Tissue-border detection in volumetric laser endomicroscopy using bi-directional gated recurrent neural networks
| Autor: | Okel, Sanne E., van der Sommen, Fons, Selmanaj, Endi, van der Putten, Joost, Struyvenberg, Maarten R., Bergman, Jacques J.G.H.M., De With, Peter H.N., Mazurowski, Maciej A., Drukker, Karen |
|---|---|
| Přispěvatelé: | Center for Care & Cure Technology Eindhoven, Video Coding & Architectures, Eindhoven MedTech Innovation Center, EAISI Health, Gastroenterology and Hepatology, Graduate School, CCA - Imaging and biomarkers, CCA - Cancer Treatment and Quality of Life, Amsterdam Gastroenterology Endocrinology Metabolism |
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
Floating point
Tissue segmentation Computer science business.industry Deep learning ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Recurrent neural network Boundary (topology) CAD Computer aided detection SDG 3 – Goede gezondheid en welzijn Laser law.invention Volumetric laser endomicroscopy Barrett's esophagus SDG 3 - Good Health and Well-being law Endomicroscopy Computer vision Artificial intelligence business |
| Zdroj: | Medical Imaging 2021: Computer-Aided Diagnosis Medical Imaging 2021 Medical Imaging 2021: Computer-Aided Diagnosis, 11597 |
| Popis: | Computer-aided detection (CAD) approaches have shown promising results for early esophageal cancer detection using Volumetric Laser Endoscopy (VLE) imagery. However, the relatively slow and computationally costly tissue segmentation employed in these approaches hamper their clinical applicability. In this paper, we propose to reframe the 2D tissue segmentation problem into a 1D tissue boundary detection problem. Instead of using an encoder-decoder architecture, we propose to follow the tissue boundary using a Recurrent Neural Network (RNN), exploiting the spatio-temporal relations within VLE frames. We demonstrate a near state-of-the-art performance using 18 times less floating point operations, enabling real-time execution in clinical practice. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|