Valentin Koch1,2, Olle Holmberg1,2,3, Edna Blum 4, Ece Sancar 1,2, Alp Aytekin4, Masaru Seguchi4, Erion Xhepa4, Jens Wiebe4, Salvatore Cassese4, Sebastian Kufner4, Thorsten Kessler 4,5, Hendrik Sager 4,5, Felix Voll 4, Tobias Rheude 4, Tobias Lenz 4, Adnan Kastrati4,5, Heribert Schunkert4,5, Julia A. Schnabel1,2,6, Michael Joner4,5, Carsten Marr1, Philipp Nicol4
1 Helmholtz Munich - German Research Center for Environmental Health, Germany; 2 School of Computation and Information Technology, Technical University of Munich, Germany; 3 Helsing GmBH, Munich, Germany; 4 German Heart Centre Munich, Technical University of Munich, Germany; 5 German Center for Cardiovascular Research, Partner Site Munich Heart Alliance, Germany; 6 School of Biomedical Engineering and Imaging Sciences, King's College London, UK;We present DeepNeo, a deep learning-based algorithm, to automate the process of segmenting and characterizing neointimal tissue of stented patients (tissue that grows over the stent) in optical coherence tomography (OCT) images. OCT is a tool that provides high-resolution imaging of stented segments after PCI (percutaneous coronary intervention, a non-surgical procedure that improves blood flow to the heart, e.g. by implanting a stent).
We provide an online tool for researchers to quickly analyze and download results,
The user interface of our webtool, which provides a user-friendly platform for the analysis of intravascular OCT images. The interface is designed with several features to facilitate accurate and efficient analysis, including an upload mask (A), which allows users to upload OCT pullback images (DICOM or .zip), a visual representation of the current OCT frame with segmentation and neointima prediction, a schematic view of quadrants (C1) (top row represents quadrant I, bottom row quadrant IV) and neointima and lumen (C2) that provides a visual representation of the tissue characteristics, including a slider (C3) that enables users to move through the pullback. In addition, the interface includes a pullback analysis (D) that provides a detailed analysis of the OCT images and a manual correction feature (E) to correct beginning and end of the stent if necessary. The webtool also allows users to download a detailed analysis of their results and provides an information tab (F) for additional guidance. Users are required to accept the research-only use on the welcome page (G) before accessing the tool.
We are very happy to provide code and models on request. Please contact Valentin Koch (valentin.koch@helmholtz-munich.de), Carsten Marr (carsten.marr@helmholtz-munich.de), or Michael Joner (joner@dhm.mhn.de) for more information.