Radiomic Quantification for MRI Assessment of Sacroiliac Joints of Patients with Spondyloarthritis.

Spondyloarthritis (SpA) is a group of diseases primarily involving chronic inflammation of the spine and peripheral joints, as evaluated by magnetic resonance imaging (MRI). Considering the complexity of SpA, we performed a retrospective study to discover quantitative/radiomic MRI-based features correlated with SpA. We also investigated different fat-suppression MRI techniques to develop detection models for inflammatory sacroiliitis. Finally, these model results were compared with those of experienced musculoskeletal radiologists, and the concordance level was evaluated. Examinations of 46 consecutive patients were obtained using SPAIR (spectral attenuated inversion recovery) and STIR (short tau inversion recovery) MRI sequences. Musculoskeletal radiologists manually segmented the sacroiliac joints for further extraction of 230 MRI features from gray-level histogram/matrices and wavelet filters. These features were associated with sacroiliitis, SpA, and the current biomarkers of ESR (erythrocyte sedimentation rate), CRP (C-reactive protein), BASDAI (Bath Ankylosing Spondylitis Activity Index), BASFI (Bath Ankylosing Spondylitis Functional Index), and MASES (Maastricht Ankylosing Spondylitis Enthesis Score). The Mann-Whitney U test showed that the radiomic markers from both MRI sequences were associated with active sacroiliitis and with SpA and its axial and peripheral subtypes (p < 0.05). Spearman's coefficient also identified a correlation between MRI markers and data from clinical practice (p < 0.05). Fat-suppression MRI models yielded performances that were statistically equivalent to those of specialists and presented strong concordance in identifying inflammatory sacroiliitis. SPAIR and STIR acquisition protocols showed potential for the evaluation of sacroiliac joints and the composition of a radiomic model to support the clinical assessment of SpA.

Comparison between STIR and T2-weighted SPAIR sequences in the evaluation of inflammatory sacroiliitis: diagnostic performance and signal-to-noise ratio.

OBJECTIVE: To compare two different fat-saturated magnetic resonance imaging (MRI) techniques-STIR and T2 SPAIR-in terms of image quality, as well as in terms of their diagnostic performance in detecting sacroiliac joints (SIJ) active inflammation. MATERIALS AND METHODS: We included 69 consecutive patients with suspected spondyloarthritis undergoing MRI between 2012 and 2014. The signal-to-noise ratio (SNR) was calculated with the method recommended by the American College of Radiology. Two readers evaluated SIJ MRI following ASAS criteria to assess diagnostic performance regarding the detection of active SIJ inflammation. T1 SPIR Gd+ sequence was used as the reference standard. RESULTS: The mean SNR was 72.8 for the T1 SPIR Gd+ sequence, compared with 14.1 and 37.6 for the STIR and T2 SPAIR sequences, respectively. The sensitivity and specificity of STIR and SPAIR T2 sequences did not show any statistically significant differences, for the diagnosis of sacroiliitis with active inflammation. CONCLUSION: Our results corroborate those in the recent literature suggesting that STIR sequences are not superior to T2 SPAIR sequences for SIJ evaluation in patients with suspected spondyloarthritis. On 1.5-T MRI, T2-weighted SPAIR sequences provide better SNRs than do STIR sequences, which reinforces that T2 SPAIR sequences may be an advantageous option for the evaluation of sacroiliitis.

OBJETIVO: O objetivo deste estudo foi comparar a qualidade da imagem das técnicas de saturação de gordura T2 SPAIR e STIR e, adicionalmente, comparar o desempenho diagnóstico das duas sequências para detecção de inflamação ativa nas imagens de ressonância magnética (RM) das sacroilíacas. MATERIAIS E MÉTODOS: Foram incluídos 69 pacientes consecutivos que realizaram RM das articulações sacroilíacas no período de 2012 a 2014. O cálculo da relação sinal/ruído (RSR) foi realizado pelo método do Colégio Americano de Radiologia. O desempenho diagnóstico foi realizado utilizando os resultados da leitura das imagens da sequência T1 SPIR Gd+ como referência. RESULTADOS: A avaliação da RSR mostrou média de 72,8 para a sequência T1 SPIR Gd+. A sensibilidade e a especificidade das sequências STIR e SPAIR 2 para o diagnóstico de sacroiliite com inflamação ativa não apresentaram diferenças estatisticamente significantes. CONCLUSÃO: Nossos resultados reforçam a literatura recente sugerindo que não há superioridade da sequência STIR em relação à sequência SPAIR T2 para a avaliação das sacroilíacas em pacientes com espondiloartrite. A sequência SPAIR T2 apresenta melhor RSR em relação à sequência STIR em imagens de RM de 1,5 T, o que reforça que a sequência SPAIR T2 possa ser utilizada como opção vantajosa na avaliação da sacroiliite inflamatória.

