Development of a convolutional neural network for diagnosing osteoarthritis, trained with knee radiographs from the ELSA-Brasil Musculoskeletal / Desenvolvimento de rede neural convolucional para o diagnóstico radiográfico de osteoartrite dos joelhos no ELSA-Brasil Musculoesquelético

Abstract Objective: To develop a convolutional neural network (CNN) model, trained with the Brazilian "Estudo Longitudinal de Saúde do Adulto Musculoesquelético" (ELSA-Brasil MSK, Longitudinal Study of Adult Health, Musculoskeletal) baseline radiographic examinations, for the automated classification of knee osteoarthritis. Materials and Methods: This was a cross-sectional study carried out with 5,660 baseline posteroanterior knee radiographs from the ELSA-Brasil MSK database (5,660 baseline posteroanterior knee radiographs). The examinations were interpreted by a radiologist with specific training, and the calibration was as established previously. Results: The CNN presented an area under the receiver operating characteristic curve of 0.866 (95% CI: 0.842-0.882). The model can be optimized to achieve, not simultaneously, maximum values of 0.907 for accuracy, 0.938 for sensitivity, and 0.994 for specificity. Conclusion: The proposed CNN can be used as a screening tool, reducing the total number of examinations evaluated by the radiologists of the study, and as a double-reading tool, contributing to the reduction of possible interpretation errors.

Resumo Objetivo: Desenvolver um modelo computacional - rede neural convolucional (RNC) - treinado com radiografias da linha de base do Estudo Longitudinal de Saúde do Adulto Musculoesquelético (ELSA-Brasil Musculoesquelético), para a classificação automática de osteoartrite dos joelhos. Materiais e Métodos: Trata-se de um estudo transversal abrangendo todos os exames da linha de base do ELSA-Brasil Musculoesquelético (5.660 radiografias dos joelhos em incidência posteroanterior). Os exames foram interpretados por médico radiologista com treinamento específico e calibração previamente publicada. Resultados: A RNC desenvolvida apresentou área sob a curva característica de operação do receptor de 0,866 (IC 95%: 0,842-0,882). O modelo pode ser calibrado para alcançar, não simultaneamente, valores máximos de 0,907 para acurácia, 0,938 para sensibilidade e 0,994 para especificidade. Conclusão: A RNC desenvolvida pode ser utilizada como ferramenta de triagem, reduzindo o número total de exames avaliados pelos radiologistas do estudo, e/ou como ferramenta de segunda leitura, contribuindo com a redução de possíveis erros de interpretação.

Development of a convolutional neural network for diagnosing osteoarthritis, trained with knee radiographs from the ELSA-Brasil Musculoskeletal.

Objective: To develop a convolutional neural network (CNN) model, trained with the Brazilian "Estudo Longitudinal de Saúde do Adulto Musculoesquelético" (ELSA-Brasil MSK, Longitudinal Study of Adult Health, Musculoskeletal) baseline radiographic examinations, for the automated classification of knee osteoarthritis. Materials and Methods: This was a cross-sectional study carried out with 5,660 baseline posteroanterior knee radiographs from the ELSA-Brasil MSK database (5,660 baseline posteroanterior knee radiographs). The examinations were interpreted by a radiologist with specific training, and the calibration was as established previously. Results: The CNN presented an area under the receiver operating characteristic curve of 0.866 (95% CI: 0.842-0.882). The model can be optimized to achieve, not simultaneously, maximum values of 0.907 for accuracy, 0.938 for sensitivity, and 0.994 for specificity. Conclusion: The proposed CNN can be used as a screening tool, reducing the total number of examinations evaluated by the radiologists of the study, and as a double-reading tool, contributing to the reduction of possible interpretation errors.

