近年來為了達到口腔癌的提早診斷及治療，已有許多口腔癌篩檢的方法被提出。在這些方法當中，以光學特性檢測的方式具有較快速且非侵入式的篩檢流程，最適合用於口腔癌的初步診斷。這類的方法主要利用癌症腫瘤造成組織在自體螢光及型態上變異，輔助觀察異常之患部，以協助醫護人員進行診斷。而在先前的研究當中，我們已開發出用於拍攝口腔自體螢光影像之儀器系統，並應用於臨床收案及測試。同時，該系統也包含了口腔癌的自動偵測演算法，透過計算螢光影像中病灶區域的影像強度平均及標準差，作為特徵值並建立分類器模型，用於正常與異常影像資料的分辨。
為了提升資料分類的正確率，本研究透過基於紋理分析的特徵演算法，對螢光影像擷取具備更佳的分群效果之特徵。我們分別使用了小波分析及希爾伯特-黃轉換(Hilbert-Huang transform, HHT)兩種空間域對頻域轉換的演算法，將螢光影像分解成不同頻率的組成訊號。基於這些高頻訊號成分常帶有紋理資訊的原理，透過計算其頻段的統計值及共生矩陣特性(co-occurrence matrix property)來描述正常與異常組織影像的紋理性質，如粗糙及平滑程度等。不同類型的紋理分析方法，包含碎形紋理分析及灰階共生矩陣分析也將會被實作並用於分析相同的口腔螢光資料，已進行比較，並Fisher’s Discriminant ratio作為評估的標準，以探討各個特徵對正常與異常資料的分類效果。由實驗結果發現，相較於最早使用最簡單的統計值，以紋理分析的方式最能區別出口腔癌所造成組織的病變，其中以頻譜分析的方式所得到的影像特徵具有最好的效果，而有助於後續建立更可靠的正常/異常資料分類器。

In recent years, oral cancer screening techniques based on optical characteristics examination have been developed, which allows a much more feasible solution for preliminary diagnosis. In earlier research, an oral cavity imaging system has been designed and applied to clinical test and data collection, which provides two different excitation light sources for oral fluorescence imaging, meanwhile, the captured fluorescence images are analyzed with a recognition algorithm to detect oral cancer. Based on the fact that neoplasia causes fluorescent and morphological changes in lesion, the algorithm computes intensity and standard deviations of ROI (Region of interest) as features for classification.

In order to achieve higher classification accuracy, we proposed feature extraction methods based on texture analysis to extract more effective and reliable image features. In this study, spatial frequency transformation methods are implemented, including wavelet-based texture analysis and 2D-HHT (Hilbert Huang Transform) texture feature extraction. Both approaches investigate texture information by decomposing original image into different frequency channels, from which statistical and co-occurrence features are extracted. For comparison, different types of texture analysis, including fractal analysis and co-occurrence matrix methods, are also implemented on the same dataset, with Fisher’s Discriminant Ratio as a criterion for evaluate discriminability of features. Compared to simplest statistical property used in former work, texture analysis successfully brings improvement in characterizing oral cancer lesions; where spatial-frequency transform methods show the best discriminability, which can be further applied to construction more reliable classifier.