Comparison between STIR and T2-weighted SPAIR sequences in the evaluation of inflammatory sacroiliitis: diagnostic performance and signal-to-noise ratio / Comparação do desempenho diagnóstico e da relação sinal-ruído das sequências SPAIR T2 e STIR na avaliação de sacroiliite inflamatória

Abstract Objective: To compare two different fat-saturated magnetic resonance imaging (MRI) techniques-STIR and T2 SPAIR-in terms of image quality, as well as in terms of their diagnostic performance in detecting sacroiliac joints (SIJ) active inflammation. Materials and Methods: We included 69 consecutive patients with suspected spondyloarthritis undergoing MRI between 2012 and 2014. The signal-to-noise ratio (SNR) was calculated with the method recommended by the American College of Radiology. Two readers evaluated SIJ MRI following ASAS criteria to assess diagnostic performance regarding the detection of active SIJ inflammation. T1 SPIR Gd+ sequence was used as the reference standard. Results: The mean SNR was 72.8 for the T1 SPIR Gd+ sequence, compared with 14.1 and 37.6 for the STIR and T2 SPAIR sequences, respectively. The sensitivity and specificity of STIR and SPAIR T2 sequences did not show any statistically significant differences, for the diagnosis of sacroiliitis with active inflammation. Conclusion: Our results corroborate those in the recent literature suggesting that STIR sequences are not superior to T2 SPAIR sequences for SIJ evaluation in patients with suspected spondyloarthritis. On 1.5-T MRI, T2-weighted SPAIR sequences provide better SNRs than do STIR sequences, which reinforces that T2 SPAIR sequences may be an advantageous option for the evaluation of sacroiliitis.

Resumo Objetivo: O objetivo deste estudo foi comparar a qualidade da imagem das técnicas de saturação de gordura T2 SPAIR e STIR e, adicionalmente, comparar o desempenho diagnóstico das duas sequências para detecção de inflamação ativa nas imagens de ressonância magnética (RM) das sacroilíacas. Materiais e Métodos: Foram incluídos 69 pacientes consecutivos que realizaram RM das articulações sacroilíacas no período de 2012 a 2014. O cálculo da relação sinal/ruído (RSR) foi realizado pelo método do Colégio Americano de Radiologia. O desempenho diagnóstico foi realizado utilizando os resultados da leitura das imagens da sequência T1 SPIR Gd+ como referência. Resultados: A avaliação da RSR mostrou média de 72,8 para a sequência T1 SPIR Gd+. A sensibilidade e a especificidade das sequências STIR e SPAIR 2 para o diagnóstico de sacroiliite com inflamação ativa não apresentaram diferenças estatisticamente significantes. Conclusão: Nossos resultados reforçam a literatura recente sugerindo que não há superioridade da sequência STIR em relação à sequência SPAIR T2 para a avaliação das sacroilíacas em pacientes com espondiloartrite. A sequência SPAIR T2 apresenta melhor RSR em relação à sequência STIR em imagens de RM de 1,5 T, o que reforça que a sequência SPAIR T2 possa ser utilizada como opção vantajosa na avaliação da sacroiliite inflamatória.

A study of MRI-based radiomics biomarkers for sacroiliitis and spondyloarthritis.

PURPOSE: To evaluate the performance of texture-based biomarkers by radiomic analysis using magnetic resonance imaging (MRI) of patients with sacroiliitis secondary to spondyloarthritis (SpA). RELEVANCE: The determination of sacroiliac joints inflammatory activity supports the drug management in these diseases. METHODS: Sacroiliac joints (SIJ) MRI examinations of 47 patients were evaluated. Thirty-seven patients had SpA diagnoses (27 axial SpA and ten peripheral SpA) which was established previously after clinical and laboratory follow-up. To perform the analysis, the SIJ MRI was first segmented and warped. Second, radiomics biomarkers were extracted from the warped MRI images for associative analysis with sacroiliitis and the SpA subtypes. Finally, statistical and machine learning methods were applied to assess the associations of the radiomics texture-based biomarkers with clinical outcomes. RESULTS: All diagnostic performances obtained with individual or combined biomarkers reached areas under the receiver operating characteristic curves ≥ 0.80 regarding SpA related sacroiliitis and and SpA subtypes classification. Radiomics texture-based analysis showed significant differences between the positive and negative SpA groups and differentiated the axial and peripheral subtypes (P < 0.001). In addition, the radiomics analysis was also able to correctly identify the disease even in the absence of active inflammation. CONCLUSION: We concluded that the application of the radiomic approach constitutes a potential noninvasive tool to aid the diagnosis of sacroiliitis and for SpA subclassifications based on MRI of sacroiliac joints.

