Multivariate anomaly detection models enhance identification of errors in routine clinical chemistry testing
Abstract
Objectives
Conventional autoverification rules evaluate analytes independently, potentially missing unusual patterns of results indicative of errors such as serum contamination by collection tube additives. This study assessed whether multivariate anomaly detection algorithms could enhance the detection of such errors.
Methods
Multivariate Gaussian, k-nearest neighbours (KNN) distance, and one-class support vector machine (SVM) anomaly detection models, along with conventional limit checks, were developed using a training dataset of 127,451 electrolyte, urea, and creatinine (EUC) results, with a 5 % flagging rate targeted for all approaches. The models were compared with limit checks for their ability to detect atypical EUC results from samples spiked with additives from collection tubes: EDTA, fluoride, sodium citrate, or acid citrate dextrose (n=200 per contaminant). The study additionally assessed the ability of the models to identify 127,449 single-analyte errors, a potential weakness of multivariate models.
Results
The KNN distance and SVM models outperformed limit checks for detecting all contaminants (p-values <0.05). The multivariate Gaussian model did not surpass limit checks for detecting EDTA contamination but was superior for detecting the other additives. All models surpassed limit checks for identifying single-analyte errors, with the KNN distance model demonstrating the highest overall sensitivity.
Conclusions
Multivariate anomaly detection models, particularly the KNN distance model, were superior to the conventional approach for detecting serum contamination and single-analyte errors. Developing multivariate approaches to autoverification is warranted to optimise error detection and improve patient safety.
Acknowledgments
The author wishes to thank Jiji Anthony, Manar Benyamin, Ithar Daniel, Katarina Drlja, Anila Hashmi, Srina Manoharan, Matthew Niven, Nirina Razafiarisoa, and Baljit Singh for assistance with analysing samples.
-
Research ethics: The local Institutional Review Board deemed the study exempt from review.
-
Informed consent: Not applicable.
-
Author contributions: The author has accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: The author states no conflict of interest.
-
Research funding: None declared.
-
Data availability: The raw data can be obtained on request from the corresponding author.
References
1. Plebani, M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006;44:750–9. https://doi.org/10.1515/cclm.2006.123.Search in Google Scholar PubMed
2. Torke, N, Boral, L, Nguyen, T, Perri, A, Chakrin, A. Process improvement and operational efficiency through test result autoverification. Clin Chem 2005;51:2406–8. https://doi.org/10.1373/clinchem.2005.054395.Search in Google Scholar PubMed
3. Shih, MC, Chang, HM, Tien, N, Hsiao, CT, Peng, CT. Building and validating an autoverification system in the clinical chemistry laboratory. Lab Med 2011;42:668–73. https://doi.org/10.1309/lm5am4iixc4oietd.Search in Google Scholar
4. Rimac, V, Lapic, I, Kules, K, Rogic, D, Miler, M. Implementation of the autovalidation algorithm for clinical chemistry testing in the laboratory information system. Lab Med 2018;49:284–91. https://doi.org/10.1093/labmed/lmx089.Search in Google Scholar PubMed
5. Fernández-Grande, E, Valera-Rodriguez, C, Sáenz-Mateos, L, Sastre-Gómez, A, García-Chico, P, Palomino-Muñoz, TJ. Impact of reference change value (RCV) based autoverification on turnaround time and physician satisfaction. Biochem Med 2017;27:342–9. https://doi.org/10.11613/bm.2017.037.Search in Google Scholar
6. Schifman, RB, Talbert, M, Souers, RJ. Delta check practices and outcomes: a Q-probes study involving 49 health care facilities and 6541 delta check alerts. Arch Pathol Lab Med 2017;141:813–23. https://doi.org/10.5858/arpa.2016-0161-cp.Search in Google Scholar
7. Whitehurst, P, Di Silvio, TV, Boyadjian, G. Evaluation of discrepancies in patients’ results – an aspect of computer-assisted quality control. Clin Chem 1975;21:87–92. https://doi.org/10.1093/clinchem/21.1.87.Search in Google Scholar
8. Cornes, MP, Ford, C, Gama, R. Undetected spurious hypernatraemia wastes health-care resources. Ann Clin Biochem 2011;48:87–8. https://doi.org/10.1258/acb.2010.010200.Search in Google Scholar PubMed
9. Lamb, EJ, Abbas, NA. Spurious hypernatraemia and Citra-Lock. Ann Clin Biochem 2007;44:579. https://doi.org/10.1258/000456307782268219.Search in Google Scholar PubMed
10. Milliere, J, Corriveau, D, Parmar, MS. Pseudohypernatraemia secondary to trisodium citrate (Citra-Lock™). Biochem Med 2016;26:260–3. https://doi.org/10.11613/bm.2016.030.Search in Google Scholar PubMed PubMed Central
11. Sharratt, CL, Gilbert, CJ, Cornes, MC, Ford, C, Gama, R. EDTA sample contamination is common and often undetected, putting patients at unnecessary risk of harm. Int J Clin Pract 2009;63:1259–62. https://doi.org/10.1111/j.1742-1241.2008.01981.x.Search in Google Scholar PubMed
