<abstract xml:lang="en" xmlns="http://www.w3.org/1999/xhtml"><p>The performance of bone age assessment is highly correlated with the extraction of bony tissue from soft tissues, and the key problem is how to successfully separate epiphyseal/metaphyseal region of interests (EMROIs) from the background and soft tissue. In our experiment, a series of image preprocessing procedures are used to exclude the background and locate the EMROIs of left-hand radiographs. Subsequently, automatic gamma parameter enhancement is applied to test the two segmentation methods (adaptive two-means clustering algorithm and gradient vector flow snake) among children of different age (the age from 2 to 16 years for 80 girls and boys). Four error measurements of misclassification error, relative foreground area error, modified Hausdorff distances, and edge mismatch, are included to evaluate the segmentation performance. The result shows that the two segmentation algorithms are corresponding to different ranges of optimal gamma parameters. Furthermore, the margin of EMROIs can be obtained more precisely by developing an automatic bone age assessment method with the gamma parameter enhancement.</p></abstract>

The performance of bone age assessment is highly correlated with the extraction of bony tissue from soft tissues, and the key problem is how to successfully separate epiphyseal/metaphyseal region of interests (EMROIs) from the background and soft tissue. In our experiment, a series of image preprocessing procedures are used to exclude the background and locate the EMROIs of left-hand radiographs. Subsequently, automatic gamma parameter enhancement is applied to test the two segmentation methods (adaptive two-means clustering algorithm and gradient vector flow snake) among children of different age (the age from 2 to 16 years for 80 girls and boys). Four error measurements of misclassification error, relative foreground area error, modified Hausdorff distances, and edge mismatch, are included to evaluate the segmentation performance. The result shows that the two segmentation algorithms are corresponding to different ranges of optimal gamma parameters. Furthermore, the margin of EMROIs can be obtained more precisely by developing an automatic bone age assessment method with the gamma parameter enhancement.

Automatic Segmentation of Phalanx and Epiphyseal/Metaphyseal Region by Gamma Parameter Enhancement Algorithm

Department of Electrical Engineering, National Chiayi University, No. 300 Syuefu Rd., Chiayi City 60004, Taiwan

Instrument Technology Research Center, National Applied Research Laboratories, 20 R&D Road VI, Hsinchu Science Park, Hsinchu 30076, Taiwan

Department of Electrical Engineering, National Tsing Hwa University, 101, Section 2, Kuang-Fu Road, Hsinchu 30013, Taiwan

Department of Radiology, Taipei Veterans General Hospital, 201, Section 2, Shih-Pai Road, Taipei 112, Taiwan

Measurement Science Review

The performance of bone age assessment is highly correlated with the extraction of bony tissue from soft tissues, and the key problem is how to successfully...

Automatic Segmentation of Phalanx and Epiphyseal/Metaphyseal Region by Gamma Parameter Enhancement Algorithm

Data publikacji: 01 mar 2012

Zakres stron: 21 - 27

DOI: https://doi.org/10.2478/v10048-012-0003-z

Słowa kluczowe
Bone age, Epiphyseal/metaphyseal, Segmentation, Gamma parameter enhancement, Adaptive two-means clustering

This content is open access.

Automatic Segmentation of Phalanx and Epiphyseal/Metaphyseal Region by Gamma Parameter Enhancement Algorithm

Data publikacji: 01 mar 2012

Zakres stron: 21 - 27

DOI: https://doi.org/10.2478/v10048-012-0003-z

Słowa kluczoweBone age, Epiphyseal/metaphyseal, Segmentation, Gamma parameter enhancement, Adaptive two-means clustering

This content is open access.

Słowa kluczowe
Bone age, Epiphyseal/metaphyseal, Segmentation, Gamma parameter enhancement, Adaptive two-means clustering