access icon free Anomaly classification in digital mammography based on multiple-instance learning

© The Institution of Engineering and Technology
Received 03/06/2017, Accepted 29/10/2017, Revised 13/10/2017, Published 03/11/2017

Cancer tissues in mammography images exhibit abnormal regions; it is of great clinical importance to label a mammography image as having cancerous regions or not, perform the corresponding image segmentation. However, the detailed annotation of the cancer region is often an ambiguous and challenging task. The authors describe a fully automatic computer-aided detection and diagnosis (CAD) system to detect and classify breast cancer as malignant or benign, by using mammography and building on the multiple-instance learning (MIL) algorithms, which has been confirmed beneficial for radiologist decision sustenance. Traditional learning methods require great effort to annotate the training data by costly manual labelling and specialised computational models to detect these annotations during the test. The proposed CAD system simultaneously performs pixel-level segmentation (suspicious versus normal tissue) and image-level classification (benign versus malignant image). The set-up of the proposed system is in order: automatically segmented regions of interest (ROIs). Then, features derived from ROIs detected such as textural features and shape features are selected and extracted from each region and combined them to classify ROIs as ‘benign’ or ‘malignant’, by implementing MIL algorithms. Experimental results demonstrate the efficiency and robustness of the proposed CAD system compared with previous work in the literature.

Inspec keywords: cancer; feature selection; medical image processing; image segmentation; learning (artificial intelligence); mammography; feature extraction; image classification

Other keywords: anomaly classification; pixel-level segmentation; image segmentation; CAD system; cancer tissues; textural feature extraction; Mammography Image Analysis Society database; Digital Database for Screening Mammography; MIL algorithms; fully automatic computer-aided detection and diagnosis system; digital mammography; feature selection; multiple-instance learning; image-level classification; shape feature extraction

Subjects: X-rays and particle beams (medical uses); Image recognition; Biology and medical computing; Computer vision and image processing techniques; Patient diagnostic methods and instrumentation; X-ray techniques: radiography and computed tomography (biomedical imaging/measurement)

References

    1. 1)
      • 10. Malich, A., Schmidt, S., Fischer, D.R., et al: ‘The performance of computer-aided detection when analyzing prior mammograms of newly detected breast cancers with special focus on the time interval from initial imaging to detection’, Eur. J. Radiol., 2009, 69, (3), pp. 574–578.
    2. 2)
      • 1. Jemal, A., Siegel, R., Xu, J., et al: ‘Cancer statistics, 2010’, CA Cancer J. Clin., 2010, 60, (5), pp. 277–300.
    3. 3)
      • 38. Reyad, Y.A., Berbar, M.A., Hussain, M.: ‘Comparison of statistical, LBP, and multi-resolution analysis features for breast mass classification’, J. Med. Syst., 2014, 38, (9), p. 100.
    4. 4)
      • 27. Jen, C.-C., Yu, S.-S.: ‘Automatic detection of abnormal mammograms in mammographic images’, Expert Syst. Appl., 2015, 42, (6), pp. 3048–3055.
    5. 5)
      • 77. Hartigan, J.A., Wong, M.A.: ‘A K-means clustering algorithm’, J. R. Stat. Soc. (Appl. Stat.), 1979, 28, (1), pp. 100–108.
    6. 6)
      • 44. Quellec, G., Lamard, M., Cazuguel, G., et al: ‘Wavelet optimization for content-based image retrieval in medical databases’, Med. Image Anal., 2010, 14, (2), pp. 227–241.
    7. 7)
      • 9. Malich, A., Sauner, D., Marx, C., et al: ‘Influence of breast lesion size and histologic findings on tumor detection rate of a computer-aided detection system’, Radiology, 2003, 228, (3), pp. 851–856.
    8. 8)
      • 34. Wei, C.-H., Chen, S.Y., Liu, X.: ‘Mammogram retrieval on similar mass lesions’, Comput. Methods Programs Biomed., 2012, 106, (3), pp. 234–248.
