-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.

190 A Computer-Aided Diagnosis Tool for the Detection of Hemarthrosis By Remote Joint Ultrasound in Patients with Hemophilia

Program: Oral and Poster Abstracts
Type: Oral
Session: 322. Disorders of Coagulation or Fibrinolysis: Clinical and Epidemiological: Looking Forward: Novel Therapies and Diagnostic Modalities for Bleeding Disorders II
Hematology Disease Topics & Pathways:
Bleeding and Clotting, artificial intelligence (AI), Research, Clinical Research, Diseases, Technology and Procedures, machine learning
Saturday, December 10, 2022: 2:45 PM

Roberta Gualtierotti1,2*, Sara Arcudi, MD1,2*, Alessandro Ciavarella, MD1,2*, Marco Colussi, marco.colussi@unimi.it3*, Sergio Mascetti, sergio.mascetti@unimi.it3*, Claudio Bettini, claudio.bettini@unimi.it3* and Flora Peyvandi, MD1,2

1Department of Pathophysiology and Transplantation, Università degli Studi di Milano, Milan, Italy
2Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy
3Department of Computer Science, Università degli Studi di Milano, Milan, Italy

Background: Musculoskeletal ultrasound (MSK-US) is a non-invasive and easily accessible diagnostic tool for joint health assessment of patients with hemophilia. The early identification of hemarthrosis is pivotal and could be achieved by a telemedicine system where MSK-US is performed by general practitioners or the patients themselves. The images so collected could be sent to the Comprehensive Care Center clinicians. Due to the great number of images collected, a computer-aided diagnosis (CAD) system for the automatic detection of joint effusion could support the physicians in prioritizing interventions.

Aims: our aim is to develop a telemedicine system for the remote monitoring of joint health. The first step was to assess the feasibility of a CAD system by developing a deep-learning algorithm to automatically recognize joint capsule distension as a proxy of hemarthrosis in MSK-US images.

Methods: images of the longitudinal scan of the subquadricipital recess of the knee at 30°-flexion were collected and labeled by an expert MSK-US operator. We chose the knee as it is an easily accessible joint even as self-assessment. The learning algorithm is based on an object detection framework that is trained to detect the normal and the distended joint recesses.

Results: we recruited two-hundred consecutive adult patients with hemophilia referring to our Center from October 2020 to December 2021, aged 44.7 ± 18.6 years and 50 sex- and age-matched healthy controls. A total of 8,634 MSK-US images were collected (2,267 knee scans). Of these, 483 were considered valid and used for the learning task. We used 330 images for training and 120 images for testing the classifier, leading to the detection of joint recess distensions with an 78% accuracy (69% sensitivity; 87% specificity). In addition, two volunteer patients were asked to perform an ultrasound examination of knees, elbows and ankles with a portable ultrasound probe connected with a smartphone after a brief training. The two patients accepted the brief training on how to use the portable probe and how to perform the ultrasound by themselves. Some difficulties were reported during the self-assessment of the elbows and ankles, which could be performed by a caregiver.

Conclusions: the CAD system for the automatic detection of joint recess distension in patients with hemophilia developed by our group is feasible. Performance may further improve with a larger training set. The MSK-US self-assessment was accepted by the two volunteer patients and their indications will be followed for the development of a telemedicine system including elbows, knees and ankles as the most commonly affected joints in hemophilic arthropathy. The telemedicine system will prompt early detection of hemarthrosis in patients with hemophilia, thus supporting clinicians in providing early personalized management with the most appropriate therapeutic intervention.

Disclosures: Gualtierotti: Roche: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Takeda: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Bayer: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfiser: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novo Nordisk: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Ciavarella: Bayer: Consultancy, Other: Bayer Hemophilia Awards Fellowship. Peyvandi: Sanofi: Honoraria; Sobi: Honoraria; Takeda: Honoraria; Roche: Honoraria; BioMarin: Honoraria; Grifols: Honoraria.

*signifies non-member of ASH