Assembly and evaluation of an inventory of guidelines that are available to support clinical hematology laboratory practice

Prediction of SARS-CoV-2-positivity from million-scale complete blood counts using machine learning

Background

The Complete Blood Count (CBC) is a commonly used low-cost test that measures white blood cells, red blood cells, and platelets in a person’s blood. It is a useful tool to support medical decisions, as intrinsic variations of each analyte bring relevant insights regarding potential diseases. In this study, we aimed at developing machine learning models for COVID-19 diagnosis through CBCs, unlocking the predictive power of non-linear relationships between multiple blood analytes.

Methods

We collected 809,254 CBCs and 1,088,385 RT-PCR tests for SARS-Cov-2, of which 21% (234,466) were positive, from 900,220 unique individuals. To properly screen COVID-19, we also collected 120,807 CBCs of 16,940 individuals who tested positive for other respiratory viruses. We proposed an ensemble procedure that combines machine learning models for different respiratory infections and analyzed the results in both the first and second waves of COVID-19 cases in Brazil.

Results

We obtain a high-performance AUROC of 90 + % for validations in both scenarios. We show that models built solely of SARS-Cov-2 data are biased, performing poorly in the presence of infections due to other RNA respiratory viruses.

Conclusions

We demonstrate the potential of a novel machine learning approach for COVID-19 diagnosis based on a CBC and show that aggregating information about other respiratory diseases was essential to guarantee robustness in the results. Given its versatile nature, low cost, and speed, we believe that our tool can be particularly useful in a variety of scenarios—both during the pandemic and after.

The complete blood count (CBC) is a medical laboratory test that provides information about cells in a person’s blood and is extensively used to support medical decisions. This study explored the ability of a computer-based approach to automatically identify active COVID-19 infections by using CBC exams. We collected a large dataset with over one million CBC exams and the matching tests currently used to detect SARS-Cov-2 or other respiratory viruses. Our results demonstrate both the potential of this approach for diagnosing SARS-Cov-2 infection by using only CBC data, and also that considering information about other respiratory diseases in the methodology is essential to guarantee that results can be trusted. This automated computational approach can be useful in a variety of contexts during the COVID-19 pandemic and after since it is fast, low-cost, and versatile.

Zuin et al. use a large dataset of blood count exams to predict SARS-CoV-2 PCR results with machine learning. The model performs well and is superior to those that do not take into account infection with other RNA respiratory viruses.

Update on diagnostic testing for platelet function disorders: What is practical and useful?

INTRODUCTION

Platelet function disorders (PFD) are an important group of bleeding disorders that require validated and practical laboratory strategies for diagnosis.

METHODS

This review summarizes the authors' experiences, current literature, and an international survey to evaluate the practices of diagnostic laboratories that offer tests for PFD.

RESULTS

Blood counts, blood film review, and aggregation tests are the most commonly performed investigations for PFD and help determine whether there is thrombocytopenia and/or defective platelet function due to a variety of causes. The performance characteristics of tests for PFD, and the level of evidence that these tests detect bleeding problems, are important issues to determine where tests are useful for diagnostic or correlative purposes, or research only uses. Platelet aggregation assays, and quantitative analysis of platelet dense granule numbers, are tests with good performance characteristics that detect abnormalities associated with increased bleeding in a significant proportion of individuals referred for PFD investigations. Lumiaggregometry estimates of platelet adenosine triphosphate release show greater variability which limits the diagnostic usefulness. Diagnostic laboratories report that fiscal and other constraints, including a lack of high-quality evidence, limit their ability to offer an expanded test menu for PFD.

CONCLUSION

PFD are clinically important bleeding disorders that remain challenging for diagnostic laboratories to investigate. While some PFD tests are well validated for diagnostic purposes, gaps in scientific evidence and resource limitations influence diagnostic laboratory decisions on which PFD tests to offer.

Appropriateness of diagnostics tests

The evolution of the concept of 'appropriateness', in the three past decades, from 'no harm' and 'no waste' to 'medical decision-making' and 'determining outcomes' highlights two main points: its foundation is evidence-based medicine, and it is a quality of every phase of the total testing process, not only for the selection of tests. Nevertheless, appropriateness in Laboratory Hematology, as well as in Laboratory Medicine, is an elusive concept: 'Appropriateness' interplays with 'patient's safety', 'healthcare costs', 'clinical decision-making', and 'effectiveness', and the criteria for appropriateness, mainly adherence to clinical guidelines, are often not evidence-based and not always consensus-based. Moreover, practising appropriateness is a complex issue because of the ambiguity of the criteria and targets, the never-ending work of implementing guidelines and their audit, and the uniqueness of the clinical situation of the individual patient. Authors agree on some practical rules: establishing a multidisciplinary and multiprofessional team, choosing carefully clinical targets, finding or building evidences, sharing guidelines with clinicians, choosing adequate tools for changing, working hard on implementation, identifying the 'right' laboratory methods and processes, checking progress indefinitely, providing information, interpretations, and consultations, and promoting feedback and audits. The success depends on the 'right' combination of educational, operative, and reinforcing interventions. Competences in organization, in implementation science, and in interpersonal relationship management are essential as well as knowledge and experience in Hematology, not only in Laboratory Hematology.