Navegando por Palavras-chave "Tomografia computadorizada por raios x"

Agora exibindo 1 - 2 de 2
  • Nenhuma Miniatura disponível
    Item
    Somente Metadadados
    Avaliação do valor da tomografia computadorizada de alta resolução no diagnóstico das doenças parenquimatosas pulmonares difusas
    (Universidade Federal de São Paulo (UNIFESP), 2013-09-19) Verrastro, Carlos Gustavo Yuji [UNIFESP]; Meirelles, Gustavo de Souza Portes [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
    Objectives: 1) to evaluate the agreement between the diagnostic hypothesis (DH) made by radiologists on high resolution computed tomography (HRCT) and the final diagnosis in patients with diffuse parenchymal lung diseases (DPLD); 2) to evaluate whether clinical data would modify the agreement between HRCT DH and the final diagnosis and 3) to assess if the degree of confidence of the radiologist for his HRCT DH influences the agreement with the final diagnosis. Methods: Two radiologists independently and blindly reviewed HRCT scans of patients with DPLD. The agreement between their first DH and between any of the three DH were evaluated before and after the access of radiologists to the clinical data, with study of the importance of the degree of confidence in the DH formulated. Results: In the whole sample the concordance of HRCT DH and final diagnosis were on average 48% and 76% when considering the first or any of the DH, respectively, without access to clinical data. Accessing clinical data improved concordance between radiologists DH and the final diagnosis, especially for hypersensitivity pneumonitis. In this case the sensitivity of HRCT improved on average from 70.5% to 89.5%. The DH formulated with high degree of confidence were correct on average in 69% of the cases. Conclusions: 1) the first DH made in HRCT analysis was in agreement with the final diagnosis in 50% of the cases. When considering any of the three DH, the agreement increased to about 75% of the cases. These numbers were higher (56% and 85%, respectively) for the most common disease in our sample, which was usual interstitial pneumonia. 2) knowledge of clinical data improved the agreement between the first DC made in HRCT analysis and the final diagnosis, but had little influence when considering any of the three DH. Access to clinical data had most significant impact in patients with hypersensitivity pneumonitis and lymphocytic interstitial pneumonia. 3) Increasing the degree of confidence of radiologists in their HRCT DH had a direct relationship with greater agreement between their DH and the final diagnosis.
  • Nenhuma Miniatura disponível
    Item
    Somente Metadadados
    Concordância entre radiologistas na quantificação de bronquiectasias pela tomografia computadorizada de alta resolução
    (Universidade Federal de São Paulo (UNIFESP), 2014-11-26) Brito, Milene Carneiro Barbosa de [UNIFESP]; Meirelles, Gustavo de Souza Portes [UNIFESP]; Universidade Federal de São Paulo (UNIFESP)
    OBJECTIVES: To determinethe agreement on the quantification of HRCT. METHODS: HRCT scans of 43 patients with bronchiectasis were analyzed by two radiologists. Findings were graded with a score, and kappa values and overall agreement were calculated. RESULTS: Measurement and appearance of bronchiectasis showed moderate interobserver (k = 0.45 and k = 0.43, respectively) and intraobserver agreement (k = 0.54 and k = 0.47, respectively). Agreement on the presence of mucous plugging was fair(central distribution: overall interobserver agreement of 68.3%, k = 0.39 for intraobserver agreement; peripheral distribution: k = 0.34 and k = 0.35 for inter- and intraobserver agreement, respectively) and forperibronchial thickening (k = 0.21 and k = 0.30 for interand intraobserver agreement, respectively). Agreement on the detection of opacities, ground-glass areas, and cysts/bullae was fair (k = 0.39, overall agreement of 64.3%, and k = 0.47 for interobserver agreement, and overall agreement of 71.9%, k = 0.24, and k = 0.44for intraobserver agreement, respectively). After qualitative analysis of the findings of bronchiectasis on HRCT, the total score for each patient was calculated, showing excellent correlation between observers (intraclass correlation coefficient of 0.85 for interobserver agreement and of 0.81 for intraobserver agreement). CONCLUSION: HRCT findings of bronchiectasis showed fair agreement between observers. After final analysis of the findings using the proposed score, we observed an excellent inter- and intra-observer correlation.