【论文阅读笔记】Deep Learning in Medical Imaging: Overview and Future Promise of an Exciting New Technique
2018-02-03 11:14
1361 查看
本文是一篇类似于当前期刊论文介绍的文章,对深度学习应用于医学图像分析方向的研究论文进行了简单的小结,论文发表在IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 35, NO. 5, MAY 2016。文章一开始从大的方面介绍了Deep Learning应用在医学图像分析方面的难点:1、缺少大量带标注的数据 2、使用CNN训练医学图像,计算资源要求很高 3、由于正负样本的不平衡,为了避免神经网络过拟合,需要大量的参数调优工作。
文章后面介绍了相关的一些论文,在此我分类的整理出来一些重要的文献放在最后,大多数发表在IEEE TRANSACTIONS ON MEDICAL IMAGING上,翻墙才能下载,感兴趣的童鞋可以留言给我。
从文中可以看出,在医学影像学深度学习处理方面,目前人们对网络结构的探索主要有以下几个方面:
1、数据维度的探索:将2D图片转换为2.5D或者3D来进行数据增强
2、学习方法的探索:监督学习和非监督学习的应用。监督学习被用在检测,分割,和标记等分类任务,非监督学习常用于图像编码、高效的图像便是和监督学习之前的数据预处理。在参考文献【8】中结合使用了卷积网络和受限波尔茨曼机,超过了单一使用监督或非监督学习更好的效果,有兴趣的可以一读。
3、训练数据的探索:医学数据通常比较难获取,尤其是带标记的数据更是难以获得。在获得的数据中通常正常样本是大多数,病理样本较少,导致训练中的大多数时间用在了正常样本上,这样既浪费时间,还可能导致过拟合,为了更高效的应用现有的少量数据,在文献【12】中提出了一中加速的CNN网络结构,有兴趣的可以一读。
4、迁移学习和优化调参:由于医学数据的难以获得,可以采取在开源大数据集上已经训练得到的最优CNN网络参数,附加自己的网络结构进行参数调优,用于医学图像训练,以弥补数据量的不足。
文章中提到的开源数据集:
VISCERAL:http://www.visceral.eu/
The Cancer Imaging Archive:http://www.cancerimagingarchive.ne
BioImage benchmark:https://grand-challenge.org/
diabetic retinopathy:https://www.kaggle.com/c/diabetic-retinopathy-detection
cardiac volumes:https://www.kaggle.com/c/second-annual-data-science-bowl
参考文献:
1、Lesion Detection
【1】“Medical image processing on the GPU–Past, present and future,” Med. Image Anal., vol. 17, no. 8, pp. 1073–94, 2013.
【2】“Pulmonary nodule detection in CT images using multiview convolutional networks,” IEEE Trans. Med. Imag., vol. 35, no. 5,pp. 1160–1169, May 2016.
【3】“Improving computer-aided detection using convolutional neural networks and random view aggregation,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1170–1181, May 2016.
【4】“Automatic detection of cerebral microbleeds from MR images via 3D convolutional neural networks,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1182–1195, May 2016.”
【5】“Locality sensitive deep learning for detection and classification of nuclei in routine colon cancer histology
images,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1196–1206, May 2016.
【6】“Lung pattern classification for interstitial lung diseases using a deep convolutionalneuralnetwork,”IEEE Trans.
Med. Imag., vol. 35, no. 5, pp. 1207–1216, May 2016.
【7】“Deep convolutional neural networks for computeraided detection: CNN architectures, dataset characteristics and transfer learning,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1285–1298, May 2016.
【8】“Combining generative and discriminative representation learning in convolutional restricted Boltzmann
machines,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1262–1272, May 2016.
2、Segmentation and Shape Modeling
【9】“Deep 3D convolutional encoder networks with shortcuts for multiscale feature integration applied to multiple sclerosis lesion segmentation,” IEEE Trans. Med. Imag., vol. 35, no. 5,pp. 1229–1239, May 2016.
【10】“Brain tumor segmentation using convolutional neural networks in MRI images,” IEEE Trans.
Med. Imag., vol. 35, no. 5, pp. 1240–1251, May 2016.
【11】“Automatic segmentation of MR brain images with a convolutional neural network,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1252–1261, May 2016.
