以前大部分的肝臟疾病檢驗都是使用活體檢驗(侵入式採樣)，透過獲得小量肝臟組織，在顯微鏡下檢查，從而確定肝病的病因及肝臟纖維化的程度。近年來，由於醫療科學越來越進步，非侵入式的檢查方式逐漸盛行，像是使用超音波來做肝臟疾病的檢查。超音波檢查不僅減輕了病患的痛苦，也減少了侵入式檢查可能帶來的感染風險。但是對於經驗還不是很豐富的檢查人員來說，直接用肉眼來分析超音波圖以片判斷肝臟疾病是有難度的。
為了克服這個問題，我們提出電腦輔助分類法，使用影像處理和型樣辨識來分析超音波影像，分辨肝癌或是肝膿瘍。
在這本論文中，使用灰階共生矩陣和灰階長度矩陣來擷取特徵，分析超音波影像，利用序列前向選擇、續裂後像選擇或F分數選擇合適的特徵，接著消除多餘的特徵點找到最佳的特徵集合。最後再使用支持向量機或是類神經網路來分類不同的肝臟疾病，藉由找到合適的特徵我們可以增加肝臟疾病的診斷準確率，本研究可以提供無經驗診療師診斷之輔助。

Most of clinical technologies of detecting dangerous liver diseases are dependent on liver biopsy sampling. With the advances of medical technology in recent years, non-invasive detecting methods (e.g., based on ultrasound images) have been widely applied to diseases diagnosis. Therefore, the liver disease diagnosis becomes easier and more comfortable than before. This could reduce the risk of pain, infection or other injuries from biopsy tests. Nevertheless, for the less experienced clinicians, it could not be easy to clearly identify liver diseases from ultrasound images just by their eyes.
In order to overcome this problem, we applied the technologies of image processing and pattern recognition to the computer-aided classification system designed for ultrasound images between hepatocellular carcinoma (hcc-the most common type of liver cancer) and liver abscess.
In this study, the feature extraction methods (Gray-Level Co-Occurrence Matrix and Gray-Level Run-Length Matrix) were used to analyze the ultrasound images. Then the feature selections (Sequential Forward Selection, Sequential Backward Selection or F-score) eliminated the redundant features to obtain the optimal feature set before classifiers (Support Vector Machine or Neural Network) discriminated the different kind of diseases. This study can provide the diagnosis help for an inexperienced clinician.

[1] P. Pisani, D. M. Parkin, F. Bray, and J. Ferlay, “Estimates of the worldwide mortality from 25 cancers in 1990,” International journal of cancer, vol. 83, pp. 18-29, 1999.
[2] D. M. Parkin, “Global cancer statistics in the year 2000,” The lancet oncology, vol. 2, pp. 533-543, 2001.
[3] C. J. Chen and D. S. Chen, “Interaction of hepatitis B virus, chemical carcinogen, and genetic susceptibility: multistage hepatocarcinogenesis with multifactorial etiology,” Hepatology, vol. 36, pp. 1046-1049, 2002.
[4] J. J. Fu, Y.-W. Yu, H.-M. Lin, J.-W. Chai, and C. C.-C. Chen, “Feature extraction and pattern classification of colorectal polyps in colonoscopic imaging,” Computerized Medical Imaging and Graphics, vol. 38, pp. 267-275, 2014.
[5] S. Ghosh and V. Chaudhary, “Supervised methods for detection and segmentation of tissues in clinical lumbar MRI,” Computerized Medical Imaging and Graphics, vol. 38, pp. 639-649, 2014.
[6] T. Hopp, N. Duric, and N. Ruiter, “Image fusion of Ultrasound Computer Tomography volumes with X-ray mammograms using a biomechanical model based 2D/3D registration,” Computerized Medical Imaging and Graphics, 2014.
[7] J. Wu, A. Gogna, B. S. Tan, L. L. Ooi, Q. Tian, F. Liu, et al., “A manifold learning method to detect respiratory signal from liver ultrasound images,” Computerized Medical Imaging and Graphics, 2014.
[8] F. Martínez, J. Ruano, M. Gómez, and E. Romero, “Estimating the size of polyps during actual endoscopy procedures using a spatio-temporal characterization,” Computerized Medical Imaging and Graphics, 2015.
[9] K. J. Opieliński, P. Pruchnicki, T. Gudra, P. Podgórski, J. Kurcz, T. Kraśnicki, et al., “Imaging results of multi-modal ultrasound computerized tomography system designed for breast diagnosis,” Computerized Medical Imaging and Graphics, 2015.