12. IBM. What is anomaly detection? [online]. https://www.ibm.com/topics/anomaly-detection [Accessed 4 Mar 2024].Search in Google Scholar
13. Venables, WN, Ripley, BD. Modern applied statistics with S, 4th ed.. New York: Springer; 2002.10.1007/978-0-387-21706-2Search in Google Scholar
14. Beygelzimer, A, Kakadet, S, Langford, J, Arya, S, Mount, D, Li, S. FNN: fast nearest neighbor search algorithms and applications. R package version 1.1.3.2; 2023. Available from: https://CRAN.R-project.org/package=FNN.Search in Google Scholar
15. Meyer, D, Dimitriadou, E, Hornik, K, Weingessel, A, Leisch, F. e1071: misc functions of the Department of Statistics, Probability Theory Group (formerly: E1071), TU Wien. R package version 1.7-13; 2023. Available from: https://CRAN.R-project.org/package=e1071.Search in Google Scholar
16. Asif, U, Whitehead, SJ, Ford, C, Gama, R. Preanalytical potassium EDTA sample contamination: open versus closed phlebotomy systems. Ann Clin Biochem 2019;56:711–4. https://doi.org/10.1177/0004563219878463.Search in Google Scholar PubMed
17. R Core Team. R: a language and environment for statistical computing. Vienna (Austria): R Foundation for Statistical Computing; 2022.Search in Google Scholar
18. Randell, EW, Yenice, S, Khine Wamono, AA, Orth, M. Autoverification of test results in the core clinical laboratory. Clin Biochem 2019;73:11–25. https://doi.org/10.1016/j.clinbiochem.2019.08.002.Search in Google Scholar PubMed
19. Randell, EW, Short, G, Lee, N, Beresford, A, Spencer, M, Kennel, M, et al.. Strategy for 90% autoverification of clinical chemistry and immunoassay test results using six sigma process improvement. Data Brief 2018;18:1740–9. https://doi.org/10.1016/j.dib.2018.04.080.Search in Google Scholar PubMed PubMed Central
Supplementary Material
This article contains supplementary material (https://doi.org/10.1515/cclm-2024-0484).
© 2024 Walter de Gruyter GmbH, Berlin/Boston
Articles in the same Issue
- Frontmatter
- Editorial
- External quality assurance (EQA): navigating between quality and sustainability
- Reviews
- Molecular allergology: a clinical laboratory tool for precision diagnosis, stratification and follow-up of allergic patients
- Nitrous oxide abuse direct measurement for diagnosis and follow-up: update on kinetics and impact on metabolic pathways
- Opinion Papers
- A vision to the future: value-based laboratory medicine
- Point-of-care testing, near-patient testing and patient self-testing: warning points
- Navigating the path of reproducibility in microRNA-based biomarker research with ring trials
- Point/Counterpoint
- Six Sigma – is it time to re-evaluate its value in laboratory medicine?
- The value of Sigma-metrics in laboratory medicine
- Genetics and Molecular Diagnostics
- Analytical validation of the amplification refractory mutation system polymerase chain reaction-capillary electrophoresis assay to diagnose spinal muscular atrophy
- Can we identify patients carrying targeted deleterious DPYD variants with plasma uracil and dihydrouracil? A GPCO-RNPGx retrospective analysis
- General Clinical Chemistry and Laboratory Medicine
- Comparison of ChatGPT, Gemini, and Le Chat with physician interpretations of medical laboratory questions from an online health forum
- External quality assessment performance in ten countries: an IFCC global laboratory quality project
- Multivariate anomaly detection models enhance identification of errors in routine clinical chemistry testing
- Enhanced patient-based real-time quality control using the graph-based anomaly detection
- Performance evaluation and user experience of BT-50 transportation unit with automated and scheduled quality control measurements
- Stability of steroid hormones in dried blood spots (DBS)
- Quantification of C1 inhibitor activity using a chromogenic automated assay: analytical and clinical performances
- Reference Values and Biological Variations
- Time-dependent characteristics of analytical measurands
- Cancer Diagnostics
- Expert-level detection of M-proteins in serum protein electrophoresis using machine learning
- An automated workflow based on data independent acquisition for practical and high-throughput personalized assay development and minimal residual disease monitoring in multiple myeloma patients
- Cardiovascular Diseases
- Analytical validation of the Mindray CL1200i analyzer high sensitivity cardiac troponin I assay: MERITnI study
- Diabetes
- Limitations of glycated albumin standardization when applied to the assessment of diabetes patients
- Patient result monitoring of HbA1c shows small seasonal variations and steady decrease over more than 10 years
- Letters to the Editor
- Inaccurate definition of Bence Jones proteinuria in the EFLM Urinalysis Guideline 2023
- Use of the term “Bence-Jones proteinuria” in the EFLM European Urinalysis Guideline 2023
- Is uracil enough for effective pre-emptive DPD testing?
