Computerized Recognition of Inflammatory Patterns of Sacroiliitis in MRI Images

• Matheus Calil Faleiros USP
• José Raniery Ferreira Junior USP
• Eddy Javala Jens USP
• Vitor Faeda Dalto USP
• Marcello Henrique Nogueira-Barbosa USP
• Paulo Mazzoncini de Azevedo-Marques USP

DOI: https://doi.org/10.5753/sbcas.2017.3731

Abstract

The reference standard to evaluate active inflammation of sacroiliac joints (SJ) in spondyloarthritis is magnetic resonance imaging (MRI). However, it may be challenging to specialists due to clinical variability. In this context, we aim to recognize inflammatory patterns of SJ using gray-level, texture, and spectral features. Features have been extracted from MRI exams of 51 patients and selected by the ReliefF feature selection method. Image classification was performed by machine learning methods and was assessed by the area under the receiver operating characteristic curve, with a 10-fold cross validation. Results have shown that the five nearest neighbors classifier presented the best performance for inflammatory pattern recognition.

References

Barros, P. (2011) “Epidemiology of Spondyloarthritis in Brazil”, The American Journal of the Medical Sciences, 341(4), 287-288.

Dalto, V., Assad, R., Crema, M., Louzada-Junior, P., Nogueira-Barbosa, M. (2017) “MRI assessment of bone marrow oedema in the sacroiliac joints of patients with spondyloarthritis: is the SPAIR T2w technique comparable to STIR?”, European Radiology, DOI: 10.1007/s00330-017-4746-7.

Frank, E., Hall, M., Witten, I. (2016) “The WEKA Workbench”. Online Appendix for Data Mining: Practical Machine Learning Tools and Techniques, Fourth Edition.

Gonzalez, R., Woods, R. (1993) Digital image processing, Addison-Wesley.

Haralick, R., Shanmugam, K., Dinstein, I. (1973) “Textural Features for Image Classification”. IEEE Transactions on Systems, Man, and Cybernetics, 3(6), 610-621.

JFeatureLib open source project. [link], 02/05/2016. Fontes: Haralick.java - Autor: graf ; Tamura.java - Autor: Marko Keuschnig & Christian Penz ; Histogram.java - Autor: graf;

Kononenko, I. (1994) “Estimating attributes: analysis and extensions of RELIEF”, Machine Learning: ECML-94. Springer Berlin Heidelberg, 171-182.

Maksymowych, W., Inman, R., Salonen, D., Dhillon, S., Williams, M., Stone, M., Spady, B., Palsat, J., Lambert, R. (2005) “Spondyloarthritis research Consortium of Canada magnetic resonance imaging index for assessment of sacroiliac joint inflammation in ankylosing spondylitis”, Arthritis & Rheumatism, 53(5), 703-709.

Pialat, J., Di Marco, L., Feydy, A., Peyron, C., Porta, B., Himpens, P., Boudrigua, A., Aubry, S. (2016) “Sacroiliac joints imaging in axial spondyloarthritis”, Diagnostic and Interventional Imaging, 97(7), 697-708.

Schneider, C., Rasband, W., Eliceiri, K. (2012) "NIH Image to ImageJ: 25 years of image analysis", Nature Methods 9(7): 671-675.

Tamura, H., Mori, S., Yamawaki, T. (1978) “Textural Features Corresponding to Visual Perception”. IEEE Transactions on Systems, Man, and Cybernetics, 8(6), 460-473.