Deep Convolutional Neural Network-Based Automatic Classification of Neonatal Hip Ultrasound Images: A Novel Data Augmentation Approach with Speckle Noise Reduction.

Neonatal hip ultrasound imaging has been widely used for a few decades in the diagnosis of developmental dysplasia of the hip (DDH). Graf's method of hip ultrasonography is still the most reproducible because of its classification system; yet, the reproducibility is questionable as a result of the high dependency on the skills of the sonographer and the evaluator. A computer-aided diagnosis system may help evaluators increase their precision in the diagnosis of DDH using Graf's method. This study describes a fully automatic computer-aided diagnosis system for the classification of hip ultrasound images captured in Graf's standard plane based on convolutional neural networks (CNNs). Automatically segmented image patches containing all of the necessary anatomical structures were given to the proposed CNN system to extract discriminative features and classify the recorded hips. For ease of evaluation, the data set was divided into three groups: normal, mild dysplasia and severe dysplasia. This study proposes a different approach to data augmentation using speckle noise reduction with an optimized Bayesian non-local mean filter. Data augmentation with this filter increased the accuracy of the proposed CNN system from 92.29% to 97.70%. This new approach for automatic classification of DDH, classifies dysplastic neonatal hips with a high accuracy rate and might help evaluators to increase their evaluation success.

[1]  Qi Zhang,et al.  Deep learning based classification of breast tumors with shear-wave elastography. , 2016, Ultrasonics.

[2]  Niamul Quader,et al.  Automatic Evaluation of Scan Adequacy and Dysplasia Metrics in 2-D Ultrasound Images of the Neonatal Hip. , 2016, Ultrasound in medicine & biology.

[3]  Y. Sağlam,et al.  Reliability of Computer‐Assisted and Manual Measurement Methods for Assessment of Graf Type 1 and Type 2 Hip Sonograms , 2016, Journal of ultrasound in medicine : official journal of the American Institute of Ultrasound in Medicine.

[4]  Nima Tajbakhsh,et al.  Computer-Aided Pulmonary Embolism Detection Using a Novel Vessel-Aligned Multi-planar Image Representation and Convolutional Neural Networks , 2015, MICCAI.

[5]  Aysun Sezer,et al.  Automatic segmentation and classification of neonatal hips according to Graf's sonographic method: A computer-aided diagnosis system , 2019, Appl. Soft Comput..

[6]  B Huynh,et al.  MO-DE-207B-06: Computer-Aided Diagnosis of Breast Ultrasound Images Using Transfer Learning From Deep Convolutional Neural Networks. , 2016, Medical physics.

[7]  Mingyue Ding,et al.  Deep learning based classification of focal liver lesions with contrast-enhanced ultrasound , 2014 .

[8]  Hongwei Liu,et al.  Convolutional Neural Network With Data Augmentation for SAR Target Recognition , 2016, IEEE Geoscience and Remote Sensing Letters.

[9]  R Graf,et al.  Fundamentals of Sonographic Diagnosis of Infant Hip Dysplasia , 1984, Journal of pediatric orthopedics.

[10]  Chen Wang,et al.  Comparison of Despeckle Filters for Breast Ultrasound Images , 2015, Circuits Syst. Signal Process..

[11]  R. Graf,et al.  Hip Sonography: Diagnosis and Management of Infant Hip Dysplasia , 2006 .

[12]  Kumaradevan Punithakumar,et al.  Toward automated classification of acetabular shape in ultrasound for diagnosis of DDH: Contour alpha angle and the rounding index , 2016, Comput. Methods Programs Biomed..

[13]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[14]  Abhilash Rakkunedeth Hareendranathan,et al.  An index for diagnosing infant hip dysplasia using 3-D ultrasound: the acetabular contact angle , 2016, Pediatric Radiology.

[15]  C. Alberola-Lopez,et al.  P6D-4 Analysis of Ultrasound Images Based on Local Statistics. Application to the Diagnosis of Developmental Dysplasia of the Hip , 2007, 2007 IEEE Ultrasonics Symposium Proceedings.