- Reply to: “Is uracil enough for effective pre-emptive DPD testing?”
- Accurate predictory role of monocyte distribution width on short-term outcome in sepsis patients
- Reply to: “Accurate predictory role of monocyte distribution width on short-term outcome in sepsis patients”
- Spurious parathyroid hormone (PTH) elevation caused by macro-PTH
- Setting analytical performance specifications for copeptin-based testing
- Serum vitamin B12 levels during chemotherapy against diffuse large B-cell lymphoma: a case report and review of the literature
- Evolution of acquired haemoglobin H disease monitored by capillary electrophoresis: a case of a myelofibrotic patient with a novel ATRX mutation
Articles in the same Issue
- Frontmatter
- Editorial
- External quality assurance (EQA): navigating between quality and sustainability
- Reviews
- Molecular allergology: a clinical laboratory tool for precision diagnosis, stratification and follow-up of allergic patients
- Nitrous oxide abuse direct measurement for diagnosis and follow-up: update on kinetics and impact on metabolic pathways
- Opinion Papers
- A vision to the future: value-based laboratory medicine
- Point-of-care testing, near-patient testing and patient self-testing: warning points
- Navigating the path of reproducibility in microRNA-based biomarker research with ring trials
- Point/Counterpoint
- Six Sigma – is it time to re-evaluate its value in laboratory medicine?
- The value of Sigma-metrics in laboratory medicine
- Genetics and Molecular Diagnostics
- Analytical validation of the amplification refractory mutation system polymerase chain reaction-capillary electrophoresis assay to diagnose spinal muscular atrophy
- Can we identify patients carrying targeted deleterious DPYD variants with plasma uracil and dihydrouracil? A GPCO-RNPGx retrospective analysis
- General Clinical Chemistry and Laboratory Medicine
- Comparison of ChatGPT, Gemini, and Le Chat with physician interpretations of medical laboratory questions from an online health forum
- External quality assessment performance in ten countries: an IFCC global laboratory quality project
- Multivariate anomaly detection models enhance identification of errors in routine clinical chemistry testing
- Enhanced patient-based real-time quality control using the graph-based anomaly detection
- Performance evaluation and user experience of BT-50 transportation unit with automated and scheduled quality control measurements
- Stability of steroid hormones in dried blood spots (DBS)
- Quantification of C1 inhibitor activity using a chromogenic automated assay: analytical and clinical performances
- Reference Values and Biological Variations
- Time-dependent characteristics of analytical measurands
- Cancer Diagnostics
- Expert-level detection of M-proteins in serum protein electrophoresis using machine learning
- An automated workflow based on data independent acquisition for practical and high-throughput personalized assay development and minimal residual disease monitoring in multiple myeloma patients
- Cardiovascular Diseases
- Analytical validation of the Mindray CL1200i analyzer high sensitivity cardiac troponin I assay: MERITnI study
- Diabetes
- Limitations of glycated albumin standardization when applied to the assessment of diabetes patients
- Patient result monitoring of HbA1c shows small seasonal variations and steady decrease over more than 10 years
- Letters to the Editor
- Inaccurate definition of Bence Jones proteinuria in the EFLM Urinalysis Guideline 2023
- Use of the term “Bence-Jones proteinuria” in the EFLM European Urinalysis Guideline 2023
- Is uracil enough for effective pre-emptive DPD testing?
- Reply to: “Is uracil enough for effective pre-emptive DPD testing?”
- Accurate predictory role of monocyte distribution width on short-term outcome in sepsis patients
- Reply to: “Accurate predictory role of monocyte distribution width on short-term outcome in sepsis patients”
- Spurious parathyroid hormone (PTH) elevation caused by macro-PTH
- Setting analytical performance specifications for copeptin-based testing
- Serum vitamin B12 levels during chemotherapy against diffuse large B-cell lymphoma: a case report and review of the literature
- Evolution of acquired haemoglobin H disease monitored by capillary electrophoresis: a case of a myelofibrotic patient with a novel ATRX mutation