3、Network Exploration
【12】M. van Grinsven, B. van Ginneken, C. Hoyng, T. Theelen, and C.Sánchez, “Fast convolutional neural network training using selective data sampling: Application to hemorrhage detection in color fundus images,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp.
1273–1284, May 2016.
4、Transfer Learning and Fine Tuning
【13】Y. Bar, I. Diamant, L. Wolf, and H. Greenspan, “Deep learning with non-medical training used for chest pathology identification,” Proc.
【14】Y. Bar, I. Diamant, L. Wolf, S. Lieberman, E. Konen, and H.、Greenspan, “Chest pathology detection using deep learning with non-medical training,” in Proc. IEEE 12th Int. Symp. Biomed. Imag.,2015, pp. 294–297.
【15】B. van Ginneken, A. A. Setio, C. Jacobs, and F. Ciompi, “Off-the-shelf convolutional neural network features for pulmonary nodule detection in computed tomography scans,” in Proc. IEEE 12th Int. Symp.Biomed. Imag., 2015, pp. 286–289.
【16】N. Tajbakhsh et al., “Convolutional neural networks for medical image analysis: Full training or fine tuning?,” IEEE Trans. Med. Imag., vol.35, no. 5, pp. 1299–1312, May 2016.
【17】”Multi-instance deep learning: Discover discriminative local anatomies for bodypart recognition,” IEEE Trans. Med. Imag.,vol. 35, no. 5, pp. 1332–1343, May 2016.
文章后面介绍了相关的一些论文,在此我分类的整理出来一些重要的文献放在最后,大多数发表在IEEE TRANSACTIONS ON MEDICAL IMAGING上,翻墙才能下载,感兴趣的童鞋可以留言给我。
从文中可以看出,在医学影像学深度学习处理方面,目前人们对网络结构的探索主要有以下几个方面:
1、数据维度的探索:将2D图片转换为2.5D或者3D来进行数据增强
2、学习方法的探索:监督学习和非监督学习的应用。监督学习被用在检测,分割,和标记等分类任务,非监督学习常用于图像编码、高效的图像便是和监督学习之前的数据预处理。在参考文献【8】中结合使用了卷积网络和受限波尔茨曼机,超过了单一使用监督或非监督学习更好的效果,有兴趣的可以一读。
3、训练数据的探索:医学数据通常比较难获取,尤其是带标记的数据更是难以获得。在获得的数据中通常正常样本是大多数,病理样本较少,导致训练中的大多数时间用在了正常样本上,这样既浪费时间,还可能导致过拟合,为了更高效的应用现有的少量数据,在文献【12】中提出了一中加速的CNN网络结构,有兴趣的可以一读。
4、迁移学习和优化调参:由于医学数据的难以获得,可以采取在开源大数据集上已经训练得到的最优CNN网络参数,附加自己的网络结构进行参数调优,用于医学图像训练,以弥补数据量的不足。
文章中提到的开源数据集:
VISCERAL:http://www.visceral.eu/
The Cancer Imaging Archive:http://www.cancerimagingarchive.ne
BioImage benchmark:https://grand-challenge.org/
diabetic retinopathy:https://www.kaggle.com/c/diabetic-retinopathy-detection
cardiac volumes:https://www.kaggle.com/c/second-annual-data-science-bowl
参考文献:
1、Lesion Detection
【1】“Medical image processing on the GPU–Past, present and future,” Med. Image Anal., vol. 17, no. 8, pp. 1073–94, 2013.
【2】“Pulmonary nodule detection in CT images using multiview convolutional networks,” IEEE Trans. Med. Imag., vol. 35, no. 5,pp. 1160–1169, May 2016.
【3】“Improving computer-aided detection using convolutional neural networks and random view aggregation,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1170–1181, May 2016.
【4】“Automatic detection of cerebral microbleeds from MR images via 3D convolutional neural networks,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1182–1195, May 2016.”
【5】“Locality sensitive deep learning for detection and classification of nuclei in routine colon cancer histology
images,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1196–1206, May 2016.
【6】“Lung pattern classification for interstitial lung diseases using a deep convolutionalneuralnetwork,”IEEE Trans.
Med. Imag., vol. 35, no. 5, pp. 1207–1216, May 2016.
【7】“Deep convolutional neural networks for computeraided detection: CNN architectures, dataset characteristics and transfer learning,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1285–1298, May 2016.