1. Health Promotion Administration, Ministry of Health and Welfare, Taiwan. Cancer Prevention and Control. Available at:
http://www.hpa.gov.tw/BHPNet/English/ClassShow.aspx?No=201401280006
2. Huang, C.-C., et al., Life expectancy and expected years of life lost to oral cancer in Taiwan: A nation-wide analysis of 22,024 cases followed for 10 years. Oral Oncology 51 (2015) 349–354.
3. Chiang, C.-J., et al., Incidence and survival of adult cancer patients in Taiwan, 2002-2012. Journal of the Formosan Medical Association (2015), 1–13.
4. M. W. Lingen, et al. Critical Evaluation of Diagnostic Aids for the Detection of Oral Cancer. Oral Oncol., p. 10–22, January 2008.
5. T.W.J. Poate, et al., An audit of the efficacy of the oral brush biopsy technique in a specialist Oral Medicine unit. Oral Oncology, (2004) 40 829–834.
6. N. J. Park, et al., Salivary microRNA: Discovery, Characterization, and Clinical Utility for Oral Cancer Detection., Clin Cancer Res September 1, 2009 15; 5473.
7. M. Missmann, et al., A reason for the use of toluidine blue staining in the presurgical management of patients with oral squamous cell carcinomas., Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and Endodontology, Volume 102, Issue 6, December 2006, P741–743.
8. E. Svistun, et al., Vision enhancement system for detection of oral cavity neoplasia based on autofluorescence. Head Neck. 2004 Mar;26(3):205-15.
9. I. Pavlova, et al., Understanding the Biological Basis of Autofluorescence Imaging for Oral Cancer Detection: High-Resolution Fluorescence Microscopy in Viable Tissue. Clin Cancer Res. 2008 Apr 15;14(8):2396-404.
10. A. Rashid, S. Warnakulasuriya, The use of light-based (optical) detection systems as adjuncts in the detection of oral cancer and oral potentially malignant disorders: a systematic review. Journal of Oral Pathology & Medicine (2015) 44: 307-328.
11. C. S. Farah, et al., Efficacy of tissue autofluorescence imaging (velscope) in the visualization of oral mucosal lesions. Head & Neck Volume 34, Issue 6, pages 856–862, June 2012
12. H.-R. Hung, et al., Application of Autofluorescence Imaging on Oral Cancer Screening.
13. L. Z. Li, et al., Quantitative Magnetic Resonance and Optical Imaging Biomarkers of Melanoma Metastatic Potential. Proceedings of the National Academy of Sciences (Impact Factor: 9.67). 05/2009; 106(16):6608-13.
14. T.-Y. Huang et al., Application of Autofluorescence Imaging On Clinical Oral Cancer Screening
15. G. Castellano, et al., Texture analysis of medical images. Clinical Radiology (2004) 59, 1061–1069.
16. N. E. Huang, Z. Wu, Review on Hilbert-Huang transform: Method and its applications to geophysical studies. Reviews of Geophysics, Volume 46, Issue 2 June 2008.
17. J. Beck, et al., Spatial Frequency Channels and Perceptual Grouping in Texture Segregation. Computer Vision, Graphics, and Image Processing Volume 37, Issue 2, February 1987, Pages 299-325.
18. S. Mika, Fisher discriminant analysis with kernels. Neural Networks for Signal Processing IX, 1999.
19. X. Tang, Texture information in run-length matrices. IEEE Transactions on Image Processing Vol. 7, No. 11.
20. D. A. Clausi, An analysis of co-occurrence texture statistics as a function of grey level quantization. Can. J. Remote Sensing, Vol. 28, No. 1, pp. 45–62, 2002
21. M. B. Ericson, et al., Bispectral fluorescence imaging combined with texture analysis and linear discrimination for correlation with histopathologic extent of basal cell carcinoma. Journal of Biomedical Optics 10(3), 034009 (May/June 2005)
22. H. D. Li, et al., Markov Random Field for Tumor Detection in Digital Mammography. IEEE Transactions on Medical Imaging, Vol. 14, No. 3, September 1995.
23. H. S. Choi, et al., Partial Volume Tissue Classification of Multichannel Magnetic Resonance Images-A Mixel Model. IEEE Transactions on Medical Imaging, Vol.10. No. 3. September 1991.
24. C.-H. Lee, et al., Segmenting Brain Tumors with Conditional Random Fields and Support Vector Machines. Computer Vision for Biomedical Image Applications Vol. 3765 of the series Lecture Notes in Computer Science pp 469-478.
25. E-L. Chen, et al., An Automatic Diagnostic System for CT Liver Image Classification. IEEE Transaction on Biomedical Engineering, Vol. 45, No. 6, June 1998 783.
26. S. E. Grigorescu, et al., Comparison of Texture Features Based on Gabor Filters. IEEE Transactions on Image Processing, Vol. 11, No. 10, October 2002.
27. S. Arivazhagan, et al., Texture classification using wavelet transform. Pattern Recognition Letters 24 (2003) 1513–1521.
28. K. Huang, et al., Wavelet Feature Selection for Image Classification. IEEE Transactions on Image Processing, Vol. 17, No. 9, September 2008.
29. C.-R. Huang, et al., Helicobacter Pylori-Related Gastric Histology Classification Using Support-Vector-Machine-Based Feature Selection. IEEE Transactions on Information Technology in Biomedicine, Vol. 12, No. 4, July 2008.
30. B. Remeseiro, et al., Automatic classification of the interferential tear film lipid layer using colour texture analysis. Computer Methods and Programs in Biomedicine III (2013) P.93-103
31. A. Mojsilovic, et al., On the Selection of an Optimal Wavelet Basis for Texture Characterization. IEEE Transactions on Image Processing, Vol. 9, No. 12, December 2000
32. N. E. Huang, et al, Applications of Hilbert–Huang transform to non-stationary financial time series analysis. Stochastic Models in Business and Industry, Vol. 19, Issue 3 September 2003 P.245–268.
33. B. Demir, et al., Empirical Mode Decomposition of Hyperspectral Images for Support Vector Machine Classification. IEEE Transactions on Geoscience and Remote Sensing Vol. 48, No. 11, November 2010
34. J.-J. Pan, et al., Texture Classification Based on BIMF Monogenic Signals. Computer Vision–ACCV 2012 Vol. 7725 of the series Lecture Notes in Computer Science pp. 177-187
35. J.-P. Hu, et al., Novel and efficient computation of Hilbert–Huang transform on surfaces. Computer Aided Geometric Design 43(2016)95–108
36. K. Langley, et al., The Riesz transform and simultaneous representations of phase, energy and orientation in spatial vision. Vision Research 50 (2010) 1748–1765