Learning Deep Representations of Medical Images using Siamese CNNs with Application to Content-Based Image Retrieval

By Yu-An Chung, Wei-Hung Weng. 2017

70 citations recorded by the source corpus.

Deep neural networks have been investigated in learning latent representations of medical images, yet most of the studies limit their approach in a single supervised convolutional neural network (CNN), which usually rely heavily on a large scale annotated dataset for training. To learn image representations with less supervision involved, we propose a deep Siamese CNN (SCNN) architecture that can be trained with only binary image pair information. We evaluated the learned image representations on a task of content-based medical image retrieval using a publicly available multiclass diabetic retinopathy fundus image dataset. The experimental results show that our proposed deep SCNN is comparable to the state-of-the-art single supervised CNN, and requires much less supervision for training.

Subfields: Artificial Intelligence, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zongben Xu
2. Jun Shu
3. Songyang Zhang
4. Deyu Meng
5. Yongfei Liu
6. Xiangyi Zhang
7. Xuming He

In Radiology, Nuclear Medicine and Imaging

1. Jun Shu
2. Songyang Zhang
3. Deyu Meng
4. Yongfei Liu
5. Xiangyi Zhang
6. Xuming He
7. Zongben Xu

Citations in the corpus

• Part-aware Prototype Network for Few-shot Semantic Segmentation (2020, background): “ sizable margin. 1 1 Introduction Semantic segmentation is a core task in modern computer vision with many potential applications ranging from autonomous navigation [4] to medical image understanding [6]. A particular challenge in deploying segmentation algorithms in real-world applications is to ad”
• Small Sample Learning in Big Data Era (2018, methodology): “ As shown in Fig.12(a), Swati et al. (2017) augmented vanilla siamese networks for chromosome classication. In order to reduce the dependency of using actual class labels annotated by human experts, Chung and Weng (2017) proposed a deep siamese CNN to learn xed-length latent image representation fro”