From spoken narratives to domain knowledge: Mining linguistic data for medical image understanding

Image understanding is an important topic studied in imaging, computing, and the cognitive sciences and incorporates the domain knowledge of target images, human vision and psychophysics, and data mining. In order to better perform computational image understanding tasks, such as object detection [1–3], shape estimation [4], or depth estimation [5], common knowledge from humans is borrowed and injected into a variety of algorithms. However, hard-coded human knowledge cannot be easily and directly applied for complex tasks such as image classification and retrieval, which involve a wide range of images and require human expertise.

Source: Artificial Intelligence in Medicine - August 19, 2014 Category: Bioinformatics Authors: Xuan Guo, Qi Yu, Cecilia Ovesdotter Alm, Cara Calvelli, Jeff B. Pelz, Pengcheng Shi, Anne R. Haake Source Type: research