Objetivo: Desenvolver um modelo computacional - rede neural convolucional (RNC) - treinado com radiografias da linha de base do Estudo Longitudinal de Saúde do Adulto Musculoesquelético (ELSA-Brasil Musculoesquelético), para a classificação automática de osteoartrite dos joelhos. Materiais e Métodos: Trata-se de um estudo transversal abrangendo todos os exames da linha de base do ELSA-Brasil Musculoesquelético (5.660 radiografias dos joelhos em incidência posteroanterior). Os exames foram interpretados por médico radiologista com treinamento específico e calibração previamente publicada. Resultados: A RNC desenvolvida apresentou área sob a curva característica de operação do receptor de 0,866 (IC 95%: 0,842-0,882). O modelo pode ser calibrado para alcançar, não simultaneamente, valores máximos de 0,907 para acurácia, 0,938 para sensibilidade e 0,994 para especificidade. Conclusão: A RNC desenvolvida pode ser utilizada como ferramenta de triagem, reduzindo o número total de exames avaliados pelos radiologistas do estudo, e/ou como ferramenta de segunda leitura, contribuindo com a redução de possíveis erros de interpretação.

Chronic inflammatory diseases, subclinical atherosclerosis, and cardiovascular diseases: Design, objectives, and baseline characteristics of a prospective case-cohort study ‒ ELSA-Brasil.

OBJECTIVES: This analysis describes the protocol of a study with a case-cohort to design to prospectively evaluate the incidence of subclinical atherosclerosis and Cardiovascular Disease (CVD) in Chronic Inflammatory Disease (CID) participants compared to non-diseased ones. METHODS: A high-risk group for CID was defined based on data collected in all visits on self-reported medical diagnosis, use of medicines, and levels of high-sensitivity C-Reactive Protein >10 mg/L. The comparison group is the Aleatory Cohort Sample (ACS): a group with 10% of participants selected at baseline who represent the entire cohort. In both groups, specific biomarkers for DIC, markers of subclinical atherosclerosis, and CVD morbimortality will be tested using weighted Cox. RESULTS: The high-risk group (n = 2,949; aged 53.6 ± 9.2; 65.5% women) and the ACS (n=1543; 52.2±8.8; 54.1% women) were identified. Beyond being older and mostly women, participants in the high-risk group present low average income (29.1% vs. 24.8%, p < 0.0001), higher BMI (Kg/m2) (28.1 vs. 26.9, p < 0.0001), higher waist circumference (cm) (93.3 vs. 91, p < 0.0001), higher frequencies of hypertension (40.2% vs. 34.5%, p < 0.0001), diabetes (20.7% vs. 17%, p = 0.003) depression (5.8% vs. 3.9%, p = 0.007) and higher levels of GlycA a new inflammatory marker (p < 0.0001) compared to the ACS. CONCLUSIONS: The high-risk group selected mostly women, older, lower-income/education, higher BMI, waist circumference, and of hypertension, diabetes, depression, and higher levels of GlycA when compared to the ACS. The strategy chosen to define the high-risk group seems adequate given that multiple sociodemographic and clinical characteristics are compatible with CID.

Chronic inflammatory diseases, subclinical atherosclerosis, and cardiovascular diseases: Design, objectives, and baseline characteristics of a prospective case-cohort study ‒ ELSA-Brasil

