Detecting cerebral microbleeds with transfer learning

Cerebral microbleeds (CMBs) are small perivascular hemosiderin deposits leaked from cerebral small vessels in normal (or near normal) tissue. It is important to detect CMBs accurately and reliably for diagnosing and researching some cerebrovascular diseases and cognitive dysfunctions. In the last decade, several approaches based on traditional machine learning and classical convolutional neural network (CNN) were developed for detecting CMBs semi-automatically and automatically. In recent years, numerous advanced variants of CNN with deeper structure have been developed for image recognition, showing better performances comparing with classical CNN. In particular, ResNet proposed recently won the championships on many important image recognition benchmarks because of its extremely deep representations. In view of this, we proposed a method based on ResNet-50 for exploring the possibility of further improving the accuracy of CMBs detection in this study. Due to our small CMB samples size, transfer learning was employed. Based on the transfer learning of ResNet-50, we achieved a high performance with a sensitivity of 95.71 ± 1.044%, a specificity of 99.21 ± 0.076%, and an accuracy of 97.46 ± 0.524% in format of average ± standard deviation, which outperformed three state-of-the-art methods.

[1]  Jean Ponce,et al.  A Theoretical Analysis of Feature Pooling in Visual Recognition , 2010, ICML.

[2]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[3]  Tao Chen,et al.  An Up-to-Date Review on Cerebral Microbleeds. , 2016, Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association.

[4]  B. M. ter Haar Romeny,et al.  Automated detection of cerebral microbleeds in patients with Traumatic Brain Injury , 2016, NeuroImage: Clinical.

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

[6]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[7]  Ling Wei,et al.  Combination of stationary wavelet transform and kernel support vector machines for pathological brain detection , 2016, Simul..

[8]  Y. Itoyama,et al.  Silent Cerebral Microbleeds on T2*-Weighted MRI: Correlation with Stroke Subtype, Stroke Recurrence, and Leukoaraiosis , 2002, Stroke.

[9]  Olivier Salvado,et al.  Computer-aided detection of cerebral microbleeds in susceptibility-weighted imaging , 2015, Comput. Medical Imaging Graph..

[10]  Huimin Lu,et al.  Wavelet energy entropy and linear regression classifier for detecting abnormal breasts , 2018, Multimedia Tools and Applications.

[11]  Yi Chen,et al.  A Feature-Free 30-Disease Pathological Brain Detection System by Linear Regression Classifier. , 2017, CNS & neurological disorders drug targets.

[12]  Hong Chen,et al.  Voxelwise detection of cerebral microbleed in CADASIL patients by leaky rectified linear unit and early stopping , 2017, Multimedia Tools and Applications.

[13]  Hong Chen,et al.  Seven-layer deep neural network based on sparse autoencoder for voxelwise detection of cerebral microbleed , 2017, Multimedia Tools and Applications.

[14]  Weiguo Fan,et al.  A new image classification method using CNN transfer learning and web data augmentation , 2018, Expert Syst. Appl..

[15]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[16]  Jian Sun,et al.  Convolutional neural networks at constrained time cost , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Sidan Du,et al.  Application of stationary wavelet entropy in pathological brain detection , 2018, Multimedia Tools and Applications.

[18]  D. Werring,et al.  Cognitive dysfunction in patients with cerebral microbleeds on T2*-weighted gradient-echo MRI. , 2004, Brain : a journal of neurology.

[19]  Yi Chen,et al.  Fruit category classification via an eight-layer convolutional neural network with parametric rectified linear unit and dropout technique , 2018, Multimedia Tools and Applications.

[20]  V. Mok,et al.  Cerebral microbleeds and white matter changes in patients hospitalized with lacunar infarcts , 2004, Journal of Neurology.

[21]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Snehashis Roy,et al.  Cerebral microbleed segmentation from susceptibility weighted images , 2015, Medical Imaging.

[23]  Shuihua Wang,et al.  RGB-D image-based detection of stairs, pedestrian crosswalks and traffic signs , 2014, J. Vis. Commun. Image Represent..

[24]  Huimin Lu,et al.  Multiple Sclerosis Detection Based on Biorthogonal Wavelet Transform, RBF Kernel Principal Component Analysis, and Logistic Regression , 2016, IEEE Access.

[25]  E Mark Haacke,et al.  Semiautomated detection of cerebral microbleeds in magnetic resonance images. , 2011, Magnetic resonance imaging.

[26]  J. Wardlaw,et al.  Spontaneous brain microbleeds: systematic review, subgroup analyses and standards for study design and reporting. , 2007, Brain : a journal of neurology.

[27]  Jürgen Schmidhuber,et al.  Highway Networks , 2015, ArXiv.

[28]  Hao Chen,et al.  Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks , 2016, IEEE Transactions on Medical Imaging.

[29]  Ming Yang,et al.  Sensorineural hearing loss detection via discrete wavelet transform and principal component analysis combined with generalized eigenvalue proximal support vector machine and Tikhonov regularization , 2018, Multimedia Tools and Applications.

[30]  Jie Liu,et al.  Classification of cerebral microbleeds based on fully-optimized convolutional neural network , 2018, Multimedia Tools and Applications.

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

[32]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

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

[34]  Niels D Prins,et al.  Cerebral small-vessel disease and decline in information processing speed, executive function and memory. , 2005, Brain : a journal of neurology.

[35]  Max A. Viergever,et al.  Efficient detection of cerebral microbleeds on 7.0T MR images using the radial symmetry transform , 2012, NeuroImage.

[36]  Ming Yang,et al.  Classification of Alzheimer’s Disease Based on Eight-Layer Convolutional Neural Network with Leaky Rectified Linear Unit and Max Pooling , 2018, Journal of Medical Systems.

[37]  Surya Ganguli,et al.  Exact solutions to the nonlinear dynamics of learning in deep linear neural networks , 2013, ICLR.

[38]  E. Haacke,et al.  Susceptibility-Weighted Imaging: Technical Aspects and Clinical Applications, Part 1 , 2008, American Journal of Neuroradiology.

[39]  Shuihua Wang,et al.  Tea category identification based on optimal wavelet entropy and weighted k-Nearest Neighbors algorithm , 2018, Multimedia Tools and Applications.

[40]  Susan M. Chang,et al.  Computer-aided detection of radiation-induced cerebral microbleeds on susceptibility-weighted MR images☆ , 2013, NeuroImage: Clinical.

[41]  Steven Warach,et al.  Cerebral Microbleeds : A Field Guide to their Detection and Interpretation , 2012 .

[42]  Ming Yang,et al.  A pathological brain detection system based on kernel based ELM , 2016, Multimedia Tools and Applications.

[43]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[44]  D. Werring,et al.  The Microbleed Anatomical Rating Scale (MARS) , 2009, Neurology.

[45]  Sidan Du,et al.  Cerebral Micro-Bleed Detection Based on the Convolution Neural Network With Rank Based Average Pooling , 2017, IEEE Access.

[46]  J. van Leeuwen,et al.  Neural Networks: Tricks of the Trade , 2002, Lecture Notes in Computer Science.

[47]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[48]  P Kapeller,et al.  MRI evidence of past cerebral microbleeds in a healthy elderly population , 1999, Neurology.

[49]  M. Seghier,et al.  Microbleed Detection Using Automated Segmentation (MIDAS): A New Method Applicable to Standard Clinical MR Images , 2011, PloS one.