Comparison of Screening Dilution and Automated Reading for Antinuclear Antibody Detection on HEP2 Cells in the Monitoring of Connective Tissue Diseases

Berger, Anne‐Emmanuelle; Moreau, A; Bossy, V.; Genin, Christian; Rinaudo, M.; Paul, Stéphane

doi:10.1002/jcla.21881

Cited by 5 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NOVA View system has reached a high agreement with manual microscopic reading at IIF 1: 80 screening dilution [ 18 ], so we used a dilution of 1:80 to determine the ANA patterns. In our laboratory, five experienced and certified medical technologists (considered experienced readers in this study) who had 3–12 years (mean: 8 years) of experience in ANA IIF reading were responsible for ANA pattern reporting.…”

Section: Methodsmentioning

confidence: 99%

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus

Sheu

Chung

et al. 2021

Diagnostics

View full text Add to dashboard Cite

Background: Antinuclear antibody pattern recognition is vital for autoimmune disease diagnosis but labor-intensive for manual interpretation. To develop an automated pattern recognition system, we established machine learning models based on the International Consensus on Antinuclear Antibody Patterns (ICAP) at a competent level, mixed patterns recognition, and evaluated their consistency with human reading. Methods: 51,694 human epithelial cells (HEp-2) cell images with patterns assigned by experienced medical technologists collected in a medical center were used to train six machine learning algorithms and were compared by their performance. Next, we choose the best performing model to test the consistency with five experienced readers and two beginners. Results: The mean F1 score in each classification of the best performing model was 0.86 evaluated by Testing Data 1. For the inter-observer agreement test on Testing Data 2, the average agreement was 0.849 (?) among five experienced readers, 0.844 between the best performing model and experienced readers, 0.528 between experienced readers and beginners. The results indicate that the proposed model outperformed beginners and achieved an excellent agreement with experienced readers. Conclusions: This study demonstrated that the developed model could reach an excellent agreement with experienced human readers using machine learning methods.

show abstract

Section: Methodsmentioning

confidence: 99%

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus

Sheu

Chung

et al. 2021

Diagnostics

View full text Add to dashboard Cite

show abstract

“…The most accepted threshold is often the dilution 1/160 for first screening dilution ( 15 – 17 ). In complement to IIF assay, which is a very sensitive technic and can now be automated ( 18 , 19 ), screening fluorescence enzyme or chemiluminescence immunoassays have been proposed in the last few years as detection assays. These multiparametric immunoassays allow simultaneous testing for 13–17 of commonest pathogenic autoantibody specificities in systemic autoimmune diseases [i.e., SSA-52kD, SSA-60kD, SSB, U1RNP (RNP 70,A,C), CENP-B, Scl70, Jo1, Fibrillarin, RNA polymerase III, ribosomal proteins, PM-Scl, PCNA, Mi2 proteins, Sm, dsDNA, and chromatin].…”

Section: Aab In Healthy Population and In Non-autoimmune Diseasesmentioning

confidence: 99%

Autoantibodies Associated With Connective Tissue Diseases: What Meaning for Clinicians?

Didier¹,

et al. 2018

View full text Add to dashboard Cite

Connective tissue diseases (CTDs) such as systemic lupus erythematosus, systemic sclerosis, myositis, Sjögren’s syndrome, and rheumatoid arthritis are systemic diseases which are often associated with a challenge in diagnosis. Autoantibodies (AAbs) can be detected in these diseases and help clinicians in their diagnosis. Actually, pathophysiology of these diseases is associated with the presence of antinuclear antibodies. In the last decades, many new antibodies were discovered, but their implication in pathogenesis of CTDs remains unclear. Furthermore, the classification of these AAbs is nowadays misused, as their targets can be localized outside of the nuclear compartment. Interestingly, in most cases, each antibody is associated with a specific phenotype in CTDs and therefore help in better defining either the disease subtypes or diseases activity and outcome. Because of recent progresses in their detection and in the comprehension of their pathogenesis implication in CTD-associated antibodies, clinicians should pay attention to the presence of these different AAbs to improve patient’s management. In this review, we propose to focus on the different phenotypes and features associated with each autoantibody used in clinical practice in those CTDs.

show abstract

“…A further contribution to the harmonization of the process concerns the choice of the cutoff titer, which is fundamental for a correct classification of the samples as positive or negative. While it would be recommended for each laboratory to determine its own screening dilution for the local population to distinguish healthy and diseased states, in practice, this procedure is not followed by the vast majority of laboratories because there is a high consensus in the literature that the titer of 1:80 can be considered the best compromise between diagnostic sensitivity and specificity (21,36,(50)(51)(52). Furthermore, since the titer 1:80 is the screening dilution adopted by all manufacturers of CAD systems for automated reading and interpretation of ANA (24), this methodological approach represents a first and concrete step to achieve the harmonization of ANA HEp-2 IIF results.…”

Section: Standardization/harmonization Of Automated Ana Hep-2 Iif Assaysmentioning

confidence: 99%

Standardization and Quality Assessment Under the Perspective of Automated Computer-Assisted HEp-2 Immunofluorescence Assay Systems

Cinquanta

Bizzaro

2021

Front. Immunol.

View full text Add to dashboard Cite

The recent availability of automated computer-assisted diagnosis (CAD) systems for the reading and interpretation of the anti-nuclear antibody (ANA) test performed with the indirect immunofluorescence (IIF) method on HEp-2 cells, has improved the reproducibility of the results and initiated a process of harmonization of this test. Furthermore, CAD systems provide quantitative expression of fluorescence intensity, allowing the introduction of objective quality control procedures to the monitoring of the entire process. The calibration of the reading systems and the automated image interpretation are essential prerequisites for obtaining reproducible and harmonized IIF test results and form the basis for standardization, regardless of the computer algorithms used in the different systems. The use of automated CAD systems, facilitating control procedures, represents a step forward for the quality certification of the laboratory.

show abstract

Comparison of Screening Dilution and Automated Reading for Antinuclear Antibody Detection on HEP2 Cells in the Monitoring of Connective Tissue Diseases

Abstract: The automated NOVA View(®) reading of slides allows saving time, and an improvement in the standardization. Nevertheless, it requires a confirmation by a qualified operator, to interpret mixed patterns in particular.

Cited by 5 publications

References 9 publications

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus

Autoantibodies Associated With Connective Tissue Diseases: What Meaning for Clinicians?

Standardization and Quality Assessment Under the Perspective of Automated Computer-Assisted HEp-2 Immunofluorescence Assay Systems

Contact Info

Product

Resources

About