【8】“Combining generative and discriminative representation learning in convolutional restricted Boltzmann
machines,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1262–1272, May 2016.
2、Segmentation and Shape Modeling
【9】“Deep 3D convolutional encoder networks with shortcuts for multiscale feature integration applied to multiple sclerosis lesion segmentation,” IEEE Trans. Med. Imag., vol. 35, no. 5,pp. 1229–1239, May 2016.
【10】“Brain tumor segmentation using convolutional neural networks in MRI images,” IEEE Trans.
Med. Imag., vol. 35, no. 5, pp. 1240–1251, May 2016.
【11】“Automatic segmentation of MR brain images with a convolutional neural network,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp. 1252–1261, May 2016.
3、Network Exploration
【12】M. van Grinsven, B. van Ginneken, C. Hoyng, T. Theelen, and C.Sánchez, “Fast convolutional neural network training using selective data sampling: Application to hemorrhage detection in color fundus images,” IEEE Trans. Med. Imag., vol. 35, no. 5, pp.
1273–1284, May 2016.
4、Transfer Learning and Fine Tuning
【13】Y. Bar, I. Diamant, L. Wolf, and H. Greenspan, “Deep learning with non-medical training used for chest pathology identification,” Proc.
【14】Y. Bar, I. Diamant, L. Wolf, S. Lieberman, E. Konen, and H.、Greenspan, “Chest pathology detection using deep learning with non-medical training,” in Proc. IEEE 12th Int. Symp. Biomed. Imag.,2015, pp. 294–297.
【15】B. van Ginneken, A. A. Setio, C. Jacobs, and F. Ciompi, “Off-the-shelf convolutional neural network features for pulmonary nodule detection in computed tomography scans,” in Proc. IEEE 12th Int. Symp.Biomed. Imag., 2015, pp. 286–289.
【16】N. Tajbakhsh et al., “Convolutional neural networks for medical image analysis: Full training or fine tuning?,” IEEE Trans. Med. Imag., vol.35, no. 5, pp. 1299–1312, May 2016.
【17】”Multi-instance deep learning: Discover discriminative local anatomies for bodypart recognition,” IEEE Trans. Med. Imag.,vol. 35, no. 5, pp. 1332–1343, May 2016.
相关文章推荐
- 【论文阅读笔记】Deep Learning based Recommender System: A Survey and New Perspectives
- 论文阅读笔记:Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey
- 论文笔记:LSTM, GRU, Highway and a Bit of Attention: An Empirical Overview for Language Modeling in Speec
- 论文笔记 A Large Contextual Dataset for Classification,Detection and Counting of Cars with Deep Learning
- 论文阅读笔记之《Learning Deep Correspondence through Prior and Posterior Feature Constancy》
- An Overview Of The New Services, Controls, And Features In ASP.NET 2.0
- 论文笔记 Ensemble of Deep Convolutional Neural Networks for Learning to Detect Retinal Vessels in Fundus
- Attention and Memory in Deep Learning and NLP(深度学习和NLP中的注意和记忆机制) 阅读笔记
- [论文阅读笔记] DeepNeuron: An Open Deep Learning Toolbox for Neuron Tracing
- [深度学习论文笔记][Weight Initialization] Exact solutions to the nonlinear dynamics of learning in deep lin
- APPLYING DEEP LEARNING TO ANSWER SELECTION: A STUDY AND AN OPEN TASK 论文阅读
- 1705.Person Re-Identification by Deep Joint Learning of Multi-Loss Classification 论文阅读笔记
- 深度神经网络的多任务学习概览(An Overview of Multi-task Learning in Deep Neural Networks) --多任务学习-2
- Feed Forward and Backward Run in Deep Convolution Neural Network 论文阅读笔记
- 论文阅读:End-to-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for H
- 1705.Person Re-Identification by Deep Joint Learning of Multi-Loss Classification 论文阅读笔记
- A survey of task allocation and load balance in distributed system阅读笔记
- The Future of Real-Time SLAM and “Deep Learning vs SLAM”
- An Overview of Pages, Blocks and FTLs in a Solid-State Drive (SSD)
- 论文阅读:Segmentation of teeth in CT volumetric dataset by panoramic projection and variational level se