Cédric De Almeida Braga, Maxence Bauvais, Pierre Sujobert, Maël Heiblig, Maxime Jullien, Baptiste Le Calvez, Camille Richard, Valentin Le Roc'h, Emmanuelle Rault, Olivier Hérault, Pierre Peterlin, Alice Garnier, Patrice Chevallier, Simon Bouzy, Yannick Le Bris, Antoine Néel, Julie Graveleau, Olivier Kosmider, Perrine Paul‐Gilloteaux, Nicolas Normand, Marion Eveillard
{"title":"Deep Learning‐Based Blood Abnormalities Detection as a Tool for VEXAS Syndrome Screening","authors":"Cédric De Almeida Braga, Maxence Bauvais, Pierre Sujobert, Maël Heiblig, Maxime Jullien, Baptiste Le Calvez, Camille Richard, Valentin Le Roc'h, Emmanuelle Rault, Olivier Hérault, Pierre Peterlin, Alice Garnier, Patrice Chevallier, Simon Bouzy, Yannick Le Bris, Antoine Néel, Julie Graveleau, Olivier Kosmider, Perrine Paul‐Gilloteaux, Nicolas Normand, Marion Eveillard","doi":"10.1111/ijlh.14368","DOIUrl":null,"url":null,"abstract":"IntroductionVEXAS is a syndrome described in 2020, caused by mutations of the <jats:italic>UBA1</jats:italic> gene, and displaying a large pleomorphic array of clinical and hematological features. Nevertheless, these criteria lack significance to discriminate VEXAS from other inflammatory conditions at the screening step. This work hence first focused on singling out dysplastic features indicative of the syndrome among peripheral blood (PB) polymorphonuclears (PMN). A deep learning algorithm is then proposed for automatic detection of these features.MethodsA multicentric dataset, comprising 9514 annotated PMN images was gathered, including <jats:italic>UBA1</jats:italic> mutated VEXAS (<jats:italic>n</jats:italic> = 25), <jats:italic>UBA1</jats:italic> wildtype myelodysplastic (<jats:italic>n</jats:italic> = 14), and <jats:italic>UBA1</jats:italic> wildtype cytopenic patients (<jats:italic>n</jats:italic> = 25). Statistical analysis on a subset of patients was performed to screen for significant abnormalities. Detection of these features on PB was then automated with a convolutional neural network (CNN) for multilabel classification.ResultsSignificant differences were observed in the proportions of PMNs with pseudo‐Pelger, nuclear spikes, vacuoles, and hypogranularity between patients with VEXAS and both cytopenic and myelodysplastic controls.Automatic detection of these abnormalities yielded AUCs in the range [0.85–0.97] and a F1‐score of 0.70 on the test set. A VEXAS screening score was proposed, leveraging the model outputs and predicting the <jats:italic>UBA1</jats:italic> mutational status with 0.82 sensitivity and 0.71 specificity on the test patients.ConclusionThis study suggests that computer‐assisted analysis of PB smears, focusing on suspected VEXAS cases, can provide valuable insights for determining which patients should undergo molecular testing. The presented deep learning approach can help hematologists direct their suspicions before initiating further analyses.","PeriodicalId":14120,"journal":{"name":"International Journal of Laboratory Hematology","volume":"25 1","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2024-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Laboratory Hematology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1111/ijlh.14368","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"HEMATOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
IntroductionVEXAS is a syndrome described in 2020, caused by mutations of the UBA1 gene, and displaying a large pleomorphic array of clinical and hematological features. Nevertheless, these criteria lack significance to discriminate VEXAS from other inflammatory conditions at the screening step. This work hence first focused on singling out dysplastic features indicative of the syndrome among peripheral blood (PB) polymorphonuclears (PMN). A deep learning algorithm is then proposed for automatic detection of these features.MethodsA multicentric dataset, comprising 9514 annotated PMN images was gathered, including UBA1 mutated VEXAS (n = 25), UBA1 wildtype myelodysplastic (n = 14), and UBA1 wildtype cytopenic patients (n = 25). Statistical analysis on a subset of patients was performed to screen for significant abnormalities. Detection of these features on PB was then automated with a convolutional neural network (CNN) for multilabel classification.ResultsSignificant differences were observed in the proportions of PMNs with pseudo‐Pelger, nuclear spikes, vacuoles, and hypogranularity between patients with VEXAS and both cytopenic and myelodysplastic controls.Automatic detection of these abnormalities yielded AUCs in the range [0.85–0.97] and a F1‐score of 0.70 on the test set. A VEXAS screening score was proposed, leveraging the model outputs and predicting the UBA1 mutational status with 0.82 sensitivity and 0.71 specificity on the test patients.ConclusionThis study suggests that computer‐assisted analysis of PB smears, focusing on suspected VEXAS cases, can provide valuable insights for determining which patients should undergo molecular testing. The presented deep learning approach can help hematologists direct their suspicions before initiating further analyses.
期刊介绍:
The International Journal of Laboratory Hematology provides a forum for the communication of new developments, research topics and the practice of laboratory haematology.
The journal publishes invited reviews, full length original articles, and correspondence.
The International Journal of Laboratory Hematology is the official journal of the International Society for Laboratory Hematology, which addresses the following sub-disciplines: cellular analysis, flow cytometry, haemostasis and thrombosis, molecular diagnostics, haematology informatics, haemoglobinopathies, point of care testing, standards and guidelines.
The journal was launched in 2006 as the successor to Clinical and Laboratory Hematology, which was first published in 1979. An active and positive editorial policy ensures that work of a high scientific standard is reported, in order to bridge the gap between practical and academic aspects of laboratory haematology.