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

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 ~ 80% (46 samples, 20 positive and 26 negative) as training and ~ 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.

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

Abstract 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 ∼ 80% (46 samples, 20 positive and 26 negative) as training and ∼ 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)

The pubo-femoral distance decreases with Pavlik harness treatment for developmental dysplasia of the hip in newborns.

OBJECTIVE: To evaluate the evolution of the ultrasonographic pubo-femoral distance (PFD) before and after Pavlik harness treatment for developmental dysplasia of the hip (DDH) in newborns. PATIENTS AND METHODS: Twenty-five patients (16.7 ± 10.4 days; 19 females, six males) diagnosed with DDH and treated using the Pavlik harness were included. Eighteen patients had bilateral, and seven unilateral DDH, with a total of 43 dysplastic hips. The seven non-dysplastic hips in unilateral cases were used for comparison. The PFD was measured in the coronal and axial planes with the hip flexed to approximately 90°, before and after an average of 93 days of treatment. The femoral head coverage was assessed in the coronal plane, and correlated with PFD values. RESULTS: In dysplastic hips, the mean PFD decreased from 6.1 ± 1.8 mm to 3.0 ± 0.7 mm in the axial (adjusted difference, 2.9 mm; p < 0.01), and from 5.9 ± 2.0 to 3.0 ± 0.6 mm in the coronal plane (adjusted difference 2.7 mm; p < 0.01). The femoral head coverage increased from 30.8 to 62.1%, and the mean differences of femoral head coverage and PFD were significantly correlated (p < 0.001). There was no difference between treated dysplastic and non-dysplastic hips. There was high intra- and inter-observer agreement for PFD measurements. CONCLUSION: The PFD decreased significantly after DDH treatment using the Pavlik harness in newborns, and showed significant correlation with the femoral head coverage improvement. PFD might be a reliable tool for monitoring DDH treatment in newborns treated using the Pavlik harness.

MRI assessment of bone marrow oedema in the sacroiliac joints of patients with spondyloarthritis: is the SPAIR T2w technique comparable to STIR?

OBJECTIVE: To compare short tau inversion-recovery (STIR) with another fat saturation method in the assessment of sacroiliac joint inflammation. METHODS: This prospective cross-sectional study comprised 76 spondyloarthritis (SpA) patients who underwent magnetic resonance imaging of the sacroiliac joints in a 1.5-T scanner, using STIR, spectral attenuated inversion recovery (SPAIR) T2w and spectral presaturation with inversion recovery (SPIR) T1w post-contrast sequences. Two independent readers (R1 and R2) assessed the images using the Spondyloarthritis Research Consortium of Canada (SPARCC) score. We assessed agreement of the SPARCC scores for SPAIR T2w and STIR with that for T1 SPIR post-contrast (reference standard) using the St. Laurent coefficient. We evaluated each sequence using the concordance correlation coefficient (CCC). RESULTS: We observed a strong agreement between STIR and SPAIR T2w sequences. Lin's CCC was 0.94 for R1 and 0.84 for R2 for STIR and 0.94 for R1 and 0.84 for R2 for SPAIR. The interobserver evaluation revealed a good CCC of 0.79 for SPAIR and 0.78 for STIR. CONCLUSION: STIR technique and SPAIR T2w sequence showed high agreement in the evaluation of sacroiliac joint subchondral bone marrow oedema in patients with SpA. SPAIR T2w may be an alternative to the STIR sequence for this purpose. KEY POINTS: • There are no studies evaluating which fat saturation technique should be used. • SPAIR T2w may be an alternative to STIR for sacroiliac joint evaluation. • The study will lead to changes in guidelines for spondyloarthritis.

Imaging assessment of glenohumeral dysplasia secondary to brachial plexus birth palsy.