Abstract Objectives This analysis describes the protocol of a study with a case-cohort to design to prospectively evaluate the incidence of subclinical atherosclerosis and Cardiovascular Disease (CVD) in Chronic Inflammatory Disease (CID) participants compared to non-diseased ones. Methods A high-risk group for CID was defined based on data collected in all visits on self-reported medical diagnosis, use of medicines, and levels of high-sensitivity C-Reactive Protein >10 mg/L. The comparison group is the Aleatory Cohort Sample (ACS): a group with 10% of participants selected at baseline who represent the entire cohort. In both groups, specific biomarkers for DIC, markers of subclinical atherosclerosis, and CVD morbimortality will be tested using weighted Cox. Results The high-risk group (n = 2,949; aged 53.6 ± 9.2; 65.5% women) and the ACS (n=1543; 52.2±8.8; 54.1% women) were identified. Beyond being older and mostly women, participants in the high-risk group present low average income (29.1% vs. 24.8%, p < 0.0001), higher BMI (Kg/m2) (28.1 vs. 26.9, p < 0.0001), higher waist circumference (cm) (93.3 vs. 91, p < 0.0001), higher frequencies of hypertension (40.2% vs. 34.5%, p < 0.0001), diabetes (20.7% vs. 17%, p = 0.003) depression (5.8% vs. 3.9%, p = 0.007) and higher levels of GlycA a new inflammatory marker (p < 0.0001) compared to the ACS. Conclusions The high-risk group selected mostly women, older, lower-income/education, higher BMI, waist circumference, and of hypertension, diabetes, depression, and higher levels of GlycA when compared to the ACS. The strategy chosen to define the high-risk group seems adequate given that multiple sociodemographic and clinical characteristics are compatible with CID.

Different Components of Subjective Well-being Are Associated With Chronic Nondisabling and Disabling Knee Pain: ELSA-Brasil Musculoskeletal Cohort.

BACKGROUND/OBJECTIVE: Chronic knee pain (CKP) is a common pain complaint in older adults that is often associated with disability. This study investigated the relationship between 2 components of subjective well-being (depressive symptoms and life satisfaction) and CKP phenotypes based on the presence of knee disability. METHODS: A cross-sectional study was performed at baseline of ELSA-Brasil Musculoskeletal cohort (2012-2014). Chronic knee pain phenotypes were identified according to the presence of CKP that was accompanied or not by disability, which was assessed by a question on pain-related limitations to perform everyday activities (overall), Western Ontario and McMaster Universities Osteoarthritis Index's physical function subscale (daily tasks) and 5-times sit-to-stand test (objective). Depressive symptoms were assessed by the Clinical Interview Schedule-Revised and life satisfaction by the Satisfaction With Life Scale. Multinomial logistic regressions used CKP phenotypes as response variables (no CKP as reference). RESULTS: The sample comprised 2898 participants (mean age, 55.9 ± 8.9 years; 52.9% were female). After adjustments for sociodemographic and clinical factors, depressive symptoms were associated with daily tasks disabling CKP (odds ratio [OR], 2.30; 95% confidence interval [CI], 1.45-3.66) and objective disabling CKP (OR, 1.95; 95% CI, 1.29-2.93) and with nondisabling CKP for the overall disability measure (OR, 1.54; 95% CI, 1.17-2.04). Life satisfaction was inversely associated with all phenotypes in fully adjusted models, with strongest magnitude of associations observed for disabling CKP. CONCLUSIONS: The association of depressive symptoms and life satisfaction with CKP phenotypes suggest the need to address both negative and positive components of subjective well-being in the assessment of individuals with knee complaints.

Performance of distinct knee osteoarthritis classification criteria in the ELSA-Brasil musculoskeletal study.

OBJECTIVE: To compare the performance of different knee osteoarthritis (OA) classification criteria in a subsample of subjects of ELSA-Brasil Musculoskeletal cohort (ELSA-Brasil MSK) in comparison with the clinical-radiographic evaluation of an experienced rheumatologist. METHOD: A cross-sectional study of the diagnostic accuracy of different knee OA classification criteria: radiographic OA, symptomatic OA, clinical and clinical-radiographic criteria of the American College of Rheumatology (ACR), and the definition proposed by the National Institute for Health and Care Excellence (NICE), UK. The study also evaluated some alternative definitions of OA including a combination of criteria. In total, 250 subjects participated. Only one knee per subject was included. When both or none of the knees were affected, one knee was randomly selected for analysis. OA prevalence, sensitivity, specificity, positive and negative predictive values, and accuracy were presented (CI 95%; α = 5%). RESULTS: The mean age was 56.1 years (SD = 8.7), 48.8% female, and 39.2% presented knee OA according to the rheumatologist. The sensitivity and the specificity of radiographic OA were 51% and 96.7%, respectively, while the NICE definition showed 57.0% and 76.3%, respectively. The other OA criteria showed good levels of specificity, but the levels of sensitivity were below 30%. Considering NICE and/or X-ray result in combination, the sensitivity increased (73.4%), while the specificity reduced slightly (73.0%) in relation to the original NICE definition. CONCLUSIONS: Radiographic OA showed the best performance, followed by NICE definition, especially in combination with X-ray results.

