Feasibility Study of MRI and Multimodality CT/MRI Radiomics for Lung Nodule Classification

• Anthony E. A. Jatobá UFAL
• Marcelo C. Oliveira UFAL
• Marcel Koenigkam-Santos USP
• Paulo de Azevedo-Marques USP

DOI: https://doi.org/10.5753/sbcas.2021.16065

Abstract

Lung cancer is the most common and lethal form of cancer, and its early diagnosis is key to the patient's survival. CT is the reference imaging scan for lung cancer screening; however, it presents the drawback of exposing the patient to ionizing radiation. Recent studies have shown the relevance of MRI in lung nodules diagnosis. In this work, we aimed to evaluate whether radiomics features from MRI are well-suited for lung nodules characterization and if the combination of CT and MRI features can yield better results than the features from the individual modalities. For such, we segmented paired CT and MRI nodules from 33 lung nodules patients, extracted 89 radiomics features from each modality, and combined it into a multimodality feature set. Those features were then used for classifying the nodules into benign and malignant by a set of machine learning algorithms, assessing the AUC across 30 trials. Our results show that MRI radiomics features are suitable for characterizing lung lesions, yielding AUC values up to 17% higher than their CT counterparts, and shedding light on MRI as a viable image modality for decision support systems. Conversely, our multimodality approach did not improve performance compared to the single-modality models, suggesting that the direct combination of multimodality features might not be an adequate strategy for dealing with multimodality medical images.

References

Beckett, K. R., Moriarity, A. K., and Langer, J. M. (2015). Safe use of contrast media: what the radiologist needs to know. Radiographics, 35(6):1738–1750.

Blandin Knight, S., Crosbie, P. A., Balata, H., Chudziak, J., Hussell, T., and Dive, C. (2017). Progress and prospects of early detection in lung cancer. Open biology, 7(9):170070.

Cawley, G. C. and Talbot, N. L. (2010). On over-tting in model selection and subsequent selection bias in performance evaluation. The Journal of Machine Learning Research, 11:2079–2107.

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: synthetic minority over-sampling technique. Journal of articial intelligence research, 16:321–357.

Claesen, M. and De Moor, B. (2015). Hyperparameter search in machine learning. arXiv preprint arXiv:1502.02127.

Demsar, J. (2006). Statistical comparisons of classiers over multiple data sets. Journal of Machine Learning Research, 7:1–30.

Fawcett, T. (2006). An introduction to roc analysis. Pattern recognition letters, 27(8):861–874.

Ferreira, J. R., Oliveira, M. C., and de Azevedo-Marques, P. M. (2018). Characterization of pulmonary nodules based on features of margin sharpness and texture. Journal of digital imaging, 31(4):451–463.

Francisco, V., Koenigkam-Santos, M., Wada, D. T., Junior, J. R. F., Fabro, A. T., Cipriano, F. E. G., Quatrina, S. G., and de Azevedo-Marques, P. M. (2019). Computer-aided In XXVI Brazilian diagnosis of lung cancer in magnetic resonance imaging exams. Congress on Biomedical Engineering, pages 121–127. Springer.

Gillies, R. J., Kinahan, P. E., and Hricak, H. (2016). Radiomics: images are more than pictures, they are data. Radiology, 278(2):563–577.

Guo, Z., Li, X., Huang, H., Guo, N., and Li, Q. (2019). Deep Learning-Based Image Segmentation on Multimodal Medical Imaging. IEEE Transactions on Radiation and Plasma Medical Sciences, 3(2):162–169.

Hatt, M., Tixier, F., Visvikis, D., and Le Rest, C. C. (2017). Radiomics in pet/ct: more than meets the eye? Journal of Nuclear Medicine, 58(3):365–366.

Jatobá, A., Lima, L., Amorim, L., and Oliveira, M. (2020). Cnn hyperparameter optimization for pulmonary nodule classication. In Anais do XX Simpósio Brasileiro de Computação Aplicada à Saúde, pages 25–36, Porto Alegre, RS, Brasil. SBC.

Knight, S. B., Crosbie, P. A., Balata, H., Chudziak, J., Hussell, T., and Dive, C. (2017). Progress and prospects of early detection in lung cancer. Open Biology, 7(9).

Li, L., Zhao, X., Lu, W., and Tan, S. (2019a). Deep learning for variational multimodality tumor segmentation in pet/ct. Neurocomputing.

Li, S., Xu, P., Li, B., Chen, L., Zhou, Z., Hao, H., Duan, Y., Folkert, M., Ma, J., Huang, S., et al. (2019b). Predicting lung nodule malignancies by combining deep convolutional neural network and handcrafted features. Physics in Medicine & Biology, 64(17):175012.