    9. 9)
      • 65. Homer, M.J.: ‘Mammographic interpretation: a practical approach’ (McGraw-Hill, Boston, MA, 1997, 2nd edn.), pp. 1–6.
    10. 10)
      • 12. Mazurowski, M.A., Habas, P.A., Zurada, J.M., et al: ‘Decision optimization of case-based computer-aided decision systems using genetic algorithms with application to mammography’, Phys. Med. Biol., 2008, 53, (4), pp. 895–908.
    11. 11)
      • 37. Kim, D.H., Lee, S.H., Ro, Y.M.: ‘Mass type-specific sparse representation for mass classification in computer-aided detection on mammograms’, BioMed. Eng. Online, 2013, 12, (Suppl 1), p. S3, http://www.biomedical-engineering-online.com/content/12/S1/S3.
    12. 12)
      • 2. Elmoufidi, A., El Fahssi, K., Jai-Andaloussi, S., et al: ‘Automatic detection of suspicious lesions in digital X-ray mammograms’. Proc. Int. Symp. on Ubiquitous Networking (UNet2016), Casablanca, Morocco, May–June 2016, pp. 375–385.
    13. 13)
      • 56. Tong, T., Wolz, R., Gao, Q., et al: ‘Multipleinstance learning for classification of dementia in brain MRI’. Proc. Int. Conf. on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2013, vol. 16, Pt 2, pp. 599–606.
    14. 14)
      • 43. Nithya, R., Santhi, B.: ‘Classification of normal and abnormal patterns in digital mammograms for diagnosis of breast cancer’, Int. J. Comput. Appl., 2011, 28, (6), pp. 21–25.
    15. 15)
      • 23. Lladó, X., Oliver, A., Freixenet, J., et al: ‘A textural approach for mass false positive reduction in mammography’, Comput. Med. Imaging Graph., 2009, 33, (6), pp. 415–422.
    16. 16)
      • 24. Hussain, M.: ‘False positive reduction using Gabor feature subset selection’. Proc. Int. Conf. on Information Science and Applications (ICISA), Suwon, South Korea, June 2013, pp. 1–5.
    17. 17)
      • 55. McCann, M.T., Bhagavatula, R., Fickus, M.C., et al: ‘Automated colitis detection from endoscopic biopsies as a tissue screening tool in diagnostic pathology’. Proc. Int. Conf. on Image Processing (ICIP), Orlando, FL, USA, September–October 2012, pp. 2809–2812.
    18. 18)
      • 39. Sharma, S., Khanna, P.: ‘Computer-aided diagnosis of malignant mammograms using Zernike moments and SVM’, J. Digit. Imaging., 2015, 28, (1), pp. 77–90.
    19. 19)
      • 67. Amores, J.: ‘Multiple instance classification: review, taxonomy and comparative study’, Artif. Intell., 2013, 201, pp. 81–105.
    20. 20)
      • 51. Xu, Y., Zhang, J., Chang, E.I.-C., et al: ‘Context constrained multiple instance learning for histopathology image segmentation’. Proc. Int. Conf. on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Nice, France, October 2012, pp. 623–630.
    21. 21)
      • 42. Zhu, W., Lou, Q., Vang, Y.S., et al: ‘Deep multi-instance networks with sparse label assignment for whole mammogram classification’. 20th Int. Conf. on Medical Image Computing and Computer-Assisted Intervention, Volume: MICCAI 2017, Beijing, China, arXiv:1612.05968v1 [cs.CV], 18 December 2016.
    22. 22)
      • 63. Lu, P., Liu, W., Xu, W., et al: ‘Multi-instance learning for mass retrieval in digitized mammograms’. Proc. Int. Conf. on SPIE Medical Imaging, San Diego, California, USA, February 2012, pp. 831523–831528.
    23. 23)
      • 19. García-Manso, A., García-Orellana, C.J., González-Velasco, H.M., et al: ‘Consistent performance measurement of a system to detect masses in mammograms based on blind feature extraction’, BioMed. Eng. Online, 2013, 12, p. 2, http://www.biomedical-engineering-online.com/content/12/1/2.
