Objective： The aim of our study was to investigate the value of dynamic contrast-enhanced MRI for evaluating differential diagnosis of pulmonary isolated lesions. Methods： Twenty-nine consecutive patients enrolled i...Objective： The aim of our study was to investigate the value of dynamic contrast-enhanced MRI for evaluating differential diagnosis of pulmonary isolated lesions. Methods： Twenty-nine consecutive patients enrolled in this study, all of whom underwent DCE-MRI examinations and received a histologic and clinical diagnosis. Among these, lung tuberculoma 7 cases, harmatoma 3 cases, peripheral lung cancer 19 cases. DCE-MRI was acquired with 3D LAVA technique, total 18 phases were acquired, scanner time of per phase was 5-7″. After contrasting agent, twice successive scanning was acquired at 10″ and 50″. Then 1′30″, 2′, 2′30″, 3′, 3′30″, 4′, 5′, 6′, 7′, 8′, 9′, 10′, 11′, 12′ performed scanning. Region of interest was placed on the Maximum level in the tumors. According to Schaefer＇s standard, four types of time signal intensity curve （TIC） were classified, which were A, B, C and D. Compared the dynamic parameters between benign and malignant nodules. Results： Lung tuberculoma may display three curves： A type 1 case, ring-shaped enhancement 4 cases （periphery ring A type, central region D type）, D type 2 cases. Harmatoma may display three curves： A type 1 case, C type 2 case. Peripheral lung cancer may display A type. Except 2 cases D type lung tuberculoma, we compared curve data of 8 cases benign nodules （including tuberculoma Atype and periphery ring Atype, harmatoma Atype and C type） and lung cancer. SlEP%： benign nodules 0.7885 ±0.5543, lung cancer 1.2623 ±0.3059, P 〈 0.05; MER： benign nodules 1.0007 ± 0.4251, lung cancer 1.3694 ±0.2740, P 〈 0.05; washout： P 〉 0.05. Conclusion： Lung MR imaging is helpful to diagnosis and differential diagnosis of isolated benign and malignant nodules. SIEP% and MER could offer valuable information. The evolution of global tuberculosis may be from A type to ring-shaped ennoblement to D type. It was easy to do right diagnosis to lung tuberculoma with ring-shaped ennoblement and D type. Peripheral lung cancer commonly displayed A type and needed identification with acute inflammation. So, it is important to anti-inflammatory follow-up for a few A type nodules.展开更多
Linguistic dynamic systems(LDS)are dynamic processes involving computing with words(CW)for modeling and analysis of complex systems.In this paper,a fuzzy neural network(FNN)structure of LDS was proposed.In addition,an...Linguistic dynamic systems(LDS)are dynamic processes involving computing with words(CW)for modeling and analysis of complex systems.In this paper,a fuzzy neural network(FNN)structure of LDS was proposed.In addition,an improved nonlinear particle swarm optimization was employed for training FNN.The experiment results on logistics formulation demonstrates the feasibility and the efficiency of this FNN model.展开更多
文摘Objective： The aim of our study was to investigate the value of dynamic contrast-enhanced MRI for evaluating differential diagnosis of pulmonary isolated lesions. Methods： Twenty-nine consecutive patients enrolled in this study, all of whom underwent DCE-MRI examinations and received a histologic and clinical diagnosis. Among these, lung tuberculoma 7 cases, harmatoma 3 cases, peripheral lung cancer 19 cases. DCE-MRI was acquired with 3D LAVA technique, total 18 phases were acquired, scanner time of per phase was 5-7″. After contrasting agent, twice successive scanning was acquired at 10″ and 50″. Then 1′30″, 2′, 2′30″, 3′, 3′30″, 4′, 5′, 6′, 7′, 8′, 9′, 10′, 11′, 12′ performed scanning. Region of interest was placed on the Maximum level in the tumors. According to Schaefer＇s standard, four types of time signal intensity curve （TIC） were classified, which were A, B, C and D. Compared the dynamic parameters between benign and malignant nodules. Results： Lung tuberculoma may display three curves： A type 1 case, ring-shaped enhancement 4 cases （periphery ring A type, central region D type）, D type 2 cases. Harmatoma may display three curves： A type 1 case, C type 2 case. Peripheral lung cancer may display A type. Except 2 cases D type lung tuberculoma, we compared curve data of 8 cases benign nodules （including tuberculoma Atype and periphery ring Atype, harmatoma Atype and C type） and lung cancer. SlEP%： benign nodules 0.7885 ±0.5543, lung cancer 1.2623 ±0.3059, P 〈 0.05; MER： benign nodules 1.0007 ± 0.4251, lung cancer 1.3694 ±0.2740, P 〈 0.05; washout： P 〉 0.05. Conclusion： Lung MR imaging is helpful to diagnosis and differential diagnosis of isolated benign and malignant nodules. SIEP% and MER could offer valuable information. The evolution of global tuberculosis may be from A type to ring-shaped ennoblement to D type. It was easy to do right diagnosis to lung tuberculoma with ring-shaped ennoblement and D type. Peripheral lung cancer commonly displayed A type and needed identification with acute inflammation. So, it is important to anti-inflammatory follow-up for a few A type nodules.
基金National Natural Science Foundation of China(No.60873179)Doctoral Program Foundation of Institutions of Higher Education of China(No.20090121110032)+3 种基金Shenzhen Science and Technology Research Foundations,China(No.JC200903180630A,No.ZYB200907110169A)Key Project of Institutes Serving for the Economic Zone on the Western Coast of the Tai wan Strait,ChinaNatural Science Foundation of Xiamen,China(No.3502Z2093018)Projects of Education Depart ment of Fujian Province of China(No.JK2009017,No.JK2010031,No.JA10196)
文摘Linguistic dynamic systems(LDS)are dynamic processes involving computing with words(CW)for modeling and analysis of complex systems.In this paper,a fuzzy neural network(FNN)structure of LDS was proposed.In addition,an improved nonlinear particle swarm optimization was employed for training FNN.The experiment results on logistics formulation demonstrates the feasibility and the efficiency of this FNN model.
