10.11.2025 von https://www.degruyterbrill.com/document/doi/10.1515/cclm-2025-0302/html?lang=de
Button zum nach oben scrollen