22 2 2017 4 Vol. 22 No. 2 JOURNAL OF HARBIN UNIVERSITY OF SCIENCE AND TECHNOLOGY Apr. 2017 1 2 3 2 1. 150080 2. 150080 3. 150080,,,,, Apriori,,,,,,,, ; ; ; ; DOI: 10. 15938 /j. jhust. 2017. 02. 023 TN911. 73 A 1007-2683 2017 02-0124- 05 Multimode Retrieval of Mammography Based on Association Rules WANG Qian 1 L Ya-nan 2 LI Dong-hong 3 SONG Li-xin 2 1. School of Computer Science and Technology Harbin University of Science and Technology Harbin 150080 China 2. School of Electrical and Electronic Engineering Harbin University of Science and Technology Harbin 150080 China 3. School of Measurement and Control Technology and Communications Engineering Harbin University of Science and Technology Harbin 150080 China Abstract The mammogram case has images of low level features and semantic features. In order to achieve efficient retrieval of breast imaging cases and enhance the certainty of computer aided diagnosis a multi-mode retrieval method based on association rules is proposed in this paper. First of all feature selection algorithm based on the association rules can be used to select the low level features associated with image semantic features to achieve the dimension reduction. The associative rules which between the selected features and the semantic features can be excavated by using the Apriori algorithm. And then associative classifier engine will be used to build the associative classification model depend on the associative rules to capture the visual semantic features. Finally take obtained semantic from the association classification as input semantic combining with the low level 2015-12 - 29 F200912 2010RFXXS026. 1965 E - mail wamgqiam@ 163. com 1989 1988.
2 125 features of image to implement the mammogram case multi-mode retrieval. We conducted experiments comparing by precision and recall rate and relevance ranking average value and so on as the results show multi - mode retrieval method proposed by this paper can effectively improve the performance of breast imaging case retrieval and provide visual semantic features of image by its low-level features. Multi-mode retrieval reduced the semantic gap between image low level features and visual semantic features improved the accuracy of image retrieval and provided more meaningful decision support for doctors. Keywords mammogram association rules feature selection associative classification multi-mode retrieval 0 1 X 2 1 3 1. 1 I = I1 I2 I3 I D 4 - semantic gap 6 7 T T A A B A B I A T IA A B A B 8 P A B sup A P A B conf 9 sup = P A B 1 conf = P A B /P A 2 Agrawal 12 Apriori 10 Apriori 11 R R D C D = Data1 Data2 Datan C = C1 C2 C3 Apriori
126 22 1. 3 R1 R2 R1 R2 1 R1 R2 conf R1 > conf R2 2 conf R1 = conf R2 R1 R2 sup R1 > sup R2 3 conf R1 = conf R2 sup R1 = sup R2 R1 R2 ACE 4 A h F h N h w min 4A h + F h w = 6 4A h + F h + N h 1. 2 W 4 13 StARMiner 1 A h 2 F h T 3 N h T X 4 w min X i X w min 0. 5 N f i X i i μ fi x 1. 4 σ fi x X f i 3 γ min 18 Δμ min Δσ max γ min H 0 Δμ min f i X Δσ max f i X 3 X f i X 32 8 1 H 0 μ fi X = μ fi T - X 3 μ fi X - μ fi T - X Δμ min 4 Δf i X Δσ max 5 ACE 17 associative classifier engine Apriori ACE 1 1 1 14 5 15 2 14 6 15 3 14 4 15 7 15 8 16 1 X DDSM digital database for screening mammography 170 170 61 62