    24. 24)
      • 59. Shi, C., Zhang, H., Chen, Y., et al: ‘Multiple instance learning for computer aided detection and diagnosis of gastric cancer with dual-energy CT imaging’, J. Biomed. Inform., 2015, 57, pp. 358–368.
    25. 25)
      • 69. Zhou, Z.-H., Sun, Y.-Y., Li, Y.-F.: ‘Multi-instance learning by treating instances as non-I.I.D. samples’. Proc. Int. Conf. on Machine Learning (ICML), Montreal, Quebec, Canada, June 2009, pp. 1249–1256.
    26. 26)
      • 62. Krishnapuram, B., Stoeckel, J., Raykar, V., et al: ‘Multiple-instance learning improves CAD detection of masses in digital mammography’. Proc. Int. Conf. on International Workshop on Digital Mammography (IWDM), Tucson, AZ, USA, July 2008, pp. 350–357.
    27. 27)
      • 41. Oliveira, J.E.E., Albuquerque Araujo, A., Deserno, T.M.: ‘Content-based image retrieval applied to BI-RADS tissue classification in screening mammography’, World J. Radiol., 2011, 3, (1), pp. 24–31.
    28. 28)
      • 80. Suckling, J., Parker, J., Dance, D.R.., et al: ‘The mammographic image analysis society digital mammogram database’. Proc. Int. Workshop on Digital Mammography Exerpta Medica, Amsterdam, 1994, pp. 375–378.
    29. 29)
      • 35. Vadivel, A., Surendiran, B.: ‘A fuzzy rule-based approach for characterization of mammogram masses into BI-RADS shape categories’, Comput. Biol. Med., 2013, 43, (4), pp. 259–267.
    30. 30)
      • 48. Kandemir, M., Hamprecht, F.A.: ‘Computer-aided diagnosis from weak supervision: a benchmarking study’, Comput. Med. Imaging Graph., 2015, 42, pp. 44–50.
    31. 31)
      • 45. Quellec, G., Lamard, M., Cazuguel, G., et al: ‘Adaptive nonseparable wavelet transform via lifting and its application to content-based image retrieval’, IEEE Trans. Image Process., 2010, 19, (1), pp. 25–35.
    32. 32)
      • 73. Dietterich, T.G., Lathrop, R.H., Lozano-Pérez, T.: ‘Solving the multiple instance problem with axis-parallel rectangles’, Artif. Intell., 1997, 89, (12), pp. 31–71.
    33. 33)
      • 66. D'Orsi, C.J., Sickles, E.A., Mendelson, E.B., et al: ‘ACR BI-RADS atlas, breast imaging reporting and data system’ (American College of Radiology, Reston, VA, 2013edn.).
    34. 34)
      • 53. Venkatesan, R., Chandakkar, P., Li, B., et al: ‘Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features’. Proc. Int. Conf. of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA, August-September 2012, pp. 1462–1465.
    35. 35)
      • 11. Ganesan, K., Acharya, U.R., Chua, C.K., et al: ‘Computer-aided breast cancer detection using mammograms: a review’, IEEE Rev. Biomed. Eng., 2013, 6, pp. 77–98.
    36. 36)
      • 14. Elmoufidi, A., El Fahssi, K., Jai-Andaloussi, S., et al: ‘Detection of regions of interests in mammograms by using local binary pattern, dynamic K-means algorithm and gray level co-occurrence matrix’. Proc. Int. Conf. on Next Generation Networks and Services (NGNS'14), Casablanca, Morocco, May 2014, pp. 28–30.
    37. 37)
      • 31. Yoon, S., Kim, S.: ‘AdaBoost-based multiple SVM-RFE for classification of mammograms in DDSM’. 2008 IEEE Int. Conf. on Bioinformatics and Biomedicine Workshops (BIBMW 2008), Philadelphia, PA, USA, November 2008, pp. 75–82.
