Deep Learning‐Based Blood Abnormalities Detection as a Tool for VEXAS Syndrome Screening
Cédric de Almeida Braga
(1)
,
Maxence Bauvais
(2)
,
Pierre Sujobert
(3, 4, 5)
,
Maël Heiblig
(4, 6, 7)
,
Maxime Jullien
(8, 9, 10)
,
Baptiste Le Calvez
(11, 12, 13)
,
Camille Richard
,
Valentin Le Roc'H
,
Emmanuelle Rault
(14)
,
Olivier Hérault
(15, 16)
,
Pierre Peterlin
(13, 8)
,
Alice Garnier
(13, 8)
,
Patrice Chevallier
(13)
,
Simon Bouzy
(17)
,
Yannick Le Bris
(13)
,
Antoine Néel
(18, 19, 20)
,
Julie Graveleau
(21)
,
Olivier Kosmider
(22, 23)
,
Perrine Paul-Gilloteaux
(24)
,
Nicolas Normand
(25, 1)
,
Marion Eveillard
1
LS2N - équipe IPI -
Image Perception Interaction
2 Nantes Univ - UFR Pharmacie - Nantes Université - UFR des Sciences Pharmaceutiques et Biologiques
3 CIRI - Centre International de Recherche en Infectiologie
4 UCBL - Université Claude Bernard Lyon 1
5 CHLS - Centre Hospitalier Lyon Sud [CHU - HCL]
6 HCL - Hospices Civils de Lyon
7 UNICANCER/CRCL - Centre de Recherche en Cancérologie de Lyon
8 CRCI2NA - Centre de Recherche en Cancérologie et Immunologie Intégrée Nantes-Angers
9 LabEX IGO Immunothérapie Grand Ouest
10 EFS Pays de la Loire - Etablissement Français du Sang [Pays de la Loire]
11 Département d'Hématologie [CHU Nantes]
12 Département d'Oncologie Pédiatrique [CHU Nantes]
13 CHU Nantes - Centre Hospitalier Universitaire de Nantes = Nantes University Hospital
14 Service d'Hématologie biologique [Tours]
15 CHRU Tours - Centre Hospitalier Régional Universitaire de Tours
16 UT - Université de Tours
17 Centre Hospitalier Universitaire de Rennes [CHU Rennes] = Rennes University Hospital [Pontchaillou]
18 Service de Médecine Interne, CHU Nantes, Nantes, France. - Service de Médecine Interne, CHU Nantes, Nantes, France.
19 U1064 Inserm - CR2TI - Centre de Recherche en Transplantation et Immunologie - Center for Research in Transplantation and Translational Immunology
20 Team 4 - U1064 Inserm - CR2TI - Team 4 : Deciphering organ immune regulation in inflammation and transplantation (DORI-t)
21 Centre hospitalier de Saint-Nazaire
22 IC UM3 (UMR 8104 / U1016) - Institut Cochin
23 Hôpital Cochin [AP-HP]
24 BioCore
25 LS2N - Laboratoire des Sciences du Numérique de Nantes
2 Nantes Univ - UFR Pharmacie - Nantes Université - UFR des Sciences Pharmaceutiques et Biologiques
3 CIRI - Centre International de Recherche en Infectiologie
4 UCBL - Université Claude Bernard Lyon 1
5 CHLS - Centre Hospitalier Lyon Sud [CHU - HCL]
6 HCL - Hospices Civils de Lyon
7 UNICANCER/CRCL - Centre de Recherche en Cancérologie de Lyon
8 CRCI2NA - Centre de Recherche en Cancérologie et Immunologie Intégrée Nantes-Angers
9 LabEX IGO Immunothérapie Grand Ouest
10 EFS Pays de la Loire - Etablissement Français du Sang [Pays de la Loire]
11 Département d'Hématologie [CHU Nantes]
12 Département d'Oncologie Pédiatrique [CHU Nantes]
13 CHU Nantes - Centre Hospitalier Universitaire de Nantes = Nantes University Hospital
14 Service d'Hématologie biologique [Tours]
15 CHRU Tours - Centre Hospitalier Régional Universitaire de Tours
16 UT - Université de Tours
17 Centre Hospitalier Universitaire de Rennes [CHU Rennes] = Rennes University Hospital [Pontchaillou]
18 Service de Médecine Interne, CHU Nantes, Nantes, France. - Service de Médecine Interne, CHU Nantes, Nantes, France.
19 U1064 Inserm - CR2TI - Centre de Recherche en Transplantation et Immunologie - Center for Research in Transplantation and Translational Immunology
20 Team 4 - U1064 Inserm - CR2TI - Team 4 : Deciphering organ immune regulation in inflammation and transplantation (DORI-t)
21 Centre hospitalier de Saint-Nazaire
22 IC UM3 (UMR 8104 / U1016) - Institut Cochin
23 Hôpital Cochin [AP-HP]
24 BioCore
25 LS2N - Laboratoire des Sciences du Numérique de Nantes
Maël Heiblig
- Fonction : Auteur
- PersonId : 1256955
- ORCID : 0000-0003-1682-8657
Maxime Jullien
- Fonction : Auteur
- PersonId : 808174
- ORCID : 0000-0001-6915-6570
Baptiste Le Calvez
- Fonction : Auteur
- PersonId : 1367955
- ORCID : 0000-0001-5148-3792
Camille Richard
- Fonction : Auteur
Valentin Le Roc'H
- Fonction : Auteur
Patrice Chevallier
- Fonction : Auteur
- PersonId : 762210
- ORCID : 0000-0003-3142-5581
Yannick Le Bris
- Fonction : Auteur
- PersonId : 818803
- ORCID : 0000-0002-0095-6999
Antoine Néel
- Fonction : Auteur
Perrine Paul-Gilloteaux
- Fonction : Auteur
- PersonId : 176916
- IdHAL : perrine-paul-gilloteaux
- ORCID : 0000-0002-4822-165X
- IdRef : 236865714
Marion Eveillard
- Fonction : Auteur
- PersonId : 781636
- ORCID : 0000-0001-7206-5135
Résumé
ABSTRACT Introduction VEXAS 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. Methods A 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. Results Significant 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. Conclusion This 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.