A fuzzy Hopfield neural network for medical image segmentation

Journal Article · Thu Aug 01 04:00:00 EDT 1996 · IEEE Transactions on Nuclear Science

OSTI ID:376043

Lin, J S; Cheng, K S; Mao, C W ^[1]

---

1. National Cheng Kung Univ., Tainan (Taiwan, Province of China)

In this paper, an unsupervised parallel segmentation approach using a fuzzy Hopfield neural network (FHNN) is proposed. The main purpose is to embed fuzzy clustering into neural networks so that on-line learning and parallel implementation for medical image segmentation are feasible. The idea is to cast a clustering problem as a minimization problem where the criteria for the optimum segmentation is chosen as the minimization of the Euclidean distance between samples to class centers. In order to generate feasible results, a fuzzy c-means clustering strategy is included in the Hopfield neural network to eliminate the need of finding weighting factors in the energy function, which is formulated and based on a basic concept commonly used in pattern classification, called the within-class scatter matrix principle. The suggested fuzzy c-means clustering strategy has also been proven to be convergent and to allow the network to learn more effectively than the conventional Hopfield neural network. The fuzzy Hopfield neural network based on the within-class scatter matrix shows the promising results in comparison with the hard c-means method.

• OSTI ID: 376043
• Journal Information: IEEE Transactions on Nuclear Science, Journal Name: IEEE Transactions on Nuclear Science Journal Issue: 4Pt2 Vol. 43; ISSN 0018-9499; ISSN IETNAE
• Country of Publication: United States
• Language: English

Related Subjects

55 BIOLOGY AND MEDICINE
BASIC STUDIES
99 GENERAL AND MISCELLANEOUS
COMPUTERIZED TOMOGRAPHY
FUZZY LOGIC
IMAGE PROCESSING
NEURAL NETWORKS
PARALLEL PROCESSING
POSITRON COMPUTED TOMOGRAPHY