首页 » 文章 » 文章详细信息
Journal of Healthcare Engineering Volume 2018 ,2018-05-02
Breast Mass Detection in Digital Mammogram Based on Gestalt Psychology
Research Article
Hongyu Wang 1 Jun Feng 1 Qirong Bu 1 Feihong Liu 1 Min Zhang 2 Yu Ren 3 Yi Lv 4
Show affiliations
DOI:10.1155/2018/4015613
Received 2017-09-28, accepted for publication 2018-03-14, Published 2018-03-14
PDF
摘要

Inspired by gestalt psychology, we combine human cognitive characteristics with knowledge of radiologists in medical image analysis. In this paper, a novel framework is proposed to detect breast masses in digitized mammograms. It can be divided into three modules: sensation integration, semantic integration, and verification. After analyzing the progress of radiologist’s mammography screening, a series of visual rules based on the morphological characteristics of breast masses are presented and quantified by mathematical methods. The framework can be seen as an effective trade-off between bottom-up sensation and top-down recognition methods. This is a new exploratory method for the automatic detection of lesions. The experiments are performed on Mammographic Image Analysis Society (MIAS) and Digital Database for Screening Mammography (DDSM) data sets. The sensitivity reached to 92% at 1.94 false positive per image (FPI) on MIAS and 93.84% at 2.21 FPI on DDSM. Our framework has achieved a better performance compared with other algorithms.

授权许可

Copyright © 2018 Hongyu Wang et al. 2018
This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

图表

Sample images from MIAS data set [17].

Sample images from MIAS data set [17].

Sample images from MIAS data set [17].

The clinical diagnosis of breast mass by the radiologist.

The framework of the proposed approach. It can be divided into three stages including sensation integration, semantic integration, and verification. Visual rules used in the framework are modeled and indicated with the labels A, B, C, D, and E.

Visual patches based on Gestalt psychology.

The densification of different patches. Visual patches of (a) mass and normal tissue.

The densification of different patches. Visual patches of (a) mass and normal tissue.

Visual patches with the glandular.

The statistical histogram of homogeneity of negative and positive visual patches.

Sample results of the saliency algorithms. Green denotes the saliency region detected by these algorithms, pink represents the ground truth region containing mass, and white denotes the crossing region between green and pink. (a) Agrawal et al. [16], (b) Achanta and Süsstrunk [55], (c) Murray et al. [56], and (d) the three stages of our method. Stage 1: the fifth column is the detection result of sensation integration. Stage 2: the sixth column is the detection result of semantic integration. Stage 3: the last column is the final detection result (verification) of our method.

Sample results of the saliency algorithms. Green denotes the saliency region detected by these algorithms, pink represents the ground truth region containing mass, and white denotes the crossing region between green and pink. (a) Agrawal et al. [16], (b) Achanta and Süsstrunk [55], (c) Murray et al. [56], and (d) the three stages of our method. Stage 1: the fifth column is the detection result of sensation integration. Stage 2: the sixth column is the detection result of semantic integration. Stage 3: the last column is the final detection result (verification) of our method.

Sample results of the saliency algorithms. Green denotes the saliency region detected by these algorithms, pink represents the ground truth region containing mass, and white denotes the crossing region between green and pink. (a) Agrawal et al. [16], (b) Achanta and Süsstrunk [55], (c) Murray et al. [56], and (d) the three stages of our method. Stage 1: the fifth column is the detection result of sensation integration. Stage 2: the sixth column is the detection result of semantic integration. Stage 3: the last column is the final detection result (verification) of our method.

Sample results of the saliency algorithms. Green denotes the saliency region detected by these algorithms, pink represents the ground truth region containing mass, and white denotes the crossing region between green and pink. (a) Agrawal et al. [16], (b) Achanta and Süsstrunk [55], (c) Murray et al. [56], and (d) the three stages of our method. Stage 1: the fifth column is the detection result of sensation integration. Stage 2: the sixth column is the detection result of semantic integration. Stage 3: the last column is the final detection result (verification) of our method.

The number and percentage of patches/ROIs are counted for each step of our method: (a) plotted on the MIAS data set and (b) plotted on the DDSM data set.

The number and percentage of patches/ROIs are counted for each step of our method: (a) plotted on the MIAS data set and (b) plotted on the DDSM data set.

FROC curves of the proposed method on MIAS and DDSM data sets.

通讯作者

Jun Feng.Department of Information Science and Technology, Northwest University, Xi’an 710127, China, nwu.edu.cn.fengjun@nwu.edu.cn

推荐引用方式

Hongyu Wang,Jun Feng,Qirong Bu,Feihong Liu,Min Zhang,Yu Ren,Yi Lv. Breast Mass Detection in Digital Mammogram Based on Gestalt Psychology. Journal of Healthcare Engineering ,Vol.2018(2018)

您觉得这篇文章对您有帮助吗?
分享和收藏
0

是否收藏?

