Evaluation of schistocyte analysis by a novel automated digital cell morphology application

Computer vision quantitation of erythrocyte shape abnormalities provides diagnostic, prognostic, and mechanistic insight

Examination of red blood cell (RBC) morphology in peripheral blood smears can help diagnose hematologic diseases, even in resource-limited settings, but this analysis remains subjective and semiquantitative with low throughput. Prior attempts to develop automated tools have been hampered by their poor reproducibility and limited clinical validation. Here, we present a novel, open-source machine-learning approach (denoted as RBC-diff) to quantify abnormal RBCs in peripheral smear images and generate an RBC morphology differential. RBC-diff cell counts showed high accuracy for single-cell classification (mean AUC, 0.93) and quantitation across smears (mean R2, 0.76 compared with experts, interexperts R2, 0.75). RBC-diff counts were concordant with the clinical morphology grading for 300 000+ images and recovered the expected pathophysiologic signals in diverse clinical cohorts. Criteria using RBC-diff counts distinguished thrombotic thrombocytopenic purpura and hemolytic uremic syndrome from other thrombotic microangiopathies, providing greater specificity than clinical morphology grading (72% vs 41%; P < .001) while maintaining high sensitivity (94% to 100%). Elevated RBC-diff schistocyte counts were associated with increased 6-month all-cause mortality in a cohort of 58 950 inpatients (9.5% mortality for schist. >1%, vs 4.7% for schist; <0.5%; P < .001) after controlling for comorbidities, demographics, clinical morphology grading, and blood count indices. RBC-diff also enabled the estimation of single-cell volume-morphology distributions, providing insight into the influence of morphology on routine blood count measures. Our codebase and expert-annotated images are included here to spur further advancement. These results illustrate that computer vision can enable rapid and accurate quantitation of RBC morphology, which may provide value in both clinical and research contexts.

Combining imaging flow cytometry and machine learning for high-throughput schistocyte quantification: A SVM classifier development and external validation cohort

Background

Schistocyte counts are a cornerstone of the diagnosis of thrombotic microangiopathy syndrome (TMA). Their manual quantification is complex and alternative automated methods suffer from pitfalls that limit their use. We report a method combining imaging flow cytometry (IFC) and artificial intelligence for the direct label-free and operator-independent quantification of schistocytes in whole blood.

Methods

We used 135,045 IFC images from blood acquisition among 14 patients to extract 188 features with IDEAS® software and 128 features from a convolutional neural network (CNN) with Keras framework in order to train a support vector machine (SVM) blood elements’ classifier used for schistocytes quantification.

Finding

Keras features showed better accuracy (94.03%, CI: 93.75-94.31%) than ideas features (91.54%, CI: 91.21-91.87%) in recognising whole-blood elements, and together they showed the best accuracy (95.64%, CI: 95.39-95.88%). We obtained an excellent correlation (0.93, CI: 0.90-0.96) between three haematologists and our method on a cohort of 102 patient samples. All patients with schistocytosis (>1% schistocytes) were detected with excellent specificity (91.3%, CI: 82.0-96.7%) and sensitivity (100%, CI: 89.4-100.0%). We confirmed these results with a similar specificity (91.1%, CI: 78.8-97.5%) and sensitivity (100%, CI: 88.1-100.0%) on a validation cohort (n=74) analysed in an independent healthcare centre. Simultaneous analysis of 16 samples in both study centres showed a very good correlation between the 2 imaging flow cytometers (Y=1.001x).

Interpretation

We demonstrate that IFC can represent a reliable tool for operator-independent schistocyte quantification with no pre-analytical processing which is of most importance in emergency situations such as TMA.

Funding

None.

2021 update of the 2012 ICSH Recommendations for identification, diagnostic value, and quantitation of schistocytes: Impact and revisions

In 2012, the International Council for Standardization in Hematology (ICSH) published recommendations for the identification, quantitation, and diagnostic value of schistocytes. In the present review, the impact of these recommendations is evaluated. This work is based on citations in peer-reviewed papers published since 2012. The first 2012 ICSH Recommendations have also been revised to incorporate newly published data in the literature and current best laboratory practice. Recommended reference ranges have been proposed for healthy adults and full-term neonates of 1% or less schistocytes. More than 1% of morphologically identified schistocytes on the blood film are considered suspicious for thrombotic microangiopathy. For preterm infants, a normal level of 5% or less is recommended. The fragment red cell count (FRC) generated by some automated hematological analyzers provides a valuable screening tool for the presence of schistocytes. Specifically, the absence of FRCs can be used as a valuable parameter to exclude the presence of schistocytes on the blood film. The validity and usefulness of microscope schistocytes and automated FRCs, respectively, are discussed in the context of the laboratory diagnostic tests used for thrombotic microangiopathies.