[10] M. M. Fernández-Carrobles, G. Bueno, O. Déniz, J. Salido, M. García-Rojo, and L. Gonzández-López, “Frequential versus spatial colour textons for breast TMA classification,” Computerized Medical Imaging and Graphics, 2014.
[11] M. Rastgoo, R. Garcia, O. Morel, and F. Marzani, “Automatic differentiation of melanoma from dysplastic nevi,” Computerized Medical Imaging and Graphics, vol. 43, pp. 44-52, 2015.
[12] X.-Y. Zhang, X.-F. Diao, T.-F. Wang, and S.-P. Chen, “Study on feature extraction for ultrasonic differentiation of liver space-occupying lesions,” 2010 4th International Conference on Bioinformatics and Biomedical Engineering (iCBBE), pp. 1-4, 2010.
[13] D. Nicholas, D. Nassiri, P. Garbutt, and C. Hill, “Tissue characterization from ultrasound B-scan data,” Ultrasound in medicine & biology, vol. 12, pp. 135-143, 1986.
[14] W. D. Richard and C. G. Keen, “Automated texture-based segmentation of ultrasound images of the prostate,” Computerized Medical Imaging and Graphics, vol. 20, pp. 131-140, 1996.
[15] S. Pavlopoulos, G. Konnis, E. Kyriacou, D. Koutsouris, P. Zoumpoulis, and I. Theotokas, “Evaluation of texture analysis techniques for quantitative characterization of ultrasonic liver images,” Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE on Engineering in Medicine and Biology Society, pp. 1151-1152 vol.3, 1996.
[16] J. Bleck, U. Ranft, M. Gebel, H. Hecker, M. Westhoff-Bleck, C. Thiesemann, et al., “Random field models in the textural analysis of ultrasonic images of the liver,” IEEE Trans. Medical Imaging, vol. 15, pp. 796-801, 1996.
[17] Y. M. Kadah, A. A. Farag, J. M. Zurada, A. M. Badawi, and A. Youssef, “Classification algorithms for quantitative tissue characterization of diffuse liver disease from ultrasound images,” IEEE Trans. Medical Imaging, vol. 15, pp. 466-478, 1996.
[18] M. Klein Gebbinck, J. Verhoeven, J. Thijssen, and T. E. Schouten, “Application of neural networks for the classification of diffuse liver disease by quantitative echography,” Ultrasonic Imaging, vol. 15, pp. 205-217, 1993.
[19] S. Pavlopoulos, E. Kyriacou, D. Koutsouris, K. Blekas, A. Stafylopatis, and P. Zoumpoulis, “Fuzzy neural network-based texture analysis of ultrasonic images,” Engineering in Medicine and Biology Magazine, IEEE, vol. 19, pp. 39-47, 2000.
[20] M.-H. Horng, Y.-N. Sun, and X.-Z. Lin, “Texture feature coding method for classification of liver sonography,” Computerized medical imaging and graphics, vol. 26, pp. 33-42, 2002.
[21] P.-M. Yang, C.-M. Chen, T.-W. Lu, and C.-P. Yen, “Computer-aided diagnosis of sonographic liver cirrhosis: A spleen-reference approach,” Medical physics, vol. 35, pp. 1180-1190, 2008.
[22] C. Aubé, F. Oberti, N. Korali, M.-A. Namour, D. Loisel, J.-Y. Tanguy, et al., “Ultrasonographic diagnosis of hepatic fibrosis or cirrhosis,” Journal of hepatology, vol. 30, pp. 472-478, 1999.
[23] G. M. Dahab, M. M. Kheriza, H. M. EL‐BELTAGI, A. M. M. FOUDA, and O. A. S. EL‐DIN, “Digital quantification of fibrosis in liver biopsy sections: description of a new method by Photoshop software,” Journal of gastroenterology and hepatology, vol. 19, pp. 78-85, 2004.
[24] V. de Lédinghen, C. Douvin, A. Kettaneh, M. Ziol, D. Roulot, P. Marcellin, et al., “Diagnosis of hepatic fibrosis and cirrhosis by transient elastography in HIV/hepatitis C virus-coinfected patients,” JAIDS Journal of Acquired Immune Deficiency Syndromes, vol. 41, pp. 175-179, 2006.
[25] J. Shao, J. Wang, K. Liu, J. Bai, X. Hu, and L. Qian, “Maximal accumulative respiration strain for the assessment of hepatic fibrosis,” IUS 2008 on Ultrasonics Symposium, IEEE, pp. 2029-2032, 2008.