参考文献
[1] P. Agrawal, M. Vatsa, R. Singh. (2014). Saliency based mass detection from screening mammograms. Signal Processing.99(6):29-47. DOI: 10.3322/ca.2007.0010.
[2] M. Heath, K. Bowyer, D. Kopans, R. Moore. et al.The digital database for screening mammography. :212-218. DOI: 10.3322/ca.2007.0010.
[3] J. Suckling, J. Parker, D. Dance. (1994). The mammographic image analysis society digital mammogram database. International Congress Series Exerpta Medica.1069:375-378. DOI: 10.3322/ca.2007.0010.
[4] G.-B. Huang. (2014). An insight into extreme learning machines: random neurons, random features and kernels. Cognitive Computation.6(3):376-390. DOI: 10.3322/ca.2007.0010.
[5] J. Tang, C. Deng, G.-B. Huang. (2016). Extreme learning machine for multilayer perceptron. IEEE transactions on neural networks and learning systems.27(4):809-821. DOI: 10.3322/ca.2007.0010.
[6] F. Jäkel, M. Singh, F. A. Wichmann, M. H. Herzog. et al.(2016). An overview of quantitative approaches in gestalt perception. Vision Research.126:3-8. DOI: 10.3322/ca.2007.0010.
[7] A. Jemal, R. Siegel, E. Ward. (2008). Cancer statistics, 2008. CA: A Cancer Journal for Clinicians.58(2):71-96. DOI: 10.3322/ca.2007.0010.
[8] K. C. Lee, J. H. Han, Y. U. Song, W. J. Lee. et al.(2015). A fuzzy logic driven multiple knowledge integration framework for improving the performance of expert systems. Intelligent Systems in Accounting Finance and Management.7(4):213-222. DOI: 10.3322/ca.2007.0010.
[9] T. Drew, K. Evans, M. L.-H. Võ, F. L. Jacobson. et al.(2013). Informatics in radiology: what can you see in a single glance and how might this guide visual search in medical images?. Radiographics.33(1):263-274. DOI: 10.3322/ca.2007.0010.
[10] W. Huda, K. M. Ogden. (2004). How do radiographic techniques affect mass lesion detection performance in digital mammography?. Proceedings of the SPIE.5372:372-382. DOI: 10.3322/ca.2007.0010.
[11] X. Gao, Y. Wang, X. Li, D. Tao. et al.(2010). On combining morphological component analysis and concentric morphology model for mammographic mass detection. IEEE Transactions on Information Technology in Biomedicine.14(2):266-273. DOI: 10.3322/ca.2007.0010.
[12] W. E. Polakowski, D. A. Cournoyer, S. K. Rogers. (1997). Computer-aided breast cancer detection and diagnosis of masses using difference of gaussians and derivative-based feature saliency. IEEE Transactions on Medical Imaging.16(6):811-819. DOI: 10.3322/ca.2007.0010.
[13] N. Petrick, H.-P. Chan, B. Sahiner, D. Wei. et al.(1996). An adaptive density-weighted contrast enhancement filter for mammographic breast mass detection. IEEE Transactions on Medical Imaging.15(1):59-67. DOI: 10.3322/ca.2007.0010.
[14] Z. Wang, G. Yu, Y. Kang, Y. Zhao. et al.(2014). Breast tumor detection in digital mammography based on extreme learning machine. Neurocomputing.128(5):175-184. DOI: 10.3322/ca.2007.0010.
[15] R. Salvador, J. L. Lirola, R. Domnguez, M. López. et al.(2004). Pseudo-angiomatous stromal hyperplasia presenting as a breast mass: imaging findings in three patients. Breast.13(5):431-435. DOI: 10.3322/ca.2007.0010.
[16] L. Liu, J. Li, Y. Wang. Breast mass detection with kernelized supervised hashing. :79-84. DOI: 10.3322/ca.2007.0010.
[17] N. M. Basheer, M. M. H. Mohammed. (2013). Segmentation of breast masses in digital mammograms using adaptive median filtering and texture analysis. International Journal of Recent Technology and Engineering.2(1):39-43. DOI: 10.3322/ca.2007.0010.
[18] G. D. Tourassi, B. Harrawood, S. Singh, J. Y. Lo. et al.(2007). Evaluation of information-theoretic similarity measures for content-based retrieval and detection of masses in mammograms. Medical Physics.34(1):140-150. DOI: 10.3322/ca.2007.0010.
[19] I. Vizcano, L. Gadea, L. Andreo. (2001). Short-term follow-up results in 795 nonpalpable probably benign lesions detected at screening mammography. Radiology.219(2):475-483. DOI: 10.3322/ca.2007.0010.