    38. 38)
      • 75. Viola, P., Platt, J., Zhang, C.: ‘Multiple instance boosting for object detection’. Proc. Int. Conf. on Neural Information Processing Systems (NIPS), Vancouver, British Columbia, Canada, December 2005, pp. 1417–1424.
    39. 39)
      • 8. Brem, R.F.., Baum, J., Lechner, M., et al: ‘Improvement in sensitivity of screening mammography with computer-aided detection’, a multiinstitutional trial, AJR Am. J. Roentgenol., 2003, 181, (3), pp. 687–693.
    40. 40)
      • 16. Bator, M., Nieniewski, M.: ‘Detection of cancerous masses in mammograms by template matching: optimization of template brightness distribution by means of evolutionary algorithm’, J. Digit. Imaging, 2012, 25, (1), pp. 162–172.
    41. 41)
      • 57. Tong, T., Wolz, R., Gao, Q., et al: ‘Multiple instance learning for classification of dementia in brain MRI’, Med. Image Anal., 2014, 18, (5), pp. 808–818.
    42. 42)
      • 74. Cortes, C., Vapnik, V.: ‘Support-vector networks’, Mach. Learn., 1995, 20, (3), pp. 273–297.
    43. 43)
      • 58. Melendez, J., van Ginneken, B., Maduskar, P., et al: ‘A novel multiple-instance learning-based approach to computer-aided detection of tuberculosis on chest X-rays’, IEEE Trans. Med. Imaging, 2015, 34, (1), pp. 179–192.
    44. 44)
      • 50. Kandemir, M., Zhang, C., Hamprecht, F.A.: ‘Empowering multiple instance histopathology cancer diagnosis by cell graphs’. Proc. Int. Conf. on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Boston, MA, USA, September 2014, pp. 228–235.
    45. 45)
      • 15. Hapfelmeier, A., Horsch, A.: ‘Image feature evaluation in two new mammography CAD prototypes’, Int. J. Comput. Assist. Radiol. Surg., 2011, 6, (6), pp. 721–735.
    46. 46)
      • 7. Birdwell, R., Ikeda, D., O'Shaughnessy, K., et al: ‘Mammographic characteristics of 115 missed cancers later detected with screening mammography and the potential utility of computer-aided detection’, Radiology, 2001, 219, (1), pp. 192–202.
    47. 47)
      • 3. Ferlay, F., Héry, C., et al: ‘Global burden of breast cancer’, in Li, C. (Ed): ‘Breast cancer epidemiology’ (Springer, 2010), pp. 1–19.
    48. 48)
      • 32. Biswas, S.K., Mukherjee, D.P.: ‘Recognizing architectural distortion in mammogram: a multiscale texture modeling approach with GMM’, IEEE Trans. Biomed. Eng., 2011, 58, (7), pp. 2023–2030.
    49. 49)
      • 13. Costa, D.D., Campos, L.F., Barros, A.K.: ‘Classification of breast tissue in mammograms using efficient coding’, Biomed. Eng. Online, 2011, 10, (1), p. 55.
    50. 50)
      • 52. Dundar, M.M., Fung, G., Krishnapuram, B., et al: ‘Multiple instance learning algorithms for computer-aided detection’, IEEE Trans. Biomed. Eng., 2008, 55, (3), pp. 1015–1021.
    51. 51)
      • 29. Elmoufidi, A., El Fahssi, K., Jai-Andaloussi, S., et al: ‘Automatically density based breast segmentation for mammograms by using dynamic K-means algorithm and seed based region growing’. Proc. IEEE Int. Conf. on Instrumentation and Measurement Technology (I2MTC), PISA, ITALY, MAY 2015, pp. 533–538.
    52. 52)
      • 76. Babenko, B., Dollár, P., Tu, Z., et al: ‘Simultaneous learning and alignment: multi-instance and multi-pose learning’. Proc. Int. European Conf. on Computer Vision (ECCV), Marseille France, October 2008, pp. 1–15.
    53. 53)
      • 49. Xu, Y., Zhu, J.-Y., Chang, E.I.-C., et al: ‘Weakly supervised histopathology cancer image segmentation and classification’, Med. Image Anal., 2014, 18, (3), pp. 591–604.
