Segmentation of Bone Metastasis in CT Images Based on Modified HED

Segmentation of the bone metastasis area in medical images can reduce the workload for diagnosis and treatment. However, there are various shapes and sizes of bone metastasis also affected by noise. As a result, it is difficult to segment using classical segmentation methods. In this paper, we propose a convolutional neural network model-based segmentation method. The proposed method easily predicts the contour and location of the lesion area using side connection and modified network. In this study, we modified again the modified HED network to match the characteristics of bone metastasis. The experimental results using the proposed method for segmenting bone metastasis in the lesion area has 79.8[%] of TP rate and 69.2[%] of IOU rate.

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

[2]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Ronald M. Summers,et al.  Detection of Sclerotic Spine Metastases via Random Aggregation of Deep Convolutional Neural Network Classifications , 2014, ArXiv.

[4]  Huimin Lu,et al.  Image registration of vertebral region from CT images based on salient region feature , 2017, 2017 17th International Conference on Control, Automation and Systems (ICCAS).

[5]  R. Coleman.  Clinical Features of Metastatic Bone Disease and Risk of Skeletal Morbidity , 2006, Clinical Cancer Research.

[6]  M. Iqbal Saripan,et al.  Segmenting CT images of bronchogenic carcinoma with bone metastases using PET intensity markers approach , 2009 .

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

[8]  Ronald M. Summers,et al.  Improving Computer-Aided Detection Using Convolutional Neural Networks and Random View Aggregation , 2015, IEEE Transactions on Medical Imaging.

[9]  Mario Ceresa,et al.  Fully automatic detection and segmentation of abdominal aortic thrombus in post‐operative CTA images using Deep Convolutional Neural Networks , 2018, Medical Image Anal..

[10]  Jiri Chmelik,et al.  Fully Automatic CAD System for Segmentation and Classification of Spinal Metastatic Lesions in CT Data , 2019 .

[11]  Roman Jakubicek,et al.  Deep convolutional neural network‐based segmentation and classification of difficult to define metastatic spinal lesions in 3D CT data , 2018, Medical Image Anal..