OBJECTIVE: To assess imaging parameters related to the morphology of the glenohumeral joint in children with unilateral brachial plexus birth palsy (BPBP), in comparison with those obtained for healthy shoulders. MATERIALS AND METHODS: We conducted a retrospective search for cases of unilateral BPBP diagnosed at our facility. Only patients with a clinical diagnosis of unilateral BPBP were included, and the final study sample consisted of 10 consecutive patients who were assessed with cross-sectional imaging. The glenoid version, the translation of the humeral head, and the degrees of glenohumeral dysplasia were assessed. RESULTS: The mean diameter of the affected humeral heads was 1.93 cm, compared with 2.33 cm for those of the normal limbs. In two cases, there was no significant posterior displacement of the humeral head, five cases showed posterior subluxation of the humeral head, and the remaining three cases showed total luxation of the humeral head. The mean glenoid version angle of the affected limbs (90-α) was -9.6º, versus +1.6º for the normal, contralateral limbs. CONCLUSION: The main deformities found in this study were BPBP-associated retroversion of the glenoid cavity, developmental delay of the humeral head, and posterior translation of the humeral head.

OBJETIVO: Avaliar os parâmetros de imagem relacionados com a morfologia da articulação glenoumeral em crianças com paralisia obstétrica do plexo braquial (POPB) unilateral, comparando-os com os observados em ombros saudáveis. MATERIAIS E MÉTODOS: Foi realizada uma busca retrospectiva de casos de POPB unilateral diagnosticados em nossa instituição. Somente foram incluídos os pacientes com diagnóstico clínico de POPB unilateral, e a amostra final do estudo consistiu em 10 pacientes consecutivos avaliados por meio de imagens transversais. Foram avaliados a retroversão da cavidade glenoide, a translação da cabeça do úmero e o grau de displasia glenoumeral. RESULTADOS: A média do diâmetro da cabeça do úmero foi 1,93 cm nos membros afetados e 2,33 cm nos membros normais. Em dois casos, não houve deslocamento posterior significativo da cabeça do úmero, cinco casos apresentaram subluxação posterior da cabeça do úmero, e os três casos restantes apresentaram luxação total da cabeça do úmero. A média do ângulo de retroversão glenoide dos membros afetados (90-α) foi -9,6º, ao passo que a dos membros contralaterais normais foi +1,6º. CONCLUSÃO: As principais deformidades encontradas neste estudo foram retroversão da cavidade glenoide relacionada com POPB, atraso no desenvolvimento da cabeça do úmero e translação posterior da cabeça do úmero.

Imaging assessment of glenohumeral dysplasia secondary to brachial plexus birth palsy / Avaliação de imagens de displasia glenoumeral secundária a paralisia obstétrica do plexo braquial

Abstract Objective: To assess imaging parameters related to the morphology of the glenohumeral joint in children with unilateral brachial plexus birth palsy (BPBP), in comparison with those obtained for healthy shoulders. Materials and Methods: We conducted a retrospective search for cases of unilateral BPBP diagnosed at our facility. Only patients with a clinical diagnosis of unilateral BPBP were included, and the final study sample consisted of 10 consecutive patients who were assessed with cross-sectional imaging. The glenoid version, the translation of the humeral head, and the degrees of glenohumeral dysplasia were assessed. Results: The mean diameter of the affected humeral heads was 1.93 cm, compared with 2.33 cm for those of the normal limbs. In two cases, there was no significant posterior displacement of the humeral head, five cases showed posterior subluxation of the humeral head, and the remaining three cases showed total luxation of the humeral head. The mean glenoid version angle of the affected limbs (90-α) was -9.6º, versus +1.6º for the normal, contralateral limbs. Conclusion: The main deformities found in this study were BPBP-associated retroversion of the glenoid cavity, developmental delay of the humeral head, and posterior translation of the humeral head.

Resumo Objetivo: Avaliar os parâmetros de imagem relacionados com a morfologia da articulação glenoumeral em crianças com paralisia obstétrica do plexo braquial (POPB) unilateral, comparando-os com os observados em ombros saudáveis. Materiais e Métodos: Foi realizada uma busca retrospectiva de casos de POPB unilateral diagnosticados em nossa instituição. Somente foram incluídos os pacientes com diagnóstico clínico de POPB unilateral, e a amostra final do estudo consistiu em 10 pacientes consecutivos avaliados por meio de imagens transversais. Foram avaliados a retroversão da cavidade glenoide, a translação da cabeça do úmero e o grau de displasia glenoumeral. Resultados: A média do diâmetro da cabeça do úmero foi 1,93 cm nos membros afetados e 2,33 cm nos membros normais. Em dois casos, não houve deslocamento posterior significativo da cabeça do úmero, cinco casos apresentaram subluxação posterior da cabeça do úmero, e os três casos restantes apresentaram luxação total da cabeça do úmero. A média do ângulo de retroversão glenoide dos membros afetados (90-α) foi -9,6º, ao passo que a dos membros contralaterais normais foi +1,6º. Conclusão: As principais deformidades encontradas neste estudo foram retroversão da cavidade glenoide relacionada com POPB, atraso no desenvolvimento da cabeça do úmero e translação posterior da cabeça do úmero.