[20] N. R. Mudigonda, R. M. Rangayyan, J. L. Desautels. (2001). Detection of breast masses in mammograms by density slicing and texture flow-field analysis. IEEE Transactions on Medical Imaging.20(12):1215-1227. DOI: 10.3322/ca.2007.0010.
[21] G. Xu, Y. Ding, C. Wu, Y. Zhai. et al.Explore maximal frequent item sets for big data pre-processing based on small sample in cloud computing. International Congress on Ultra Modern Telecommunications and Control Systems and Workshops:235-239. DOI: 10.3322/ca.2007.0010.
[22] A. U. Rehman, N. Chouhan, A. Khan. Diverse and discriminative features based breast cancer detection using digital mammography. :234-239. DOI: 10.3322/ca.2007.0010.
[23] T. Kooi, G. Litjens, G. B. Van. (2016). Large scale deep learning for computer aided detection of mammographic lesions. Medical Image Analysis.35:303-312. DOI: 10.3322/ca.2007.0010.
[24] G. Litjens, T. Kooi, B. E. Bejnordi. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis.42(9):60-68. DOI: 10.3322/ca.2007.0010.
[25] X. He, R. S. Zemel, D. Ray. (2006). Learning and incorporating top-down cues in image segmentation. Lecture Notes in Computer Science.3951:338-351. DOI: 10.3322/ca.2007.0010.
[26] P. Casti, A. Mencattini, M. Salmeri. (2016). Contour-independent detection and classification of mammographic lesions. Biomedical Signal Processing and Control.25:165-177. DOI: 10.3322/ca.2007.0010.
[27] J. Chu, H. Min, L. Liu, W. Lu. et al.(2015). A novel computer aided breast mass detection scheme based on morphological enhancement and SLIC superpixel segmentation. Medical Physics.42(7):3859-3869. DOI: 10.3322/ca.2007.0010.
[28] N. Murray, M. Vanrell, X. Otazu, C. A. Parraga. et al.Saliency estimation using a non-parametric low-level vision model. :433-440. DOI: 10.3322/ca.2007.0010.
[29] A. Oliver, J. Freixenet, J. Marti. (2010). A review of automatic mass detection and segmentation in mammographic images. Medical Image Analysis.14(2):87-110. DOI: 10.3322/ca.2007.0010.
[30] N. Dhungel, G. Carneiro, A. P. Bradley. Automated mass detection in mammograms using cascaded deep learning and random forests. :1-8. DOI: 10.3322/ca.2007.0010.
[31] A. Mencattini, M. Salmeri. (2012). Breast masses detection using phase portrait analysis and fuzzy inference systems. International Journal of Computer Assisted Radiology and Surgery.7(4):573-583. DOI: 10.3322/ca.2007.0010.
[32] N. Dhungel, G. Carneiro, A. P. Bradley. Fully automated classification of mammograms using deep residual neural networks. :310-314. DOI: 10.3322/ca.2007.0010.
[33] S. L. Fischer. (2012). The gestalt research tradition: figure and ground. Gestalt Review.16(1):3-6. DOI: 10.3322/ca.2007.0010.
[34] R. J. Ferrari, R. M. Rangayyan, J. L. Desautels, R. Borges. et al.(2004). Automatic identification of the pectoral muscle in mammograms. IEEE Transactions on Medical Imaging.23(2):232-245. DOI: 10.3322/ca.2007.0010.
[35] P. Casti, A. Mencattini, M. Salmeri. (2013). Estimation of the breast skin-line in mammograms using multidirectional Gabor filters. Computers in Biology and Medicine.43(11):1870-1881. DOI: 10.3322/ca.2007.0010.
[36] U. Rutishauser, D. Walther, C. Koch, P. Perona. et al.Is bottom-up attention useful for object recognition?. :37-44. DOI: 10.3322/ca.2007.0010.
[37] D. S. Gowri, T. Amudha. A review on mammogram image enhancement techniques for breast cancer detection. :47-51. DOI: 10.3322/ca.2007.0010.
[38] C.-C. Jen, S.-S. Yu. (2015). Automatic detection of abnormal mammograms in mammographic images. Expert Systems with Applications.42(6):3048-3055. DOI: 10.3322/ca.2007.0010.
[39] J. Arevalo, M. A. G. Lopez, M. A. G. Lopez. (2016). Representation learning for mammography mass lesion classification with convolutional neural networks. Computer Methods and Programs in Biomedicine.127:248-257. DOI: 10.3322/ca.2007.0010.