    54. 54)
      • 21. de Oliveira, F.S.S., de Carvalho Filho, A.O., Silva, A.C., et al: ‘Classification of breast regions as mass and non-mass based on digital mammograms using taxonomic indexes and SVM’, Comput. Biol. Med., 2015, 57, (C), pp. 42–53.
    55. 55)
      • 64. Li, C., Lam, K.M., Zhang, L., et al: ‘Mammogram microcalcification cluster detection by locating key instances in a multi-instance learning framework’. Proc. IEEE Int. Conf. on Signal Processing, Communication and Computing (ICSPCC), Hong Kong, China, August 2012, pp. 175–179.
    56. 56)
      • 6. Quellec, G., Lamard, M., Cozic, M., et al: ‘Multiple-instance learning for anomaly detection in digital mammography’, IEEE Trans. Med. Imaging, 2016, 35, (7), pp. 1604–1614.
    57. 57)
      • 28. Jai-Andaloussi, S., Sekkaki, A., Quellec, G., et al: ‘Mass segmentation in mammograms by using bidimensional empirical mode decomposition BEMD’. Proc. Int. Conf. of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, Japan, July 2013, pp. 5441–5444.
    58. 58)
      • 17. Elmoufidi, A., El Fahssi, K., Jai-Andaloussi, S., et al: ‘Detection of regions of interest in mammograms by using local binary pattern and dynamic K-means algorithm’, Int. J. Image Video Process.: Theory Appl., 2014, 1, (1), pp. 11–18.
    59. 59)
      • 79. Heath, M., Bowyer, K., Kopans, D., et al: ‘The digital database for screening mammography’. Proc. Int. Workshop on Digital Mammography, Toronto Canada, June 2000, pp. 212–218.
    60. 60)
      • 25. Braz Junior, G., Vieira da Rocha, S., Gattass, M., et al ‘A mass classification using spatial diversity approaches in mammography images for false positive reduction’, Expert Syst. Appl., 2013, 40, (18), pp. 7534–7543.
    61. 61)
      • 18. Gargouri, N., Dammak Masmoudi, A., Sellami Masmoudi, D., et al: ‘A new GLLD operator for mass detection in digital mammograms’, Int. J. Biomed. Imaging, 2012, 2012, (765649), p. 13, http://dx.doi.org/10.1155/2012/765649.
    62. 62)
      • 20. Pereira, D.C., Ramos, R.P., do Nascimento, M.Z.: ‘Segmentation and detection of breast cancer in mammograms combining wavelet analysis and genetic algorithm’, Comput. Methods Programs Biomed., 2014, 114, (1), pp. 88–101.
    63. 63)
      • 47. Quellec, G., Lamard, M., Cochener, B., et al: ‘Realtime task recognition in cataract surgery videos using adaptive spatiotemporal polynomials’, IEEE Trans. Med. Imaging, 2015, 34, (4), pp. 877–887.
    64. 64)
      • 61. Ding, J., Cheng, H.D., Huang, J., et al: ‘Breast ultrasound image classification based on multiple-instance learning’, J. Digit. Imaging, 2012, 25, (5), pp. 620–627.
    65. 65)
      • 68. Foulds, J., Frank, E.: ‘A review of multi-instance learning assumptions’, Knowl. Eng. Rev., 2010, 25, (1), pp. 1–25.
    66. 66)
      • 72. Maron, O., Lozano-Pérez, T.: ‘A framework for multiple-instance learning’. Proc. Int. Conf. on Advances in Neural Information Processing Systems (NIPS), Denver, Colorado, USA, 1998, pp. 570–576.
    67. 67)
      • 78. Mason, L., Baxter, J., Bartlett, P.L., et al: ‘Boosting algorithms as gradient descent’. Proc. Int. Conf. on Advances in Neural Information Processing Systems, Denver, CO, USA, 29 November–4 December 1999, pp. 512–518.