[16]  Ping Chen,et al.  Segmentation of Fetal Left Ventricle in Echocardiographic Sequences Based on Dynamic Convolutional Neural Networks , 2017, IEEE Transactions on Biomedical Engineering.

[17]  Hao Chen,et al.  Standard Plane Localization in Fetal Ultrasound via Domain Transferred Deep Neural Networks , 2015, IEEE Journal of Biomedical and Health Informatics.

[18]  Pierrick Coupé,et al.  Nonlocal Means-Based Speckle Filtering for Ultrasound Images , 2009, IEEE Transactions on Image Processing.

[19]  Kumaradevan Punithakumar,et al.  A technique for semiautomatic segmentation of echogenic structures in 3D ultrasound, applied to infant hip dysplasia , 2015, International Journal of Computer Assisted Radiology and Surgery.

[20]  Scott T. Acton,et al.  Speckle reducing anisotropic diffusion , 2002, IEEE Trans. Image Process..

[21]  Jinlian Ma,et al.  A pre‐trained convolutional neural network based method for thyroid nodule diagnosis , 2017, Ultrasonics.

[22]  Yoni Donner,et al.  Fully Automating Graf's Method for DDH Diagnosis Using Deep Convolutional Neural Networks , 2016, LABELS/DLMIA@MICCAI.

[23]  S. Dudea,et al.  A new sign in the standard hip ultrasound image of the Graf method. , 2015, Medical ultrasonography.

[24]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[25]  Miguel Ángel Guevara-López,et al.  Representation learning for mammography mass lesion classification with convolutional neural networks , 2016, Comput. Methods Programs Biomed..

[26]  Andrew Zisserman,et al.  Representing shape with a spatial pyramid kernel , 2007, CIVR '07.

[27]  Tony F. Chan,et al.  Active contours without edges , 2001, IEEE Trans. Image Process..

[28]  G. Zielhuis,et al.  Ultrasonographic screening for developmental dysplasia of the hip in infants. Reproducibility of assessments made by radiographers. , 2003, The Journal of bone and joint surgery. British volume.

[29]  Hakan Omeroğlu,et al.  Use of ultrasonography in developmental dysplasia of the hip , 2014, Journal of children's orthopaedics.

[30]  Eibe Frank,et al.  Evaluating the Replicability of Significance Tests for Comparing Learning Algorithms , 2004, PAKDD.

[31]  Nico Karssemeijer,et al.  Large scale deep learning for computer aided detection of mammographic lesions , 2017, Medical Image Anal..

[32]  R. Glaab,et al.  Inter-observer agreement of ultrasonographic measurement of alpha and beta angles and the final type classification based on the Graf method. , 2004, Swiss medical weekly.

[33]  Dorin Comaniciu,et al.  Marginal Space Deep Learning: Efficient Architecture for Volumetric Image Parsing , 2016, IEEE Transactions on Medical Imaging.

[34]  Jacob L Jaremko,et al.  Potential for change in US diagnosis of hip dysplasia solely caused by changes in probe orientation: patterns of alpha-angle variation revealed by using three-dimensional US. , 2014, Radiology.

[35]  Hermann Ney,et al.  Bag-of-visual-words models for adult image classification and filtering , 2008, 2008 19th International Conference on Pattern Recognition.

[36]  Marios Anthimopoulos,et al.  Lung Pattern Classification for Interstitial Lung Diseases Using a Deep Convolutional Neural Network , 2016, IEEE Transactions on Medical Imaging.

[37]  Nicholas Ayache,et al.  Fine-tuned convolutional neural nets for cardiac MRI acquisition plane recognition , 2017, Comput. methods Biomech. Biomed. Eng. Imaging Vis..

[38]  Y. Ueki,et al.  Reproducibility of acquiring ultrasonographic infant hip images by the Graf method after an infant hip ultrasound training course , 2018, Journal of medical ultrasonics.