[40] H. Cheng, X. Shi, R. Min, L. Hu. et al.(2006). Approaches for automated detection and classification of masses in mammograms. Pattern Recognition.39(4):646-668. DOI: 10.3322/ca.2007.0010.
[41] T. A. Ngo, Z. Lu, G. Carneiro. (2017). Combining deep learning and level set for the automated segmentation of the left ventricle of the heart from cardiac cine magnetic resonance. Medical Image Analysis.35:159. DOI: 10.3322/ca.2007.0010.
[42] H. Weedon-Fekjær, P. R. Romundstad, L. J. Vatten. (2014). Modern mammography screening and breast cancer mortality: population study. British Medical Journal.348:g3701-g3708. DOI: 10.3322/ca.2007.0010.
[43] C. Gallego-Ortiz, A. L. Martel. (2015). Improving the accuracy of computer-aided diagnosis for breast MR imaging by differentiating between mass and nonmass lesions. Radiology.278(3):679-688. DOI: 10.3322/ca.2007.0010.
[44] N. Dhungel, G. Carneiro, A. P. Bradley. (2017). A deep learning approach for the analysis of masses in mammograms with minimal user intervention. Medical Image Analysis.37:114-128. DOI: 10.3322/ca.2007.0010.
[45] P. S. Vikhe, V. R. Thool. (2016). Mass detection in mammographic images using wavelet processing and adaptive threshold technique. Journal of Medical Systems.40(4):82. DOI: 10.3322/ca.2007.0010.
[46] N. Karssemeijer, G. M. te Brake. (1996). Detection of stellate distortions in mammograms. IEEE Transactions on Medical Imaging.15(5):611-619. DOI: 10.3322/ca.2007.0010.
[47] A. Krizhevsky, I. Sutskever, G. E. Hinton. Imagenet classification with deep convolutional neural networks. :1097-1105. DOI: 10.3322/ca.2007.0010.
[48] G. Kom, A. Tiedeu, M. Kom. (2007). Automated detection of masses in mammograms by local adaptive thresholding. Computers in Biology and Medicine.37(1):37-48. DOI: 10.3322/ca.2007.0010.
[49] N. H. Eltonsy, G. D. Tourassi, A. S. Elmaghraby. (2007). A concentric morphology model for the detection of masses in mammography. IEEE Transactions on Medical Imaging.26(6):880-889. DOI: 10.3322/ca.2007.0010.
[50] J. Wagemans, J. Feldman, S. Gepshtein. (2012). A century of gestalt psychology in visual perception: II. Conceptual and theoretical foundations. Psychological Bulletin.138(6):1172-1217. DOI: 10.3322/ca.2007.0010.
[51] S. H. Kang. (2010). From gestalt theory to image analysis: a probabilistic approach [book review of mr2365342]. Siam Review.52(2):399-400. DOI: 10.3322/ca.2007.0010.
[52] P. Li, J. Feng, Q. Bu, F. Liu. et al.(2015). Multi-object segmentation for abdominal CT image based on visual patch classification. CCF Chinese Conference on Computer Vision:130-138. DOI: 10.3322/ca.2007.0010.
[53] R. Achanta, A. Shaji, K. Smith, A. Lucchi. et al.(2012). SLIC superpixels compared to state-of-the-art superpixel methods. IEEE Transactions on Pattern Analysis and Machine Intelligence.34(11):2274-2282. DOI: 10.3322/ca.2007.0010.
[54] Y. Kinoshita, M. Koppen, K. Yoshida. Perception of image similarity considering gestalt theory. :171-177. DOI: 10.3322/ca.2007.0010.
[55] A. Esteva, B. Kuprel, R. A. Novoa. (2017). Corrigendum: dermatologist-level classification of skin cancer with deep neural networks. Nature.542(7639):115-118. DOI: 10.3322/ca.2007.0010.
[56] R. Bharath, L. Z. J. Nicholas, X. Cheng. Scalable scene understanding using saliency-guided object localization. :1503-1508. DOI: 10.3322/ca.2007.0010.
[57] B. Kurt, V. V. Nabiyev, K. Turhan. (2014). A novel automatic suspicious mass regions identification using Havrda & Charvat entropy and Otsu’s N thresholding. Comput Methods Programs Biomed.114(3):349-360. DOI: 10.3322/ca.2007.0010.
[58] Y. Lecun, Y. Bengio, G. Hinton. (2015). Deep learning. Nature.521(7553):436-444. DOI: 10.3322/ca.2007.0010.
[59] R. Achanta, S. Süsstrunk. Saliency detection using maximum symmetric surround. :2653-2656. DOI: 10.3322/ca.2007.0010.
文献评价指标
浏览 205次
下载全文 56次
评分次数 0次
用户评分 0.0分
分享 0次