[26] W. B. Wooten, B. Green, and H. M. Goldstein, “Ultrasonography of Necrotic Hepatic Metastases 1,” Radiology, vol. 128, pp. 447-450, 1978.
[27] J. C. Sheu, J. L. Sung, D. S. Chen, J.-Y. Yu, T. Wang, C. Su, et al., “Ultrasonography of small hepatic tumors using high-resolution linear-array real-time instruments,” Radiology, vol. 150, pp. 797-802, 1984.
[28] M. J. Tong, L. M. Blatt, and V. W. Kao, “Surveillance for hepatocellular carcinoma in patients with chronic viral hepatitis in the United States of America,” Journal of gastroenterology and hepatology, vol. 16, pp. 553-559, 2001.
[29] A. Laghi, R. Iannaccone, P. Rossi, I. Carbone, R. Ferrari, F. Mangiapane, et al., “Hepatocellular Carcinoma: Detection with Triple-Phase Multi–Detector Row Helical CT in Patients with Chronic Hepatitis 1,” Radiology, vol. 226, pp. 543-549, 2003.
[30] B. I. Choi, “The current status of imaging diagnosis of hepatocellular carcinoma,” Liver transplantation, vol. 10, pp. S20-S25, 2004.
[31] N. Newlin, T. M. Silver, K. J. Stuck, and M. Sandler, “Ultrasonic features of pyogenic liver abscesses,” Radiology, vol. 139, pp. 155-159, 1981.
[32] P. Kitiwanwanich, J. Chaiyakum, V. Laopaiboom, J. Kanpittaya, N. Chamadol, E. o. Mairieng, et al., “Ultrasonographic features of liver abscesses,” Srinagarind Medical Journal (SMJ), vol. 8, pp. 146-152, 2011.
[33] J. A. A. Pérez, J. J. González, R. F. Baldonedo, L. Sanz, G. Carreño, A. Junco, et al., “Clinical course, treatment, and multivariate analysis of risk factors for pyogenic liver abscess,” The American journal of surgery, vol. 181, pp. 177-186, 2001.
[34] C. Cruz-Gomez, P. Wiederhold, and M. Gudino-Zayas, “Automatic liver tissue segmentation in microscopic images using fusion color space and multiscale morphological reconstruction,” Electronics and Computer Engineering (TAEECE), 2013 International Conference on Technological Advances in Electrical, pp. 88-92, 2013.
[35] P. S. Maclin and J. Dempsey, “A neural network to diagnose liver cancer,” IEEE International Conference on Neural Networks, pp. 1492-1497 vol.3, 1993.
[36] L. Lijuan and Y. Xiaomei, “Neural Network Optimized by Improved Genetic Algorithms to Diagnosis,” 2012 International Conference on Industrial Control and Electronics Engineering (ICICEE), pp. 1185-1187, 2012.
[37] S. Huang, M. Ding, and S. Zhang, “Feature statistic analysis of ultrasound images of liver cancer,” in International Symposium on Multispectral Image Processing and Pattern Recognition, pp. 67890N-67890N-7, 2007.
[38] D. Gaitini, M. Lederman, Y. Baruch, E. Ghersin, E. Veitsman, H. Kerner, et al., “Computerised analysis of liver texture with correlation to needle biopsy,” Ultraschall in der Medizin (Stuttgart, Germany: 1980), vol. 26, pp. 197-202, 2005.
[39] W.-C. Yeh, S.-W. Huang, and P.-C. Li, “Liver fibrosis grade classification with B-mode ultrasound,” Ultrasound in medicine & biology, vol. 29, pp. 1229-1235, 2003.
[40] D. S. Elizabeth, A. Kannan, and H. K. Nehemiah, “Computer‐aided diagnosis system for the detection of bronchiectasis in chest computed tomography images,” International Journal of Imaging Systems and Technology, vol. 19, pp. 290-298, 2009.
[41] Z. Jiang, K. Yamauchi, K. Yoshioka, K. Aoki, S. Kuroyanagi, A. Iwata, et al., “Support vector machine-based feature selection for classification of liver fibrosis grade in chronic hepatitis C,” Journal of medical systems, vol. 30, pp. 389-394, 2006.
[42] W.-L. Lee, Y.-C. Chen, and K.-S. Hsieh, “Ultrasonic liver tissues classification by fractal feature vector based on M-band wavelet transform,” IEEE Trans. Medical Imaging, vol. 22, pp. 382-392, 2003.
