The development and evaluation of a computerized diagnosis scheme for pneumoconiosis on digital chest radiographs

BioMedical Engineering OnLine

Biyun Zhu¹  Wei Luo¹  Baoping Li⁴  Budong Chen³  Qiuying Yang²  Yan Xu³  Xiaohua Wu³  Hui Chen²  Kuan Zhang² 

[1] School of Biomedical Engineering, Capital Medical University, Beijing 100069, China

[2] Beijing Key Laboratory of Fundamental Research on Biomechanics in Clinical Application, Capital Medical University, Beijing 100069, China

[3] Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing 100053, China

[4] Department of Radiology, Coal General Hospital, Beijing 100028, China

关键词: Support vector machine;    Texture feature;    Bootstrap resampling;    Pneumoconiosis;    Digital radiograph;   

Others  :  1084293
DOI  :  10.1186/1475-925X-13-141

 received in 2014-08-08, accepted in 2014-09-24,  发布年份 2014

【 摘 要 】

Purpose

To diagnose pneumoconiosis using a computer-aided diagnosis system based on digital chest radiographs.

Methods

Lung fields were first extracted by combining the traditional Otsu-threshold method with a morphological reconstruction on digital radiographs (DRs), and then subdivided into six non-overlapping regions (region (a-f)). Twenty-two wavelet-based energy texture features were calculated exclusively from each region and selected using a decision tree algorithm. A support vector machine (SVM) with a linear kernel was trained using samples with texture features to classify an individual region of a healthy subject or a pneumoconiosis patient. The final classification results were obtained by integrating these individual classifiers with the weighted voting method. All models were developed on a dataset of 85 healthy controls and 40 stage I or II pneumoconiosis patients and validated by using the bootstrap resampling with replacement method.

Results

The areas under receiver operating characteristic curves (AUCs) of regions (c) and (f) were 0.688 and 0.563, which were worse than those of the other four regions. Region (c) and (f) were both excluded from the individual classifiers that were going to be assembled further. When built on the selected texture features, each individual SVM showed a higher diagnostic performance for the training set and the test set. The classification performance after an ensemble was 0.997 and 0.961 of the AUC value for the training and test sets, respectively. The final results were 0.974 ± 0.018 for AUC value and 0.929 ± 0.018 for accuracy.

Conclusion

The integrated SVM model built on the selected feature set showed the highest diagnostic performance among all individual SVM models. The model has good potential in diagnosing pneumoconiosis based on digital chest radiographs.

