Machine learning techniques for computer-aided classification of active inflammatory sacroiliitis in magnetic resonance imaging

Autor(es)

Faleiros, Matheus Calil ^[1] ; Nogueira-Barbosa, Marcello Henrique ^{[2, 3, 4, 5]} ; Dalto, Vitor Faeda ^[3] ; Ferreira Junior, Jose Raniery ^{[2, 4]} ; Magalhaes Tenorio, Ariane Priscilla ^[2] ; Luppino-Assad, Rodrigo ^[2] ; Louzada-Junior, Paulo ^[2] ; Rangayyan, Rangaraj Mandayam ^[6] ; de Azevedo-Marques, Paulo Mazzoncini ^{[2, 4]}

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Sao Carlos Sch Engn, Sao Carlos, SP - Brazil

^[2] Univ Sao Paulo, Ribeirao Preto Med Sch, Ribeirao Preto, SP - Brazil

^[3] Ribeirao Preto Med Sch, Musculoskeletal Imaging Res Lab, Ribeirao Preto - Brazil

^[4] Ribeirao Preto Med Sch, MAlnLab Med Artificial Intelligence Lab, Ribeirao Preto - Brazil

^[5] Ribeirao Preto Med Sch, CCIFM, Radiol Div, Av Bandeirantes 3900, BR-14048900 Ribeirao Preto, SP - Brazil

^[6] Univ Calgary, Schulich Sch Engn, Elect & Comp Engn, Calgary, AB - Canada

• Tipo de documento: Artigo Científico
• Fonte: ADVANCES IN RHEUMATOLOGY; v. 60, n. 1 MAY 7 2020.
• Citações Web of Science: 1

Resumo

Background Currently, magnetic resonance imaging (MRI) is used to evaluate active inflammatory sacroiliitis related to axial spondyloarthritis (axSpA). The qualitative and semiquantitative diagnosis performed by expert radiologists and rheumatologists remains subject to significant intrapersonal and interpersonal variation. This encouraged us to use machine-learning methods for this task. Methods In this retrospective study including 56 sacroiliac joint MRI exams, 24 patients had positive and 32 had negative findings for inflammatory sacroiliitis according to the ASAS group criteria. The dataset was randomly split with similar to 80% (46 samples, 20 positive and 26 negative) as training and similar to 20% as external test (10 samples, 4 positive and 6 negative). After manual segmentation of the images by a musculoskeletal radiologist, multiple features were extracted. The classifiers used were the Support Vector Machine, the Multilayer Perceptron (MLP), and the Instance-Based Algorithm, combined with the Relief and Wrapper methods for feature selection. Results Based on 10-fold cross-validation using the training dataset, the MLP classifier obtained the best performance with sensitivity = 100%, specificity = 95.6% and accuracy = 84.7%, using 6 features selected by the Wrapper method. Using the test dataset (external validation) the same MLP classifier obtained sensitivity = 100%, specificity = 66.7% and accuracy = 80%. Conclusions Our results show the potential of machine learning methods to identify SIJ subchondral bone marrow edema in axSpA patients and are promising to aid in the detection of active inflammatory sacroiliitis on MRI STIR sequences. Multilayer Perceptron (MLP) achieved the best results. (AU)

• Processo FAPESP: 18/07765-6 - Visão computacional e reconhecimento de padrão para identificação de biomarcadores radiômicos
• Beneficiário: Paulo Mazzoncini de Azevedo Marques
• Linha de fomento: Auxílio à Pesquisa - Pesquisador Visitante - Internacional
• Processo FAPESP: 16/17078-0 - Mineração, indexação e visualização de Big Data no contexto de sistemas de apoio à decisão clínica (MIVisBD)
• Beneficiário: Agma Juci Machado Traina
• Linha de fomento: Auxílio à Pesquisa - Temático