Machine learning and augmented human intelligence use in histomorphology for haematolymphoid disorders

Advances in digital pathology have allowed a number of opportunities such as decision support using artificial intelligence (AI). The application of AI to digital pathology data shows promise as an aid for pathologists in the diagnosis of haematological disorders. AI-based applications have embraced benign haematology, diagnosing leukaemia and lymphoma, as well as ancillary testing modalities including flow cytometry. In this review, we highlight the progress made to date in machine learning applications in haematopathology, summarise important studies in this field, and highlight key limitations. We further present our outlook on the future direction and trends for AI to support diagnostic decisions in haematopathology.

Differential diagnosis of thalassemia and iron deficiency anemia using the CellaVision Advanced Red Blood Cell software

INTRODUCTION

Red blood cells (RBCs) in patients with thalassemia and iron deficiency anemia (IDA) exhibit different patterns of morphological changes. However, manual quantitative analysis of the morphological changes in the RBCs is time-consuming and subjective, limiting its use in differential diagnosis. The aim of this study was to evaluate the CellaVision Advanced RBC software as a prescreening tool for differential diagnosis of thalassemia and IDA.

METHODS

The study cohort consisted of 54 thalassemia and 46 IDA cases in the training group and 36 thalassemia and 31 IDA patients in the validation group. The CellaVision DM96 Advanced RBC software was used to analyze the RBC morphology.

RESULTS AND CONCLUSION

Specific patterns of quantitative changes in RBC shapes were found in thalassemia and IDA patients. As a single parameter, target cell was the best morphological cell type to distinguish thalassemia from IDA, with an area under the curve (AUC) of 0.79, followed by hypochromatic cells with an AUC of 0.70. Combination of target and hypochromatic cells expressed as a ratio of the percentage of target cells to percentage of hypochromatic cells (T/H ratio) presented better differential diagnostic ability with an AUC of 0.88. A cutoff value of 1.755 for T/H ratio showed a sensitivity and specificity of 80.43% and 81.48% in the training group and 88.89% and 80.65% in the validation group, respectively. Assessment of the T/H ratio using the CellaVision Advanced RBC software represents a relatively simple and economical screening procedure for diagnostic testing of thalassemia and IDA.

The ADVIA2120i parameter Revised %MICRO is a surrogate marker of schistocyte formation after hematopoietic stem cell transplantation

Objectives : Hematopoietic stem cell transplantation (HSCT)-associated thrombotic microangiopathy (TA-TMA) is an important early post-treatment condition. This study evaluated the Revised %MICRO, a parameter obtained from the ADVIA 2120i automated blood cell counter, as a surrogate marker of the schistocyte ratio. We hypothesized that individual differences between the %MICRO value and schistocyte ratio would remain constant. Design and Methods: EDTA-2K-treated peripheral blood samples were collected from 19 patients who underwent allogeneic HSCT from April 2014 to September 2018. First, the baseline difference, X, was calculated using a sample from the first day after HSCT as X = %MICRO (first day) - schistocyte ratio (first day). Next, the Revised %MICRO for each subsequent day was calculated as Revised %MICRO = %MICRO - X. We evaluated correlations of the schistocyte ratio with the calculated %MICRO and Revised %MICRO and the RBC fragment, RBC distribution width, %MICRO and Revised %MICRO data obtained from the ADVIA 2120i. Results : The mean schistocyte percentage and Revised %MICRO were both 0.4% ± 0.6. RBC fragments correlated weakly with the %MICRO and schistocyte ratio, respectively (r = 0.162 and r = 0.771, respectively), whereas the Revised %MICRO correlated strongly with the schistocyte ratio (r = 0.893). Conclusion : The Revised %MICRO appears to be a good surrogate of the schistocyte ratio in a clinical setting. J. Med. Invest. 67 : 250-254, August, 2020.

