Normal and Abnormal Tissue Classification in Positron Emission Tomography Oncological Studies
Positron Emission Tomography (PET) imaging is increasingly used in radiotherapy environment as well as for staging and assessing treatment response. The ability to classify PET tissues, as normal versus abnormal tissues, is crucial for medical analysis and interpretation. For this reason, a system for classifying PET area is implemented and validated. The proposed classification is carried out using k-nearest neighbor (KNN) method with the stratified K-Fold Cross-Validation strategy to enhance the classifier reliability. A dataset of eighty oncological patients are collected for system training and validation. For every patient, lesion (abnormal tissue) and background (normal tissue around the lesion) are contoured on PET images using a semi-automatic method. Then 160 vectors are obtained to train and validate the KNN. Each vector is composed by thirty Standardized Uptake Values (SUVs) characterizing the area under investigation (lesion or background). In one case, vectors are labeled as normal or abnormal tissues by a nuclear medicine physician using a semi-automatic method; in other cases, Fuzzy C-means (FCM) and k-means are used for labelling vectors in an unsupervised manner. This study aims to evaluate the performance of the proposed classifier comparing it to the Linear Kernel Support Vector Machine (KSVM). The method accuracy is evaluated by comparison with the gold standard in terms of correct classification. Experimental results show that the KNN method achieves the highest classification accuracy using the semi-automatic labelling training (Sensitivity: 86.25%; Specificity: 90.00%; Negative Predictive Value: 88.37%; Precision: 89.81%; Accuracy: 88.12%; Error: 11.87%). In addition, the proposed method shows real-time performance; it could be applied to the field classification of PET images assisting physicians into discrimination of normal and abnormal tissue during radiation treatment planning.
Keywordstissue classification K-nearest neighbor Kernel support vector machine Fuzzy C-Means K-Fold cross-validation positron emission tomography
Unable to display preview. Download preview PDF.
- 2.L. K. Shankar, J. M. Hoffman, S. Bacharach, M. M. Graham, J. Karp, A. A. Lammertsma, S. Larson, D. A. Mankoff, B. A. Siegel, A. Van den Abbeele, et al., “Consensus recommendations for the use of 18ffdg pet as an indicator of therapeutic response in patients in national cancer institute trials,” J. Nucl. Med. 47, 1059–1066 (2006).Google Scholar
- 3.L. Rundo, A. Stefano, C. Militello, G. Russo, M. G. Sabini, C. D’Arrigo, F. Marletta, M. Ippolito, G. Mauri, S. Vitabile, et al., “A fully automatic approach for multimodal pet and mr image segmentation in gamma knife treatment planning,” Comput. Methods Programs Biomed. 144, 77–96 (2017).CrossRefGoogle Scholar
- 4.S. M. Larson, Y. Erdi, T. Akhurst, M. Mazumdar, H. A. Macapinlac, R. D. Finn, C. Casilla, M. Fazzari, N. Srivastava, H. W. D. Yeung, et al., “Tumor treatment response based on visual and quantitative changes in global tumor glycolysis using pet-fdg imaging: the visual response score and the change in total lesion glycolysis,” Clin. Positron Imag. 2, 159–171 (1999).CrossRefGoogle Scholar
- 15.Lei Bi, Jinman Kim, Lingfeng Wen, Dagan Feng, and M. Fulham, “Automated thresholded region classification using a robust feature selection method for pet-ct,” in Proc. 12th IEEE Int. Symp. on Biomedical Imaging (ISBI) (IEEE, 2015), pp. 1435–1438.Google Scholar
- 17.R. Boellaard, R. Delgado-Bolton, W. J. G. Oyen, F. Giammarile, K. Tatsch, W. Eschner, F. J. Verzijlbergen, S. F. Barrington, L. C. Pike, W. A. Weber, et al., “Fdg pet/ct: Eanm procedure guidelines for tumour imaging: version 2.0,” Europ. J. Nucl. Med. Mol. Imag. 42, 328–354 (2015).CrossRefGoogle Scholar