Digital Image Processing

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Feature Extraction from Phalangeal Bones in Radiographs


Affiliations
1 Computer Science and Engineering (PG) Department, Sri Ramakrishna Engineering College, Coimbatore, India
2 Akshaya College of Engineering and Technology, Coimbatore, India
3 Electrical Engineering Department, Coimbatore Institute of Engineering and Technology, Coimbatore, India
     

   Subscribe/Renew Journal


This paper proposes an automated system for bone age estimation from left hand-wrist radiographs. The input image is first preprocessed and segmented using Particle Swarm Optimization (PSO) consisting of graph-based and tetrolet-based segmentation procedures. Then the phalangeal bones specially used as Regions of Interest (ROIs) for the bone age estimation process are identified. From the phalangeal ROIs identified, 24 morphological features are extracted. Based on the results of feature extraction, the performances of the two segmentation approaches are compared. The system is validated with a data set of 100 images with 50 radiographs of female subjects and 50 of male subjects. The accuracy of feature extraction from the ROI segmented using both the segmentation procedures is calculated and the results are discussed.

Keywords

Bone Age, Bone Age Assessment (BAA), Particle Swarm Optimization (PSO), Graph-Based Segmentation, Tetrolets, Tetrolet-Based Segmentation, Feature Extraction.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 222

PDF Views: 2




  • Feature Extraction from Phalangeal Bones in Radiographs

Abstract Views: 222  |  PDF Views: 2

Authors

P. Thangam
Computer Science and Engineering (PG) Department, Sri Ramakrishna Engineering College, Coimbatore, India
K. Thanushkodi
Akshaya College of Engineering and Technology, Coimbatore, India
T. V. Mahendiran
Electrical Engineering Department, Coimbatore Institute of Engineering and Technology, Coimbatore, India

Abstract


This paper proposes an automated system for bone age estimation from left hand-wrist radiographs. The input image is first preprocessed and segmented using Particle Swarm Optimization (PSO) consisting of graph-based and tetrolet-based segmentation procedures. Then the phalangeal bones specially used as Regions of Interest (ROIs) for the bone age estimation process are identified. From the phalangeal ROIs identified, 24 morphological features are extracted. Based on the results of feature extraction, the performances of the two segmentation approaches are compared. The system is validated with a data set of 100 images with 50 radiographs of female subjects and 50 of male subjects. The accuracy of feature extraction from the ROI segmented using both the segmentation procedures is calculated and the results are discussed.

Keywords


Bone Age, Bone Age Assessment (BAA), Particle Swarm Optimization (PSO), Graph-Based Segmentation, Tetrolets, Tetrolet-Based Segmentation, Feature Extraction.