    68. 68)
      • 40. Jiang, Y., Nishikawa, R.M.., Wolverton, D.E.., et al: ‘Malignant and benign clustered microcalcifications: automated feature analysis and classification’, Radiology, 1996, 198, (3), pp. 671–678.
    69. 69)
      • 70. Wang, J., Zucker, J.-D.: ‘Solving the multiple-instance problem: a lazy learning approach’. Proc. Int. Conf. on Machine Learning (ICML), Stanford, California, USA, 29June —02 July 2000, pp. 1119–1125.
    70. 70)
      • 4. Jalalian, A., Mashohor, S.B., Mahmud, H.R., et al: ‘Computer-aided detection/diagnosis of breast cancer in mammography and ultrasound: a review’, Clin. Imaging, 2013, 37, (3), pp. 420–426.
    71. 71)
      • 71. Andrews, S., Tsochantaridis, I., Hofmann, T.: ‘Support vector machines for multiple-instance learning’. Proc. Int. Conf. on Neural Information Processing Systems (NIPS), MIT, Cambridge, MA, December 2002, pp. 577–584.
    72. 72)
      • 22. Elmoufidi, A., El Fahssi, K., Jai Andaloussi, S., et al: ‘Automatic diagnosing of suspicious lesions in digital mammograms’, Int. J. Adv. Comput. Sci. Appl., 2016, 7, (5), pp. 510–518.
    73. 73)
      • 46. Quellec, G., Lamard, M., Abràmoff, M. D., et al: ‘A multiple-instance learning framework for diabetic retinopathy screening’, Med. Image Anal., 2012, 16, (6), pp. 1228–1240.
    74. 74)
      • 5. Burhenne, L., Wood, S., D'Orsi, C., et al: ‘Potential contribution of computer-aided detection to the sensitivity of screening mammography’, Radiology, 2000, 215, (2), pp. 554–562.
    75. 75)
      • 36. Moura, D.C., Guevara Lopez, M.A.: ‘An evaluation of image descriptors combined with clinical data for breast cancer diagnosis’, Int. J. Comput. Assist. Radiol. Surg., 2013, 8, (4), pp. 561–574.
    76. 76)
      • 54. Jiang, H., Zheng, R., Yi, D., et al: ‘A novel multi-instance learning approach for liver cancer recognition on abdominal CT images based on CPSO-SVM and IO’, Comput. Math. Methods. Med., 2013, 2013, Article ID 434969, p. 10, http://dx.doi.org/10.1155/2013/434969.
    77. 77)
      • 33. Zhang, Y., Tomuro, N., Furst, J., et al: ‘Building an ensemble system for diagnosing masses in mammograms’, Int. J. Comput. Assist. Radiol. Surg., 2012, 7, (2), pp. 323–329.
    78. 78)
      • 60. Azar, J.C., Simonsson, M., Bengtsson, E., et al: ‘Automated classification of glandular tissue by statistical proximity sampling’, Int. J. Biomed. Imaging, 2015, Article ID 943104, p. 11, http://dx.doi.org/10.1155/2015/943104.
    79. 79)
      • 30. Elmoufidi, A., El Fahssi, K., Jai-Andaloussi, S., et al: ‘Evaluate dynamic Kmeans algorithm for the segmentation of different breast regions in mammogram based on density by using seed region growing technique’, J. Theor. Appl. Inf. Technol., 2015, 72, (2), pp. 280–288.
    80. 80)
      • 26. Zyout, I., Czajkowska, J., Grzegorzek, M.: ‘Multi-scale textural feature extraction and particle swarm optimization based model selection for false positive reduction in mammography’, Comput. Med. Imaging Graph., 2015, 46, (2), pp. 95–107.
    81. 81)
      • 81. Tan, M., Jiantao, Pu., Bin, Z.: ‘Optimization of breast mass classification using sequential forward floating selection (SFFS) and a support vector machine (SVM) mode’, Int. J. Comput. Assist. Radiol. Surg., 2015, 9, (6), pp. 1005–1020.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-ipr.2017.0536
Loading

Related content

content/journals/10.1049/iet-ipr.2017.0536
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading