臺灣博碩士論文加值系統

近年來，由於用藥疏失的案例發生，使得民眾對於用藥安全的相關知識需求逐漸增加。口服藥物可以根據不同成份、含量、用途分為許多種類。但民眾並不是醫療人員具有專業的藥物資訊能夠辨識藥物。因此只能利用藥物名稱與描述藥物外觀的特徵等關鍵字，藉此搜尋藥物的相關資訊。目前各大醫療機構如區域型醫院，大多都有提供關鍵字搜尋的藥物辨識系統。然而，人們在描述藥物特徵都是依照各自的觀點來描述藥物特徵，如彩色、形狀的描述等。這些藥物的描述並無一致性說法，使得系統無法成功辨識進而造成錯誤。本研究介紹現有的藥物辨識系統，並提出一套以內容為基礎的影像擷取技術，透過取得口服藥物影像中的色彩、形狀、比例、大小、紋理特徵。利用權重歐基理德距離公式計算藥物的相似度測量，且根據模糊規則調整權重，建立一個自動化藥物影像辨識系統(Automatic Drug Image Identification System, ADIIS)提供藥物辨識。使用者只需拍攝藥物的數位影像，系統便可以自動辨識藥物。為能夠完善的提供藥物資訊與避免辨識的錯誤，本系統列舉出前十名辨識率最高的藥物做為結果，輔助使用者有效的辨識藥物。經由實驗證明，在第一名的正確率能達到92.6%，且在順序七以前皆能成功辨識，證明本研究在於口服藥物辨識是有效且具可行性。

In order to take medicines safely, it’s better for a patient to know drugs information first. Drugs can be categorized into many kinds, such as different compositions, content and shapes. However, users do not always possess or comprehend professional drug facts. Many drug recognition systems offer keyword searches, but are difficult for users to understand the medications’ names. One feasible way would be letting users describe the features of drugs according to their appearance, such as color, shape, etc. In this paper, we propose an automatic drug image identification system (ADIIS) based on multiple image features. ADIIS is able to improve drug identification errors as well as provide more accurate drug information. In our primary experiments, by using an image, the system is able to retrieve the top ten similar drugs for the user to identify the specific one. Experiment results show that the first one of the ten drugs identified by ADIIS up to 92.6% correct rates. In addition, the system can successfully identify drug image in the top 7 of outcome.

Table of Contents
中文摘要 I
Abstract II
誌謝 III
Table of Contents VI
List of Tables VIII
List of Figures IX
1. Introduction 1
1.1 Background and Motivation 1
1.2 Objective 2
1.3 The Framework of Thesis 3
2. Literature Review 4
2.1 Content Based Image Retrieval 4
2.2 Related of Drug Identification System 5
3. Automatic Drug Image Identification System 9
3.1 Pre-processing 9
3.2 Feature Extraction 11
3.2.1 Color 12
3.2.2 Shape 14
3.2.3 Ratio 17
3.2.4 Magnitude 18
3.2.5 Texture 20
3.3 Similarity Measure 21
3.4 Relevance Feedback 28
4. Experimental and Discussions 34
4.1 Experimental Data and Environment 34
4.2 Experimental 35
4.3 Discussion 47
5. Conclusions and Future Works 50
References 51
Publication 57

List of Tables
Table 1 Related Drug Identification Research Collate and Description, and Use Features of The Research 8
Table 2 Ten Kinds of Colors Mapping Value 13
Table 3 Ten Kinds of Shapes Mapping Value 16
Table 4 A List of Experimental with Computer Hardware 34
Table 5 Tem Times Test of Results 38
Table 6 The Relevance Feedback by Randomly Selecting Result 39
Table 7 Ten Times Test of White Circular Drug Results 40
Table 8 The Relevance Feedback by White Circular Drug Results 41
Table 9 Five Times Test of Non-white Circular Drug Results 42
Table 10 The Relevance Feedback by Non-white Circular Drug Results 43
Table 11 Ten Times Test of Circular Drug Results 44
Table 12 The Relevance Feedback by Circular Drug Results 45
Table 13 Five Times Test of Capsules Results 46
Table 14 The Relevance Feedback by Capsules Results 47

List of Figures
Figure 1 The Medication Information Web Sit of Drugs Identification Use Name, From, Color and Mark. 5
Figure 2 The Workflow of Automatic Drug Image Identification System 9
Figure 3 The Drug Rotate Angle for Preprocessing 10
Figure 4 Preprocessing Instructions Related to Processes 11
Figure 5 Five Different Types of EHD 15
Figure 6 Back-propagation Neural Network for Shape of Drugs Classification 17
Figure 7 Same Shape But Different Ratio Drugs 18
Figure 8 The Background to Take Drug Images, and Each Grid is 1×1 mm 19
Figure 9 A Gabor Filter 21
Figure 10 The Relationship of White Circular Drugs Weight 24
Figure 11 Fuzzy Membership about Oval-type Drugs Weight 25
Figure 12 The Workflow of Similarity Measure Based on Rules 27
Figure 13 The Workflow of Relevance Feedback in Our System 30
Figure 14 The Output of ADIIS System. 32
Figure 15 The Result of Users Selected The Feedback to ADIIS 32
Figure 16 The Finish Result of The System. 33
Figure 17 A Environment of Taken Drugs 35
Figure 18 Three Times Test to Randomly Selecting of Results 37
Figure 19 The Accuracy Rate for Randomly Selected Tests 48
Figure 20 The Accuracy Rate for White Circular Drug Tests 49

References
[1]A. Abdullah, R. C. Veltkamp, and M. A. Wiering (2010), “Fixed partitioning and salient points with MPEG-7 cluster correlograms for image categorization,” Pattern Recognition, Vol. 43, No. 3, pp. 650-662.
[2]S. Arivazhagan, L. Ganesan, and S. P. Priyal (2006), “Texture classification using Gabor wavelets based rotation invariant features,” Pattern Recognition Letters, Vol. 27, No. 16, pp. 1976-1982.
[3]W. Chen, P. J. Chao, and H. L. Lin (2007), “Drug identification by network adaptive content-based image retrieval,” The Journal of Health Science, Vol. 9, No. 2, pp. 133-145.
[4]J. X. Du, D. S. Huang, X. F. Wang, and X. G (2007), “Shape recognition based on neural networks trained by differential evolution algorithm,” Neurocomputing, Vol. 70, No. 4-6, pp. 896-903.
[5]T. Gevers and A. W. M. Smeulders (1999), “Color-based object recognition,” The Journal of The Pattern Recognition Society, Vol. 32, pp. 453-464.
[6]X. Gao, B. Xiao, D. Tao, and X. Li (2008), “Image categorization: graph edit distance + Edge direction histogram,” Pattern Recognition, Vol. 41, No. 10, pp. 3179-3191.
[7]Z. Geradts, H. Hardy, A. Poorman, and J. Bijhold (2001), “Evaluation of contents based image retrieval methods for a database of logos on drug tablets,” Image Analysis and Characterization of SPIE, Vol. 4232, pp. 553-562.
[8]D. Grandt (2009), “Improving medication safety,” U.S. National Library of Medicine National Institutes of Health, Vol. 52, No. 12, pp. 1161-1165.
[9]C. H. Hsieh (2005), “Applying content based image retrieval to drug identification research,” Master Thesis, Graduate Institute of Medical Information, Taipei Medical University, Taipei, Taiwan, Republic of China.
[10]Y. R. Huang (2007), “The study of uterine cervical cellular images based on PC-based cytopathologic image analysis system and support vector machine,” Master Thesis, Department of Information Management, Chaoyang University of Technology, Taichung, Taiwan, Republic of China.
[11]H. Jegou, M. Douze, and C. Schmid (2010), “Improving bag-of-features for large scale image search,” International Journal of Computer Vision, Vol. 87, No. 3, pp. 316-336.
[12]X. Jiang, T. Sun, B. Chen, R. Li, and B. Feng (2009), “A novel video content understanding scheme based on feature combination strategy,” Journal of Computers, Vol. 4, No. 7, pp. 615-622.
[13]C. H. Lin, R. T. Chen, and Y. K. Chan (2009), “A Smart content-based image retrieval system based on color and texture feature,” Image and Vision Computing, Vol. 27, No. 6, pp. 658-665.
[14]C. H. Lin, E. W. Huang, and W. W. Jiang (2007), “Pill image retrieval using neural networks,” The Journal of Taiwan Association for Medical Informatics, Vol. 16, No. 7, pp. 29-42.
[15]G. H. Liu, L. Zhang, Y. K. Hou, Z. Y. Li, and J. Y. Yang (2010), “Image retrieval based on multi-texton histogram,” Pattern Recognition, Vol. 43, No. 7, pp. 2380-2389.
[16]Y. Liu, D. Zhang, G. Lu, and W. Y. Ma (2007), “A survey of content-based image retrieval with high-level semantics,” Pattern Recognition, Vol. 40. No. 1, pp. 262-282.
[17]Z. Liu and C. Liu (2010), “Fusion of color, local spatial and global frequency information for face recognition,” Pattern Recognition, Vol. 43, No. 8, pp. 2882-2890.
[18]R. Min and H. D. Cheng (2009), “Effective image retrieval using dominant color descriptor and fuzzy support vector machine,” Pattern Recognition, Vol. 42, No. 1, pp. 147-157.
[19]M. A. Moreno and F. Furtner (2010), “Medication safety for children,” Archives of Pediatrics and Adolescent Medicine, Vol. 16, No. 2, pp. 208.
[20]M. Pavlou and N. M. Allinson (2009), “Automated encoding of footwear patterns for fast indexing,” Image and Vision Computing, Vol. 27, No. 4, pp. 402-409.
[21]R. Phan, and D. Androutsos (2009), “Content-based retrieval of logo and trademarks in unconstrained color image databases using color edge gradient co-occurrence histograms,” Computer Vision and Image Understanding, Vol. 114, No, 1, pp. 66-84.
[22]R. Sandler and M. Lindenbaum (2009), “Optimizing gabor filter design for texture edge detection and classification,” International Journal of Computer Vision, Vol. 84, No. 3, pp. 308-324.
[23]L.L. Shen, and Z. Ji (2009), “Gabor wavelet selection and SVM classification for object recognition,” Acta Automatica Sinica, Vol. 35, No. 4, pp. 350-355.
[24]J. A. Santos, C. D. Ferreira, R. S. Torres, M. A. Goncalves, and R. A. C. Lamparelli (2010), “A relevance feedback method based on genetic programming for classification of remote sensing images,” Information Sciences, pp. 1-14.
[25]H. V. Sijs, T. V. Gelder, A. Vulto, M. Berg, and J. Aarts (2009), “Understanding handling of drug safety alerts: a simulation study,” International Journal of Medical Informatics, Vol. 79, No. 5, pp. 361-369.
[26]T. Y. L. Sung, F. H Hung, and H. W. Chiu (2008), “Implementation of an integrated drug information system for inpatients to reduce medication errors in administrating stage,” Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 743-746.
[27]E. Tapia, P. Bulacio, and L. Angelone (2010), “Recursive ECOC classification,” Pattern Recognition Letters, Vol. 31, No. 3, pp. 210-215.
[28]G. Woods (2002), Digital image processing, Gonzalez, Rafael Inc.
[29]Y. Z Xie (2008), “Study of drug identification system using the image processing technique,” Master Thesis, Department of Electronic Engineering, I-Shou University, Kaohsiung, Taiwan, Republic of China.
[30]Y. Zheng (2009), “Breast cancer detection with gabor features from digital mammograms,” Algorithms, Vol. 31, No. 1, pp. 44-62.
[31]Medication Information of Changhua Christen Hospital (2010), http://www.cch.org.tw/DRUG_01.aspx.
[32]Medication Information of Chang Gung Medical Foundtion (2010), http://www.cgmh.org.tw/stor/drug001.asp.
[33]Medication Information of Department of Health, Executive Yuan (2010), http://drug.doh.gov.tw/, Taiwan, Republic of China.
[34]Medication Information of Taichung Armed Force General Hospital (2010), http://web-reg-server.803.org.tw:8090/Med_Web/index.asp.
[35]MPEG-7 Overview (2010), http://mpeg.chiariglione.org/standards/mpeg-7/
mpeg-7.htm#E11E4.
[36]Wikipedia (2010), http://en.wikipedia.org/wiki/Portal:Contents.