Evaluation of the CellaVision Advanced RBC Application for Detecting Red Blood Cell Morphological Abnormalities

BACKGROUND

The Advanced RBC Application of the CellaVision DM9600 system (CellaVision AB, Lund, Sweden) automatically characterizes and classifies red blood cells (RBCs) into 21 morphological categories based on their size, color, shape, and inclusions. We evaluated the diagnostic performance of the CellaVision Advanced RBC Application with respect to the classification and grading of RBC morphological abnormalities in accordance with the 2015 International Council for Standardization in Haematology (ICSH) guidelines.

METHODS

A total of 223 samples, including 123 with RBC morphological abnormalities and 100 from healthy controls, were included. Seven RBC morphological abnormalities and their grading obtained with CellaVision DM9600 pre- and post-classification were compared with the results obtained using manual microscopic examination. The grading cut-off percentages were determined in accordance with the 2015 ICSH guidelines. The sensitivity and specificity of the CellaVision DM9600 system were evaluated using the manual microscopic examination results as a true positive.

RESULTS

In pre-classification, >90% sensitivity was observed for target cells, tear drop cells, and schistocytes, while >90% specificity was observed for acanthocytes, spherocytes, target cells, and tear drop cells. In post-classification, the detection sensitivity and specificity of most RBC morphological abnormalities increased, except for schistocytes (sensitivity) and acanthocytes (specificity). The grade agreement rates ranged from 35.9% (echinocytes) to 89.7% (spherocytes) in pre-classification and from 46.2% (echinocytes) to 90.1% (spherocytes) in post-classification. The agreement rate of samples with within-one grade difference exceeded 90% in most categories, except for schistocytes and echinocytes.

CONCLUSIONS

The Advanced RBC Application of CellaVision DM9600 is a valuable screening tool for detecting RBC morphological abnormalities.

Artificial Intelligence and Digital Microscopy Applications in Diagnostic Hematopathology

Digital Pathology is the process of converting histology glass slides to digital images using sophisticated computerized technology to facilitate acquisition, evaluation, storage, and portability of histologic information. By its nature, digitization of analog histology data renders it amenable to analysis using deep learning/artificial intelligence (DL/AI) techniques. The application of DL/AI to digital pathology data holds promise, even if the scope of use cases and regulatory framework for deploying such applications in the clinical environment remains in the early stages. Recent studies using whole-slide images and DL/AI to detect histologic abnormalities in general and cancer in particular have shown encouraging results. In this review, we focus on these emerging technologies intended for use in diagnostic hematology and the evaluation of lymphoproliferative diseases.

Performance of automated digital cell imaging analyzer Sysmex DI-60

BACKGROUND

The Sysmex DI-60 system (DI-60, Sysmex, Kobe, Japan) is a new automated digital cell imaging analyzer. We explored the performance of DI-60 in comparison with Sysmex XN analyzer (XN, Sysmex) and manual count.

METHODS

In a total of 276 samples (176 abnormal and 100 normal samples), white blood cell (WBC) differentials, red blood cell (RBC) classification and platelet (PLT) estimation by DI-60 were compared with the results by XN and/or manual count. RBC morphology between pre-classification and verification was compared according to the ICSH grading criteria. The manual count was performed according to the Clinical and Laboratory Standards Institute guidelines (H20-A2).

RESULTS

The overall concordance between DI-60 and manual count for WBCs was 86.0%. The agreement between DI-60 pre-classification and verification was excellent (weighted κ=0.963) for WBC five-part differentials. The correlation with manual count was very strong for neutrophils (r=0.955), lymphocytes (r=0.871), immature granulocytes (r=0.820), and blasts (r=0.879). RBC grading showed notable differences between DI-60 and manual counting on the basis of the ICSH grading criteria. Platelet count by DI-60 highly correlated with that by XN (r=0.945). However, DI-60 underestimated platelet counts in samples with marked thrombocytosis.

CONCLUSIONS

The performance of DI-60 for WBC differential, RBC classification, and platelet estimation seems to be acceptable even in abnormal samples with improvement after verification. DI-60 would help optimize the workflow in hematology laboratory with reduced manual workload.