3 March 2009 Localization of tissues in high-resolution digital anatomic pathology images
Author Affiliations +
Proceedings Volume 7260, Medical Imaging 2009: Computer-Aided Diagnosis; 726016 (2009) https://doi.org/10.1117/12.811430
Event: SPIE Medical Imaging, 2009, Lake Buena Vista (Orlando Area), Florida, United States
High resolution digital pathology images have a wide range of variability in color, shape, size, number, appearance, location, and texture. The segmentation problem is challenging in this environment. We introduce a hybrid method that combines parametric machine learning with heuristic methods for feature extraction as well as pre- and post-processing steps for localizing diverse tissues in slide images. The method uses features such as color, intensity, texture, and spatial distribution. We use principal component analysis for feature reduction and train a two layer back propagation neural network (with one hidden layer). We perform image labeling at pixel-level and achieve higher than 96% automatic localization accuracy on 294 test images.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Raja S. Alomari, Raja S. Alomari, Ron Allen, Ron Allen, Bikash Sabata, Bikash Sabata, Vipin Chaudhary, Vipin Chaudhary, } "Localization of tissues in high-resolution digital anatomic pathology images", Proc. SPIE 7260, Medical Imaging 2009: Computer-Aided Diagnosis, 726016 (3 March 2009); doi: 10.1117/12.811430; https://doi.org/10.1117/12.811430

Back to Top