Abstract: Deep Learning-based Detection of Vessel Occlusions on CT-Angiography in Patients with Suspected Acute Ischemic Stroke

Swift diagnosis and treatment play a decisive role in the clinical outcome of patients with acute ischemic stroke (AIS), and computer-aided diagnosis (CAD) systems can accelerate the underlying diagnostic processes. Here, we developed an artifical neural network (ANN) which allows automated detection of abnormal vessel findings. Pseudoprospective external validation was performed in consecutive patients with suspected AIS from 4 different hospitals during a 6-month timeframe and demonstrated high sensitivity (≥87%) and negative predictive value (≥93%). Benchmarking against two CE- and FDA-approved software solutions showed significantly higher performance for our ANN with improvements of 25–45% for sensitivity and 4–11% for NPV.We provide an imaging platform (https://stroke.neuroAI-HD.org) for online processing of medical imaging data with the developed ANN, including provisions for data crowdsourcing. Notably, this work has previously been published in Nature Communications [1].

Authors and Affiliations

1. Department of Neuroradiology, Heidelberg University Hospital, Heidelberg, Deutschland

   Gianluca Brugnara, Edwin D. Scholze, Hagen Meredig, Aditya Rastogi, Mustafa A. Mahmutoglu, Christian Ulfert, Ulf Neuberger, Zeynep Bendella, Markus Möhlenbruch, Martin Bendszus & Philipp Vollmuth

2. Division of Medical Image Computing, German Cancer Research Center (DKFZ), Heidelberg, Deutschland

   Michael Baumgartner, Klaus Kades, Jonas Scherer, Stefan Denner, Ralf Floca & Klaus Maier-Hein

3. Department of Neuroradiology, Bonn University Hospital, Bonn, Deutschland

   Katerina Deike-Hofmann, Carsten Schmeel & Alexander Radbruch

4. Neurology Clinic, Heidelberg University Hospital, Heidelberg, Deutschland

   Silvia Schönenberger, Wolfgang Wick & Peter A. Ringleb

5. Dep. of Diagnostic and Interventional Radiology with Nuclear Medicine, Thoraxklinik, Heidelberg, Deutschland

   Kai Schlamp

6. Institute for Applied Mathematics, University of Bonn, Bonn, Deutschland

   Thomas Pinetz

Corresponding author

Correspondence to Philipp Vollmuth .

Editors and Affiliations

1. Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Bayern, Deutschland

   Andreas Maier

2. Peter L. Reichertz Institut für Medizinische Informatik, Technische Universität Braunschweig, Braunschweig, Niedersachsen, Deutschland

   Thomas M. Deserno

3. Institut für Medizinische Informatik, Universität zu Lübeck, Lübeck, Schleswig-Holstein, Deutschland

   Heinz Handels

4. Medical Image Computing, E230, Deutsches Krebsforschungszentrum (DKFZ), Heidelberg, Baden-Württemberg, Deutschland

   Klaus Maier-Hein

5. Informatik und Mathematik, OTH Regensburg, Regensburg, Deutschland

   Christoph Palm

6. Institut für Medizinische Informatik, Charité - Universitätsmedizin Berlin, Berlin, Berlin, Deutschland

   Thomas Tolxdorff

Reprints and permissions

© 2024 Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature

Policies and ethics