The McKenzie method for low back pain: a systematic review of the literature with a meta-analysis approach.

STUDY DESIGN AND OBJECTIVES: Meta-analysis of randomized controlled trials to evaluate the effectiveness of the McKenzie method for low back pain (LBP). SUMMARY OF BACKGROUND DATA: The McKenzie method is a popular classification-based treatment for LBP. The faulty equation of McKenzie to extension exercises (generic McKenzie) is common in randomized trials. METHODS: MEDLINE, EMBASE, PEDro, and LILACS were searched up to August 2003. Two independent reviewers extracted the data and assessed methodologic quality. Pooled effects were calculated among homogeneous trials using the random effects model. A sensitivity analysis excluded trials reporting on generic McKenzie. RESULTS: Eleven trials of mostly high quality were included. McKenzie reduced pain (weighted mean difference [WMD] on a 0- to 100-point scale, -4.16 points; 95% confidence interval, -7.12 to -1.20) and disability (WMD on a 0- to 100-point scale, -5.22 points; 95% confidence interval, -8.28 to -2.16) at 1 week follow-up when compared with passive therapy for acute LBP. When McKenzie was compared with advice to stay active, a reduction in disability favored advice (WMD on a 0- to 100-point scale, 3.85 points; 95% confidence interval, 0.30 to 7.39) at 12 weeks of follow-up. Heterogeneity prevented pooling of studies on chronic LBP as well as pooling of studies included in the sensitivity analysis. CONCLUSIONS: There is some evidence that the McKenzie method is more effective than passive therapy for acute LBP; however, the magnitude of the difference suggests the absence of clinically worthwhile effects. There is limited evidence for the use of McKenzie method in chronic LBP. The effectiveness of classification-based McKenzie is yet to be established.

Comparaçäo clínica e funcional de pacientes com fibromialgia e dor miofascial / Clinical and functional comparison of patients with fibromyalgia and mioyfalcial pain

Objetivo: Estabelecer as diferenças e semelhanças entre mulheres com fibromialgia (FM) e mulheres com dor miofascial regional (DM) do ponto de vista clínico, funcional e qualidade de vida. Material e métodos: Estudo transversal de 26 mulheres que preencheram os critérios de classificaçäo de fibromialgia da American College of Rheumatology e 18 mulheres com dor músculo-esquelética regional associada à pontos-gatilho dolorosos na área afetada pela dor. Foram analisados estatisticamente pela disciplina de Bioestatística da Faculdade os seguintes parâmetros:intensidade da dor (escala analógica numérica de dor-0 a 10-END); intensidade de fadiga (escala analógica numérica fadiga-0 a 10-END);no de pontos dolorosos/gatilho; capacidade funcional (Health Assessment Questionnaire - 0 a 3 - HAQ); qfualidade do sono (Post-sleep Inventory - 0 a 120 - PSI); intensidade de sintomas depressivos (Bech Depression Inventory - 0 a 60 -BDI); e qualidade de vida global (Fibromyalgia Impacti Questionnaire - 0 a 100 - FIQ). Conclusöes: As semelhanças observadas entre pacientes com FM e DM säo: dados demográficos, intensidade da dor e fadiga, presença de fatores desencadeantes e nível de atividade física. As diferenças observadas foram: escolaridade, capacidade funcional, qualidade do sono, intensidade da depressäo e qualidade de vida.