Deep learning-based differentiation of peripheral high-flow and low-flow vascular malformations in T2-weighted short tau inversion recovery MRI

Article type: Research Article

Authors: Hammer, Simone^{a; *} | Nunes, Danilo Weber^b | Hammer, Michael^b | Zeman, Florian^c | Akers, Michael^a | Götz, Andrea^a | Balla, Annika^a | Doppler, Michael Christian^d | Fellner, Claudia^a | Platz Batista da Silva, Natascha^a | Thurn, Sylvia^a | Verloh, Niklas^d | Stroszczynski, Christian^a | Wohlgemuth, Walter Alexander^e | Palm, Christoph^{b; f} | Uller, Wibke^d

Affiliations: [a] Department of Radiology, Faculty of Medicine, Medical Center University of Regensburg, University of Regensburg, Regensburg, Germany | [b] Regensburg Medical Image Computing (ReMIC), Ostbayerische Technische Hochschule Regensburg (OTH Regensburg), Regensburg, Germany | [c] Faculty of Medicine, Center for Clinical Trials, Medical Center University of Regensburg, University of Regensburg, Regensburg, Germany | [d] Department of Diagnostic and Interventional Radiology, Faculty of Medicine, Medical Center University of Freiburg, University of Freiburg, Freiburg, Germany | [e] Department of Radiology, Faculty of Medicine, Medical Center University of Halle (Saale), University of Halle (Saale), Halle, Germany | [f] Regensburg Center of Biomedical Engineering (RCBE), OTH Regensburg and University of Regensburg, Regensburg, Germany

Correspondence: [*] Corresponding author: Dr. Simone Hammer, Department of Radiology, Medical Center University of Regensburg, Faculty of Medicine, University of Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany. E-mail: [email protected].

Abstract: BACKGROUND: Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease. OBJECTIVE: A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI. METHODS: 527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN’s diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN’s prediction on the radiologists’ performance and diagnostic certainty was evaluated. 3) Junior radiologists’ performance after self-training was compared with that of the CNN. RESULTS: Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p < 0.001; 92% vs. 77%, p < 0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists’ specificity and accuracy improved and were comparable to that of the CNN. CONCLUSIONS:Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective.

Keywords: Vascular malformation, deep learning, magnetic resonance imaging

DOI: 10.3233/CH-232071

Journal: Clinical Hemorheology and Microcirculation, vol. 87, no. 2, pp. 221-235, 2024