Ming-Huan Zhang1,2, Jun-Shan Ma1, Ying Shen3, Ying Chen4. 1. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai, 200093, People's Republic of China. 2. Lab of Big Data Analyses and Process, Information Science and Technology, Shanghai Sanda University, 2727 Jinhai Road, Pudong New District, Shanghai, 201209, People's Republic of China. 3. XinHua Hospital Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, 200092, People's Republic of China. 4. Lab of Big Data Analyses and Process, Information Science and Technology, Shanghai Sanda University, 2727 Jinhai Road, Pudong New District, Shanghai, 201209, People's Republic of China. tongyiichen@163.com.
Abstract
BACKGROUND: This study aimed to investigate the optimal support vector machines (SVM)-based classifier of duchenne muscular dystrophy (DMD) magnetic resonance imaging (MRI) images. METHODS: T1-weighted (T1W) and T2-weighted (T2W) images of the 15 boys with DMD and 15 normal controls were obtained. Textural features of the images were extracted and wavelet decomposed, and then, principal features were selected. Scale transform was then performed for MRI images. Afterward, SVM-based classifiers of MRI images were analyzed based on the radical basis function and decomposition levels. The cost (C) parameter and kernel parameter [Formula: see text] were used for classification. Then, the optimal SVM-based classifier, expressed as [Formula: see text]), was identified by performance evaluation (sensitivity, specificity and accuracy). RESULTS: Eight of 12 textural features were selected as principal features (eigenvalues [Formula: see text]). The 16 SVM-based classifiers were obtained using combination of (C, [Formula: see text]), and those with lower C and [Formula: see text] values showed higher performances, especially classifier of [Formula: see text]). The SVM-based classifiers of T1W images showed higher performance than T1W images at the same decomposition level. The T1W images in classifier of [Formula: see text]) at level 2 decomposition showed the highest performance of all, and its overall correct sensitivity, specificity, and accuracy reached 96.9, 97.3, and 97.1 %, respectively. CONCLUSION: The T1W images in SVM-based classifier [Formula: see text] at level 2 decomposition showed the highest performance of all, demonstrating that it was the optimal classification for the diagnosis of DMD.
BACKGROUND: This study aimed to investigate the optimal support vector machines (SVM)-based classifier of duchenne muscular dystrophy (DMD) magnetic resonance imaging (MRI) images. METHODS: T1-weighted (T1W) and T2-weighted (T2W) images of the 15 boys with DMD and 15 normal controls were obtained. Textural features of the images were extracted and wavelet decomposed, and then, principal features were selected. Scale transform was then performed for MRI images. Afterward, SVM-based classifiers of MRI images were analyzed based on the radical basis function and decomposition levels. The cost (C) parameter and kernel parameter [Formula: see text] were used for classification. Then, the optimal SVM-based classifier, expressed as [Formula: see text]), was identified by performance evaluation (sensitivity, specificity and accuracy). RESULTS: Eight of 12 textural features were selected as principal features (eigenvalues [Formula: see text]). The 16 SVM-based classifiers were obtained using combination of (C, [Formula: see text]), and those with lower C and [Formula: see text] values showed higher performances, especially classifier of [Formula: see text]). The SVM-based classifiers of T1W images showed higher performance than T1W images at the same decomposition level. The T1W images in classifier of [Formula: see text]) at level 2 decomposition showed the highest performance of all, and its overall correct sensitivity, specificity, and accuracy reached 96.9, 97.3, and 97.1 %, respectively. CONCLUSION: The T1W images in SVM-based classifier [Formula: see text] at level 2 decomposition showed the highest performance of all, demonstrating that it was the optimal classification for the diagnosis of DMD.
Entities:
Keywords:
Classifier; Duchenne muscular dystrophy; Performance; Support vector machines
Authors: Dawn E Bowles; Scott W J McPhee; Chengwen Li; Steven J Gray; Jade J Samulski; Angelique S Camp; Juan Li; Bing Wang; Paul E Monahan; Joseph E Rabinowitz; Joshua C Grieger; Lakshmanan Govindasamy; Mavis Agbandje-McKenna; Xiao Xiao; R Jude Samulski Journal: Mol Ther Date: 2011-11-08 Impact factor: 11.454
Authors: Kevin M Flanigan; Diane M Dunn; Andrew von Niederhausern; Payam Soltanzadeh; Michael T Howard; Jacinda B Sampson; Kathryn J Swoboda; Mark B Bromberg; Jerry R Mendell; Laura E Taylor; Christine B Anderson; Alan Pestronk; Julaine M Florence; Anne M Connolly; Katherine D Mathews; Brenda Wong; Richard S Finkel; Carsten G Bonnemann; John W Day; Craig McDonald; Robert B Weiss Journal: Hum Mutat Date: 2011-03 Impact factor: 4.878
Authors: Linda J Popplewell; Carl Adkin; Virginia Arechavala-Gomeza; Annemieke Aartsma-Rus; Christa L de Winter; Steve D Wilton; Jennifer E Morgan; Francesco Muntoni; Ian R Graham; George Dickson Journal: Neuromuscul Disord Date: 2010-01-15 Impact factor: 4.296
Authors: Katharine Bushby; Richard Finkel; David J Birnkrant; Laura E Case; Paula R Clemens; Linda Cripe; Ajay Kaul; Kathi Kinnett; Craig McDonald; Shree Pandya; James Poysky; Frederic Shapiro; Jean Tomezsko; Carolyn Constantin Journal: Lancet Neurol Date: 2009-11-27 Impact factor: 44.182
Authors: Luc A Heijnen; Monique Maas; Max J Lahaye; Ulrich Lalji; Doenja M J Lambregts; Milou H Martens; Robert G Riedl; Geerard L Beets; Regina G H Beets-Tan Journal: Eur Radiol Date: 2014-06-05 Impact factor: 5.315
Authors: Baocheng Chu; Annette Kampschulte; Marina S Ferguson; William S Kerwin; Vasily L Yarnykh; Kevin D O'Brien; Nayak L Polissar; Thomas S Hatsukami; Chun Yuan Journal: Stroke Date: 2004-04-01 Impact factor: 7.914