An Ensemble-Based Densely-Connected Deep Learning System for Assessment of Skeletal Maturity

Assessment of skeletal maturity plays an essential role in the clinical management of the adolescent disease. This task is very challenging when using machine learning method due to the limited data and large anatomical variations among different subjects. In this paper, we propose a deep learning pipeline to automatically assess the distal radius and ulna (DRU) maturity from left hand radiographs. The model we proposed acquires two convincing advantages: first, our model preserves the maximum information flow and has a much faster convergence rate. Second, our model avoids overfitting even if training with limited data. The proposed method achieves 83.33% and 90.31% for radius and ulna classification respectively.

[1]  Bin Liang,et al.  Using Convolutional Neural Networks and Transfer Learning for Bone Age Classification , 2017, 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA).

[2]  Simon Haykin,et al.  GradientBased Learning Applied to Document Recognition , 2001 .

[3]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  W. Greulich,et al.  Radiographic Atlas of Skeletal Development of the Hand and Wrist , 1999 .

[5]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[7]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  M. Yao,et al.  Prognostic value of a computer-aided diagnosis system involving bone scans among men treated with docetaxel for metastatic castration-resistant prostate cancer , 2016, BMC Cancer.

[9]  James M. Tanner,et al.  Assessment of skeletal maturity and prediction of adult height:(TW3 Method) , 2001 .

[10]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[11]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[12]  Jürgen Schmidhuber,et al.  Training Very Deep Networks , 2015, NIPS.

[13]  Matthew Chen.  Automated Bone Age Classification with Deep Neural Networks , 2016 .

[14]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[15]  P M Kemp,et al.  Bone age assessment: a large scale comparison of the Greulich and Pyle, and Tanner and Whitehouse (TW2) methods , 1999, Archives of disease in childhood.

[16]  M. Yao,et al.  Prognostic value of the bone scan index using a computer-aided diagnosis system for bone scans in hormone-naive prostate cancer patients with bone metastases , 2016, BMC Cancer.

[17]  J. Tanner,et al.  Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. , 1976, Archives of disease in childhood.

[18]  Kwang Gi Kim,et al.  Applying Deep Learning in Medical Images: The Case of Bone Age Estimation , 2018, Healthcare informatics research.

[19]  Dino Samartzis,et al.  Assessment of skeletal maturity in scoliosis patients to determine clinical management: a new classification scheme using distal radius and ulna radiographs. , 2014, The spine journal : official journal of the North American Spine Society.