[43] D. Sabih and M. Hussain, “Automated classification of liver disorders using ultrasound images,” Journal of medical systems, vol. 36, pp. 3163-3172, 2012.
[44] M. E. Karar, D. R. Merk, V. Falk, and O. Burgert, “A simple and accurate method for computer-aided transapical aortic valve replacement,” Computerized Medical Imaging and Graphics.
[45] C. R. Hatt, G. Ng, and V. Parthasarathy, “Enhanced needle localization in ultrasound using beam steering and learning-based segmentation,” Computerized Medical Imaging and Graphics, vol. 41, pp. 46-54, 4// 2015.
[46] R. T. Ribeiro, R. Tato Marinho, and J. M. Sanches, “An Ultrasound-Based Computer-Aided Diagnosis Tool for Steatosis Detection,” IEEE Journal of Biomedical and Health Informatics, vol. 18, pp. 1397-1403, 2014.
[47] S. Moldovanu, L. Moraru, and D. Bibicu, “Computerized decision support in liver steatosis investigation,” Int. J. Biol. Biomed. Eng, vol. 6, pp. 69-76, 2012.
[48] S. Poonguzhali and G. Ravindran, “Automatic classification of focal lesions in ultrasound liver images using combined texture features,” Information Technology Journal, vol. 7, pp. 205-209, 2008.
[49] S. Onal, S. K. Lai-Yuen, P. Bao, A. Weitzenfeld, and S. Hart, “MRI-Based Segmentation of Pubic Bone for Evaluation of Pelvic Organ Prolapse,” IEEE Journal of Biomedical and Health Informatics, vol. 18, pp. 1370-1378, 2014.
[50] W. Gomez, W. C. A. Pereira, and A. F. C. Infantosi, “Analysis of Co-Occurrence Texture Statistics as a Function of Gray-Level Quantization for Classifying Breast Ultrasound,” IEEE Trans. Medical Imaging, vol. 31, pp. 1889-1899, 2012.
[51] G. Yi, W. Yuanyuan, K. Dehong, and S. Xianhong, “Automatic Classification of Intracardiac Tumor and Thrombi in Echocardiography Based on Sparse Representation,” IEEE Journal of Biomedical and Health Informatics, vol. 19, pp. 601-611, 2015.
[52] R. M. Haralick, K. Shanmugam, and I. H. Dinstein, “Textural features for image classification,” IEEE Trans. Systems, Man and Cybernetics, pp. 610-621, 1973.
[53] A. Chaudhari and J. Kulkarni, “Local entropy based brain MR image segmentation,” 2013 IEEE 3rd International on Advance Computing Conference (IACC), pp. 1229-1233, 2013.
[54] L. Huang, J. Shi, R. Wang, and S. Zhou, “Shearlet-Based Ultrasound Texture Features for Classification of Breast Tumor,” 2013 Seventh International Conference on Internet Computing for Engineering and Science (ICICSE), pp. 116-121, 2013.
[55] S. Kumar, R. Moni, and J. Rajeesh, “Liver tumor diagnosis by gray level and contourlet coefficients texture analysis,” 2012 International Conference on in Computing, Electronics and Electrical Technologies (ICCEET), pp. 557-562, 2012.
[56] M. Lupsor, R. Badea, S. Nedevschi, D. Mitrea, and M. Florea, “Ultrasonography contribution to hepatic steatosis quantification. possibilities of improving this method through computerized analysis of ultrasonic image,” 2006 IEEE International Conference on Automation, Quality and Testing, Robotics, pp. 478-483, 2006.
[57] D. Mitrea, S. Nedevschi, and R. Badea, “The role of the superior order GLCM in improving the automatic diagnosis of the hepatocellular carcinoma based on ultrasound images,” 2011 34th International Conference on Telecommunications and Signal Processing (TSP), pp. 602-606, 2011.
[58] D. Mitrea, S. Nedevschi, and R. Badea, “The role of the superior order GLCM and of the generalized cooccurrence matrices in the characterization and automatic diagnosis of the hepatocellular carcinoma, based on ultrasound images,” 2011 IEEE International Conference on Intelligent Computer Communication and Processing (ICCP), pp. 197-204, 2011.
[59] D. Mitrea, S. Nedevschi, and R. Badea, “The role of the multiresolution textural features in improving the characterization and recognition of the liver tumors, based on ultrasound images,” 2012 14th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC), pp. 192-199, 2012.