2 127 47 35 170 65 2 2 0. 8182 0. 8500 0. 8750 0. 8511 19 5 10 3 1. 5 3 a 2 3 b 3 2 1 2 7 i i 8 d X i X j = 2 n 2 槡 l = 1 x il - x jl 7 X i X j i j l X l d k i s i j = 2 k = 1 w k d k i j 2 w k k = 1 i 8 j j k 4 k = 1 2 a - w k 0 1 0 3-20
128 22 4-181. 10 b 11 J. 2006 12 AGRAWAL R 1994 1215 487-499. 13 BUGATTI P H RIBEIRO M X 3 1 TANG H L HANKA R IP H H S. Histological Image Retrieval Based on Semantic Content Analysis J. Information Technology in Biomedicine IEEE Transactions on 2003 7 1 26-36. 2 HUANG Y L CHEN D R. Watershed Segmentation for Breast Tumor in 2-D Sonography J. Ultrasound in medicine & biology 2004 30 5 625-632. 3. J. 2013 43 12 1641-1653. 4. J. 2005 35 5 536-540. 5 SMEULDERS A W M WORRING M SANTINI S et al. Content-based Image Retrieval at the End of the Early Years J. Pattern Analysis and Machine Intelligence IEEE Transactions on 2000 22 12 1349-1380. 6. J. 2008 20 8 1085-1096. 7. J. 2003 27 4 234-237. 8. J. 2009 26 6 1237-1240. 9. J. 2012 44 S1 177 J. 2005 25 6 1408-1409... 24 3 401-404. SRIKANT R. Fast Algorithms for Mining Association Rules C / / 20th int. conf. Very Large Data Bases VLDB. TRAINA A J M et al. Contentbased Retrieval of Medical Images by Continuous Feature Selection C / / Computer-Based Medical Systems 2008 CBMS'08 21st IEEE International Symposium on. IEEE 2008 272-277. 14 ZHENG B LU A HARDESTY L A et al. A Method to Improve Visual Similarity of Breast Masses for an Interactive Computer-ai- ded Diagnosis Environment J 111-117. 15 PETRICK N CHAN H P WEI D et al. Automated Detection of Breast Masses on Mammograms Using Adaptive Contrast Enhancement and Texture Classification J. Medical physics 1996 23 10 1685-1696. 16 JIN R MENG B SONG E et al. Computer-aided Detection of Mammographic Masses Based on Content-based Image Retrieval C / /Medical Imaging. International Society for Optics and Photonics 2007 65141W - 65141W - 8. 134. Medical Physics 2006 33 1
134 22 for Next Generation Computer-aided Mammography Reference Image Databases and Evaluation Studies J. International Journal of Computer Assisted Radiology & Surgery 2011 6 6 749-767. 7 PU Jiantao ZHENG Bin KEN Joseph Leader et al. An Ellipse-fitting Based Method for Efficient Registration of Breast Masses on Two Mammographic Views J. Medical Physics 2008 35 2 489 ~ 490. 8. J. 2003 21 10 606 J. 2010 27 4 1578-1580. 9. D. 17. 2011 30-52. 10. D. 18. J. 2013. 2014 50 20 171-175. 11. 19. J. J. 2010 31 5 1121-1124. 2007 30 6 1023-1024. 12. Sech 20. J. J. 2014 14 1 139-142. 13. J. 2015 20 5 25-29. 14 LI Ke YING Ying CHEN NanLi. Infiltrative Breast Cancer Initial Detection Infiltrative Breast Cancer Initial Detection C / / International Conference on Information & Automation Shenyang China 2012 887-890. 15. NCC J. 2013 14 128-129. 16. - 607. J. 2006 33 1 118-119.. 2014 40 2 224-228. ( : ) 檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪檪 128 17 RIBEIRO M X BUGATTI P H TRAINA Jr C et al. Supporting Content-based Image Retrieval and Computer-aided Diagnosis Systems with Association Rule-based Techniques J. Data & Knowledge Engineering 2009 68 12 1370-1382. 18 BI-RADS X J. 2014 25 9 615-618. 19 MENG F J GUO B L. Research on Content-Based Image Retrieval Technology J. Applications Research of Computer 2004 21 7 21-27. 20. X J. 2011 16 1 97-102. ( : )