【 授权许可】

   
2014 Zhu et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

【 参考文献 】

• [1]Chong S, Lee KS, Chun MJ, Han J, Kwon OJ, Kim TS: Pneumoconiosis: comparison of imaging and pathologic findings. Radiographics 2006, 26:59-77.
• [2]International Labor Organization (ILO): Guidelines for the Use of the ILO International Classification of Radiographs of Pneumoconiosis. Occupational Safety and Health Series, No. 22 (Rev.). Geneva Switzerland: International Labor Office; 1980.
• [3]Savol AM, Li CC, Hoy RJ: Computer-aided recognition of small rounded pneumoconiosis opacities in chest X-rays. IEEE Trans Pattern Anal Mach Intell 1980, 2:479-482.
• [4]Yu P, Xu H, Zhu Y, Yang C, Sun X, Zhao J: An automatic computer-aided detection scheme for pneumoconiosis on digital chest radiographs. J Digit Imaging 2011, 24:382-393.
• [5]Yu P, Zhao J, Xu H, Yang C, Sun X, Chen S: Computer Aided Detection for Pneumoconiosis based on Histogram Analysis. In Proceedings of the 1st International Conference on Information Science and Engineering (ICISE). Nanjing, China: IEEE; 2009.
• [6]Yu P, Zhao J, Xu H, Sun X, Mao L: Computer Aided Detection for Pneumoconiosis based on Co-Occurrence Matrices Analysis. In Proceedings of the Second International Conference on Biomedical Engineering and Informatics. Tianjin, China: IEEE; 2011.
• [7]Katsuragawa S, Doi K, MacMahon H, Monnier-Cholley L, Morishita J, Ishida T: Quantitative analysis of geometric-pattern features of interstitial infiltrates in digital chest radiographs: preliminary results. J Digit Imaging 1996, 9:137-144.
• [8]Mir AH, Hanmandlu M, Tandon SN: Texture analysis of CT images. IEEE Eng Med Biol 1995, 14:781-786.
• [9]Zhu Y, Tan YQ, Hua YQ, Wang MP, Zhang G, Zhang JG: Feature selection and performance evaluation of support vector machine (SVM)-based classifier for differentiation benign and malignant pulmonary nodules by computed tomography. J Digit Imaging 2010, 23:51-65.
• [10]Yuan QF: Cancer Diagnosis by Using Support Vector Machine. MD Thesis. Chongqing University; 2007.
• [11]Lu Y: Automatic topic identification of health-related messages in online health community using text classification. Springer plus 2013, 2:309. BioMed Central Full Text
• [12]Zhu BY, Chen H: Morphological reconstruction based segmentation of lung fields on digital radiographs. Adv Mater Res 2013, 605:2155-2159.
• [13]Hering KG, Jacobsen M, Bosch-Galetke E, Elliehausen HJ, Hieckel HG, Hofmann-Preiss K, Jacques W, Jeremie U, Kotschy-Lang N, Kraus T, Menze B, Raab W, Raithel HJ, Schneider WD, Strassburger K, Tuengerthal S, Woitowitz HJ: Further development of the International Pneumoconiosis Classification--from ILO 1980 to ILO 2000 and to ILO 2000/German Federal Republic version. Pneumologie (Stuttgart, Germany) 2003, 57:576-584.
• [14]Arivazhagan S, Ganesan L: Texture segmentation using wavelet transform. Pattern Recogn Lett 2003, 24:3197-3203.
• [15]Kociołek M, Materka A, Strzelecki M, Szczypiński P: Discrete Wavelet Transform –Derived Features for Digital Image Texture Analysis. In Proceedings of International Conference on Signals and Electronic Systems. Lodz, Poland: IEEE; 2001.
• [16]Wu PC, Chen LG: An efficient architecture for two-dimensional discrete wavelet transform. IEEE Trans Circuits and Syst Video Tech 2001, 11:536-545.
• [17]Fukuda S, Hirosawa N: Land Cover Classification from Multi-Frequency Polarmetric Synthetic Aperture Radar Data using Wavelet-based Texture Information. In Proceedings of IEEE International Geoscience and Remote Sensing Symposium: 6-10 July 1998. Seattle, WA: IEEE; 1998:357-359.
• [18]Zhu BY, Chen H, Chen BD, Xu Y, Zhang K: Support vector machine model for diagnosis pneumoconiosis based on wavelet texture features of digital chest radiographs. J Digit Imaging 2014, 27:90-97.
• [19]Kira K, Rendell LA: The Feature Selection Problem: Traditional Methods and a New Algorithm. In Proceedings of the tenth National Conference on Artificial Intelligence: 12-16 July 1992. San Jose, CA: AAAI Press; 1992:129-134.
• [20]Bian ZQ, Zhang XG: Pattern Recognition 2nd ed. Beijing: Tsinghua University Publisher; 2000.
• [21]Quinlan JR: Induction Decision Tree. Mach Learn 1986, 1:81-106.
• [22]Li C, Zhi X, Ma J, Cui Z, Zhu Z, Zhang C: Performance comparison between logistic regression, decision trees, and multilayer perception in predicting peripheral neuropathy in type 2 diabetes mellitus. Chin Med J (Engl) 2012, 125:851-857.
• [23]Foster KR, Koprowski R, Skufca JD: Machine learning, medical diagnosis, and biomedical engineering research commentary. BioMed Eng OnLine 2014, 13:94-103. BioMed Central Full Text
• [24]Furkan K, Alexander S, Kivance K, Tulin E, Enis CA, Rengul C: Image classification of human carcinoma cells using complex wavelet-based covariance descriptors. PLoS One 2013, 8:e52807.
• [25]Fawcett T: An introduction to ROC analysis. Pattern Recogn Lett 2006, 27:861-874.
• [26]Efron B, Tibshirani R: An Introduction to the Bootstrap. New York: Chapman & Hall; 1993.
• [27]Sahiner B, Chan HP, Hadjiiski L: Classifier performance prediction for computer-aided diagnosis using a limited dataset. Med Phys 2008, 35:1559-1570.
• [28]Xu H, Tao X, Sundararajan R: Computer Aided Detection for Pneumoconiosis Screening on Digital Chest Radiographs. In Proceedings of the Third International Workshop on Pulmonary Image Analysis, September 20, 2010. Beijing, China: CreateSpace Independent Publishing Platform; 2010:129-138.
• [29]Okumura E, Kawashita I, Ishida T: Computerized analysis of pneumoconiosis in digital chest radiography: effect of artificial neural network trained with power spectra. J Digit Imaging 2011, 24:1126-1132.
• [30]Cai CX, Zhu BY, Chen H: Computer-aided diagnosis for pneumoconiosis based on texture analysis on digital chest radiographs. Appl Mech Mater 2013, 241-244:244-247.
• [31]Kondo K, Zhao B, Mino M: Automated Quantitative Analysis for Pneumoconiosis. In Proceedings of International Symposium on Multispectral Image Processing: 21-23 October 1998. Wuhan, China: SPIE; 1998.
• [32]Cohen R, Velho V: Update on respiratory disease from coal mine and silica dust. Clin Chest Med 2002, 23:811-826.
• [33]Cohen RA, Patel A, Green FH: Lung disease caused by exposure to coal mine and silica dust. Semin Respir Crit Care Med 2008, 29:651-661.
• [34]Castranova V, Vallyathan V: Silicosis and coal workers’ pneumoconiosis. Environ Health Perspect 2000, 108(Suppl 4):675-684.