[60] D. Mitrea, S. Nedevschi, M. Lupsor, and R. Badea, “Exploring the textural parameters obtained from ultrasound images for modeling the liver pathological stages in the evolution towards hepatocellular carcinoma,” IEEE International Conference on Automation, Quality and Testing, Robotics, pp. 128-133, 2008.
[61] D. Mitrea, S. Nedevschi, M. Lupsor, M. Socaciu, and R. Badea, “Improving the textural model of the hepatocellular carcinoma using dimensionality reduction methods,” 2nd International Congress on Image and Signal Processing, pp. 1-5, 2009.
[62] D. Mitrea, S. Nedevschi, M. Lupsor, M. Socaciu, and R. Badea, “Advanced classification methods for improving the automatic diagnosis of the hepatocellular carcinoma, based on ultrasound images,” in Proceedings of the 2010 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR)-Volume 02, 2010, pp. 1-6.
[63] D. Mitrea, M. Socaciu, R. Badea, and A. Golea, “Texture based characterization and automatic diagnosis of the abdominal tumors from ultrasound images using third order GLCM features,” 2011 4th International Congress on Image and Signal Processing (CISP), pp. 1558-1562, 2011.
[64] A. A. Moghe, J. Singhai, and S. Shrivastava, “Elastic registration of 2D abdominal CT images using hybrid feature point selection for liver lesions,” 2010 IEEE 2nd International on Advance Computing Conference (IACC), pp. 337-341, 2010.
[65] M. Mohamed Fathima, D. Manimegalai, and S. Thaiyalnayaki, “Automatic detection of tumor subtype in mammograms based On GLCM and DWT features using SVM,” 2013 International Conference on Information Communication and Embedded Systems (ICICES), pp. 809-813, 2013.
[66] N. M. Saad, S. Abu-Bakar, S. Muda, M. Mokji, and L. Salahuddin, “Brain lesion segmentation of Diffusion-weighted MRI using gray level co-occurrence matrix,” 2011 IEEE International Conference on Imaging Systems and Techniques (IST), pp. 284-289, 2011.
[67] D. Mitrea, S. Nedevschi, M. Lupsor, M. Socaciu, and R. Badea, “Experimenting various classification techniques for improving the automatic diagnosis of the malignant liver tumors, based on ultrasound images,” 2010 3rd International Congress on Image and Signal Processing (CISP), pp. 1853-1858, 2010.
[68] M. Galloway, “Texture analysis using grey level run lengths,” NASA STI/Recon Technical Report N, vol. 75, p. 18555, 1974.
[69] A. Chu, C. M. Sehgal, and J. F. Greenleaf, “Use of gray value distribution of run lengths for texture analysis,” Pattern Recognition Letters, vol. 11, pp. 415-419, 1990.
[70] B. V. Dasarathy and E. B. Holder, “Image characterizations based on joint gray level—run length distributions,” Pattern Recognition Letters, vol. 12, pp. 497-502, 1991.
[71] X. Tang, “Texture information in run-length matrices,” IEEE Trans. Image Processing, vol. 7, pp. 1602-1609, 1998.
[72] P. S. Bradley, O. L. Mangasarian, and W. N. Street, “Feature selection via mathematical programming,” INFORMS Journal on Computing, vol. 10, pp. 209-217, 1998.
[73] K. Kira and L. A. Rendell, “The feature selection problem: Traditional methods and a new algorithm,” in AAAI, pp. 129-134, 1992.
[74] A. Jain and D. Zongker, “Feature selection: Evaluation, application, and small sample performance,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, pp. 153-158, 1997.
[75] M. Kudo and J. Sklansky, “Comparison of algorithms that select features for pattern classifiers,” Pattern recognition, vol. 33, pp. 25-41, 2000.
[76] R. Kohavi and G. H. John, “Wrappers for feature subset selection,” Artificial intelligence, vol. 97, pp. 273-324, 1997.
[77] R. Kohavi and D. Sommerfield, “Feature Subset Selection Using the Wrapper Method: Overfitting and Dynamic Search Space Topology,” in KDD, 1995, pp. 192-197.
[78] A. Miller, Subset selection in regression: CRC Press, 2002.
[79] J. Neter, M. H. Kutner, C. J. Nachtsheim, and W. Wasserman, Applied linear statistical models, vol. 4: Irwin Chicago, 1996.
[80] R.-E. Fan, P.-H. Chen, and C.-J. Lin, “Working set selection using second order information for training support vector machines,” The Journal of Machine Learning Research, vol. 6, pp. 1889-1918, 2005.