Association Between Lumbar Spine Sagittal Alignment and L4-L5 Disc Degeneration Among Asymptomatic Young Adults.

STUDY DESIGN: Cross-sectional observational study on the relationship between the degrees of disc degeneration and sagittal alignment in asymptomatic healthy individuals. OBJECTIVE: This study sought to determine whether the sagittal spine alignment subtype is related to the prevalence of lumbar disc degeneration. SUMMARY AND BACKGROUND DATA: Sagittal balance and spinopelvic parameters might be risk factors for disc degeneration. METHODS: A total of 70 asymptomatic participants (36 women and 34 men) without regular physical activity were categorized according to the four subtypes of sagittal alignment proposed by Roussouly. All participants underwent magnetic resonance imaging of the lumbar spine (1.5T) and panoramic radiography of the spine. The degree of disc degeneration was graded using T2-weighted images according to the Pfirrmann classification. Spinopelvic parameters and vertebral curvatures were measured on digital panoramic radiographs using Surgimap software. Interobserver analyses for the Pfirrmann classification and spinopelvic parameters were assessed using the weighted Kappa and intraclass correlation coefficient (ICC), respectively. RESULTS: The Kappa associated with disc degeneration classification was 0.79 (95% confidence intervals 0.72-0.87). The ICCs were excellent, with small confidence intervals for all spinopelvic parameters. The type II group (flat lordosis) showed a higher frequency of degenerated discs at L4-L5 (P = 0.03) than the type IV group (long and curved lumbar spine). No significant differences in disc degeneration were observed among the four subtypes at the other disc levels. We found a negative, moderate correlation between the spinopelvic parameters and the occurrence of disc degeneration in the type II group. CONCLUSION: The Roussouly subtype II sagittal alignment is significantly associated with disc degeneration at L4-L5 in asymptomatic young adults. Our results support the hypothesis that spinal sagittal alignment plays a role in early disc degeneration. LEVEL OF EVIDENCE: 3.

Avaliação computadorizada de sacroiliíte em imagens de ressonância magnética / Computer evaluation of sacroiliitis in magnetic resonance imaging

OBJETIVOS: avaliar e classificar a atividade inflamatória nas articulações sacroilíacas de pacientes com espondiloartrite em imagens de ressonância magnética, utilizando atributos de textura e de histograma de níveis de cinza. MÉTODOS: imagens de 51 pacientes foram avaliadas retrospectivamente e segmentadas manualmente por um radiologista. Trinta e nove atributos de brilho e de textura foram utilizados para caracterizar a presença ou ausência de processo inflamatório. A classificação foi realizada utilizando-se diferentes classificadores e avaliada por um método de validação cruzada com 10-fold. RESULTADOS: uma rede neural multicamadas, utilizando o conjunto total de atributos, alcançou o melhor desempenho no estudo, obtendo 0,915 de área sob a curva ROC, 0,864 de sensibilidade e 0,724 de especificidade. CONCLUSÕES: o processamento computadorizado implementado possui bom potencial como base para o desenvolvimento de uma ferramenta de auxílio ao diagnóstico de processo inflamatório de articulações sacroilíacas de pacientes com espondiloartrites.

GOAL: to evaluate and classify the inflammatory process in sacroiliac joints of patients with spondyloarthritis in magnetic resonance imaging using attributes of texture and gray-level histogram. METHODS: images from 51 patients were retrospectively evaluated and manually segmented by a radiologist. Thirty nine attributes of histogram and texture were used to characterize the presence or absence of the inflammatory process. Classification was performed by several classifiers and evaluated with a 10-fold cross-validation. RESULTS: a multilayer neural network and all extracted attributes obtained highest diagnostic performance in the study with 0.915 of area under the ROC curve, 0.864 of sensitivity and 0.724of specificity. CONCLUSIONS: the implemented computerized processing presents good potential as a starting point for the development of a tool to aid the diagnosis of inflammatory process of sacroiliac joints of patients with spondyloarthritis.