-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

3648 Deep Learning Based Blood Abnormalities Detection As a Tool for Vexas Syndrome Screening

Program: Oral and Poster Abstracts
Session: 803. Emerging Tools, Techniques and Artificial Intelligence in Hematology: Poster II
Hematology Disease Topics & Pathways:
Research, artificial intelligence (AI), Translational Research, Diseases, Myeloid Malignancies, Technology and Procedures
Sunday, December 10, 2023, 6:00 PM-8:00 PM

Cedric De Almeida Braga, Ph D in training1*, Maxence Bauvais2*, Pierre Sujobert, MD, PhD3*, Mael Heiblig4*, Maxime Jullien, MD5*, Baptiste Le Calvez6*, Camille Richard7*, Valentin Le Roc'h7*, Emmanuelle Rault8*, Olivier Herault, MD9*, Pierre Peterlin10*, Alice Garnier, MD11*, Patrice Chevallier, MD12, Simon Bouzy, PharmD13*, Yannick LE Bris, PhD, PharmD14*, Antoine Neel, Pr15*, Julie Graveleau16*, Perrine Paul-Gilloteaux, Ph D17*, Olivier Kosmider, PharmD, PhD18*, Nicolas Normand19* and Marion Eveillard20,21*

1Nantes Université, Ecole Centrale Nantes, LS2N, UMR 6004, Nantes, France, Nantes, France
2Nantes University Hospital, Hematology biology, Nantes, France
3Laboratory of Hematology, Centre Hospitalier Lyon Sud, Pierre Bénite, France
4Hematology Department, Hôpital Lyon Sud, Vourles, Rhône Alpes, France
5Nantes University Hospital, NANTES, France
6Pediatric Hematology, Nantes, France
7Hematology biology, Nantes, France
8CHU Tours, Tours, France
9Centre Hospitalier Universitaire, Tours, FRA
10Hematology Department, Hôpital Hotel Dieu, NANTES CEDEX 1, France
11Hematology clinic, Nantes University Hospital, Nantes, France
12Service d'hématologie, CHU de Nantes, Nantes, France
13Hematology biology, Nantes University Hospital, Nantes, France
14CHU Nantes, Nantes, France
15Service de Médecine Interne-Médecine Vasculaire, CHU de Nantes, Nantes, France
16Saint Nazaire Hospital, Saint Nazaire, France
17Nantes University, MicroPICell, INSERM, Nantes, France
18Laboratory of Hematology, Université Paris Cité and Assistance Publique-Hôpitaux de Paris. Centre, Hôpital Cochin, Paris, France
19Polytech, Nantes Université, école centrale Nantes, CNRS, LS2N, UMR 6004, Nantes, France
20Hematology biology, CHU Nantes, Nantes, France
21CRCI²NA, INSERM, Nantes, France


The VEXAS syndrome (vacuoles, E1 enzyme, X-linked, auto-inflammatory, somatic) described in 2020 caused by mutations of the UBA1 gene, displayed a large pleomorphic array of clinical and biological features. Nevertheless, these criteria do not allow to discriminate VEXAS from other myeloid malignancies on complete blood count + differential, notably due to the absence of peripheral blood characterization of the disease.

This study aimed at singling out dysplastic features indicative of VEXAS among peripheral blood (PB) polymorphonuclears (PMN) from VEXAS patients compared to healthy patients and myelodysplastic (MDS) patients. As such task is tedious and subject to operator bias, a multicentric dataset has been used to design a deep learning algorithm for automatic detection of these features, finally tested on an external validation cohort.

Patient, material and methods

Patients written consents were obtained and five academic centers (number from 1 to 5) were involved in this 3 steps study.

Firstly, 3 academic centers (number 1 to 3) enrolled 20 patients distributed as follow: 9 patients with UBA1mut and 6 patients UBA1wt with no other genetic mutation as well as 5 samples from MDS patients. A total of 25 PB smears and especially PMN images were gathered and screened for various abnormalities. Two independent morphologists blindly quantified PMN predefined characteristics. This dataset of 2,824 multilabelled PMN was evaluated and tested by a two-sample Wilcoxon Rank Sum test for statistical significance between UBA1 mut and the others.

Secondly, in order to automate the detection of these significant features on PMN images, a convolutional neural network (CNN) was trained using a multicentric image dataset gathered from 4 academic centers (number 2 to 5). Patients were selected based on clinical and biological symptoms suggesting VEXAS syndrome, namely fever, skin lesions, chondrites, vasculitis and/or anemia. After evaluation of the UBA1 mutation status, this multicentric cohort was separated into confirmed VEXAS (n = 19) and UBA1WT (n=20). A total of 6,615 annotated PMN images were collected from DI systems (Cellavision ®, Lund, Sweden) with 1 to 12 PB smears per patient, yielding a 48 UBA1mut and 33 UBA1wt smears. All patients were males, the median age was 74 year-old (IQR 67-77) and that of control patients 72 year- old (61-79).

Finally, the CNN was evaluated on an external cohort from another academic hospital (center 1), including inflammatory patients, and patients screened for VEXAS syndrome for a total cohort of 15 patients and 1,887 images.


Four specific abnomalities were observed in PMNs from VEXAS patients as compared with healthy or myelodysplastic controls. The automatic recognition of anomalies was then cast in a multilabel classification task, where each PMN image could be assigned to one or more labels during interference. In order to take into account this specificity, a CNN was trained, composed of different layers of convolutions and pooling (Figure1).

The model was set-up to output an independent binary prediction for the presence or absence of each of the four anomalies of interest.

Automatic detection of these 4 anomalies by the proposed model yielded area under the curve (AUC) of 0.827; 0.837; 0.927 and 0.947 (Figure 2). Regarding the general performance on the external validation cohort, the CNN achieved a Hamming loss of 0.141, and macro and micro F1 scores of 0.588 and 0.668 respectively.


This study suggests that computer–assisted analysis of PB smears, focusing on suspected VEXAS cases, could provide valuable insights to determine which patients should undergo molecular testing.

A deep-learning approach leveraging previously identified peripheral blood indicators and automatic analyzers is thus presented, which can help hematologists orient their suspicion before initiating further analysis.

Disclosures: Heiblig: Pfizer Inc.: Honoraria; Astellas: Honoraria; AbbVie: Honoraria; Jazz Pharmaceuticals: Honoraria; Servier: Honoraria. Chevallier: Sanofi: Honoraria; Mallinckrodt Pharmaceuticals: Honoraria; Incyte: Honoraria, Research Funding; Takeda: Honoraria; Immedica Pharma: Honoraria; Servier: Honoraria.

*signifies non-member of ASH