Mu, W., Qi, J., Lu, H., Schabath, M., Balagurunathan, Y., Tunali, I., and Gillies, R. J. (2018). Radiomic biomarkers from pet/ct multi-modality fusion images for the prediction of immunotherapy response in advanced non-small cell lung cancer patients. In Medical Imaging 2018: Computer-Aided Diagnosis, volume 10575, page 105753S. International Society for Optics and Photonics.

Ohno, Y., Kauczor, H.-U., Hatabu, H., Seo, J. B., van Beek, E. J., and for Pulmonary Functional Imaging (IWPFI), I. W. (2018). Mri for solitary pulmonary nodule and mass assessment: current state of the art. Journal of Magnetic Resonance Imaging, 47(6):1437–1458.

Parekh, V. S. and Jacobs, M. A. (2019). Deep learning and radiomics in precision medicine. Expert review of precision medicine and drug development, 4(2):59–72.

Pastorino, U., Rossi, M., Rosato, V., Marchianò, A., Sverzellati, N., Morosi, C., Fabbri, A., Galeone, C., Negri, E., Sozzi, G., et al. (2012). Annual or biennial ct screening versus observation in heavy smokers: 5-year results of the mild trial. European Journal of Cancer Prevention, 21(3):308–315.

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830.

Rampinelli, C., De Marco, P., Origgi, D., Maisonneuve, P., Casiraghi, M., Veronesi, G., Spaggiari, L., and Bellomi, M. (2017). Exposure to low dose computed tomography for lung cancer screening and risk of cancer: secondary analysis of trial data and riskbenet analysis. bmj, 356:j347.

Ribeiro, M. X., Balan, A. G., Felipe, J. C., Traina, A. J., and Traina, C. (2009). Mining statistical association rules to select the most relevant medical image features. In Mining complex data, pages 113–131. Springer.

Siegel, R. L., Miller, K. D., and Jemal, A. (2018). Cancer statistics, 2018. CA: A Cancer Journal for Clinicians, 68(1):7–30.

Sieren, J. C., Ohno, Y., Koyama, H., Sugimura, K., and McLennan, G. (2010). Recent technological and application developments in computed tomography and magnetic resonance imaging for improved pulmonary nodule detection and lung cancer staging. Journal of Magnetic Resonance Imaging, 32(6):1353–1369.

Vaidya, M., Creach, K. M., Frye, J., Dehdashti, F., Bradley, J. D., and El Naqa, I. (2012). Combined PET/CT image characteristics for radiotherapy tumor response in lung cancer. Radiotherapy and Oncology, 102(2):239–245.

Valliéres, M., Freeman, C. R., Skamene, S. R., and El Naqa, I. (2015). A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities. Physics in Medicine and Biology, 60(14):5471– 5496.

Van Griethuysen, J. J., Fedorov, A., Parmar, C., Hosny, A., Aucoin, N., Narayan, V., Beets-Tan, R. G., Fillion-Robin, J.-C., Pieper, S., and Aerts, H. J. (2017). Computational radiomics system to decode the radiographic phenotype. Cancer research, 77(21):e104–e107.

Wei, L., Osman, S., Hatt, M., and El Naqa, I. (2019). Machine learning for radiomicsbased multimodality and multiparametric modeling. The quarterly journal of nuclear medicine and molecular imaging : ofcial publication of the Italian Association of Nuclear Medicine (AIMN) [and] the International Association of Radiopharmacology (IAR), [and] Section of the Society of.., 63(4):323–338.

Wild, C. P., Weiderpass, E., and Stewart, B. W. (2020). WORLD CANCER REPORT: cancer research for cancer development. International Agency for Research on Cancer.

Yang, Y., Feng, X., Chi, W., Li, Z., Duan, W., Liu, H., Liang, W., Wang, W., Chen, P., He, J., and Liu, B. (2018). Deep learning aided decision support for pulmonary nodules diagnosing: A review. Journal of Thoracic Disease, 10(Suppl 7):S867–S875.

Yi, C. A., Shin, K. M., Lee, K. S., Kim, B.-T., Kim, H., Kwon, O. J., Choi, J. Y., and Chung, M. J. (2008). Non–small cell lung cancer staging: efcacy comparison of integrated pet/ct versus 3.0-t whole-body mr imaging. Radiology, 248(2):632–642.

Zhu, L., Kolesov, I., Gao, Y., Kikinis, R., and Tannenbaum, A. (2014). An effective interactive medical image segmentation method using fast growcut. In MICCAI workshop on interactive medical image computing.