<italic>WSAN</italic>: An Effective Model of Weakly Supervised Similarity Analysis Network for the Lung CT Images

Abstract

With the rapid advancement of medical imaging technologies, the high-resolution CT image data is becoming increasingly valuable for both medical research and clinical diagnosis. The paper takes lung CT image as an example. Retrieving images similar to the input one can help physicians with clinical diagnosis. In comparison to traditional content-based image retrieval, similarity retrieval of lung CT images requires higher retrieval accuracy, with similar requirements in external shape as well as internal vascular and lesion location similarity. In the state-of-the-art supervised deep learning networks, the learning of the network is based on labeling. The labeling of medical images, however, requires time and effort from professionals to label each image, which is prohibitively expensive. In this paper, we propose a weakly supervised deep learning network model for similarity analysis of lung CT images that is called a $W$ eakly $S$ upervised $s$ imilarity $A$ nalysis $N$ etwork ( $WSAN$ ). Extensive experiments show that the $WSAN$ model achieves satisfactory results in measuring the similarity between lung CT images and can be used for similarity retrieval tasks.

Keywords

• Content-based retrieval
• CT~image
• deep learning
• similarity retrieval