27 February 2018 Saliency U-Net: A regional saliency map-driven hybrid deep learning network for anomaly segmentation
Author Affiliations +
Deep learning networks are gaining popularity in many medical image analysis tasks due to their generalized ability to automatically extract relevant features from raw images. However, this can make the learning problem unnecessarily harder requiring network architectures of high complexity. In case of anomaly detection, in particular, there is often sufficient regional difference between the anomaly and the surrounding parenchyma that could be easily highlighted through bottom-up saliency operators. In this paper we propose a new hybrid deep learning network using a combination of raw image and such regional maps to more accurately learn the anomalies using simpler network architectures. Specifically, we modify a deep learning network called U-Net using both the raw and pre-segmented images as input to produce joint encoding (contraction) and expansion paths (decoding) in the U-Net. We present results of successfully delineating subdural and epidural hematomas in brain CT imaging and liver hemangioma in abdominal CT images using such network.
Conference Presentation
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Alex Karargyros, Alex Karargyros, Tanveer Syeda-Mahmood, Tanveer Syeda-Mahmood, } "Saliency U-Net: A regional saliency map-driven hybrid deep learning network for anomaly segmentation", Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis, 105751T (27 February 2018); doi: 10.1117/12.2293976; https://doi.org/10.1117/12.2293976

Back to Top