RESUMO
The present study describes an auxiliary tool in the diagnosis of left ventricular (LV) segmental wall motion (WM) abnormalities based on color-coded echocardiographic WM images. An artificial neural network (ANN) was developed and validated for grading LV segmental WM using data from color kinesis (CK) images, a technique developed to display the timing and magnitude of global and regional WM in real time. We evaluated 21 normal subjects and 20 patients with LVWM abnormalities revealed by two-dimensional echocardiography. CK images were obtained in two sets of viewing planes. A method was developed to analyze CK images, providing quantitation of fractional area change in each of the 16 LV segments. Two experienced observers analyzed LVWM from two-dimensional images and scored them as: 1) normal, 2) mild hypokinesia, 3) moderate hypokinesia, 4) severe hypokinesia, 5) akinesia, and 6) dyskinesia. Based on expert analysis of 10 normal subjects and 10 patients, we trained a multilayer perceptron ANN using a back-propagation algorithm to provide automated grading of LVWM, and this ANN was then tested in the remaining subjects. Excellent concordance between expert and ANN analysis was shown by ROC curve analysis, with measured area under the curve of 0.975. An excellent correlation was also obtained for global LV segmental WM index by expert and ANN analysis (R2 = 0.99). In conclusion, ANN showed high accuracy for automated semi-quantitative grading of WM based on CK images. This technique can be an important aid, improving diagnostic accuracy and reducing inter-observer variability in scoring segmental LVWM.
Assuntos
Ecocardiografia/métodos , Hipertrofia Ventricular Esquerda/diagnóstico por imagem , Interpretação de Imagem Assistida por Computador/métodos , Redes Neurais de Computação , Disfunção Ventricular Esquerda/diagnóstico por imagem , Adulto , Idoso , Estudos de Casos e Controles , Feminino , Humanos , Modelos Lineares , Masculino , Pessoa de Meia-Idade , Curva ROC , Índice de Gravidade de DoençaRESUMO
The present study describes an auxiliary tool in the diagnosis of left ventricular (LV) segmental wall motion (WM) abnormalities based on color-coded echocardiographic WM images. An artificial neural network (ANN) was developed and validated for grading LV segmental WM using data from color kinesis (CK) images, a technique developed to display the timing and magnitude of global and regional WM in real time. We evaluated 21 normal subjects and 20 patients with LVWM abnormalities revealed by two-dimensional echocardiography. CK images were obtained in two sets of viewing planes. A method was developed to analyze CK images, providing quantitation of fractional area change in each of the 16 LV segments. Two experienced observers analyzed LVWM from two-dimensional images and scored them as: 1) normal, 2) mild hypokinesia, 3) moderate hypokinesia, 4) severe hypokinesia, 5) akinesia, and 6) dyskinesia. Based on expert analysis of 10 normal subjects and 10 patients, we trained a multilayer perceptron ANN using a back-propagation algorithm to provide automated grading of LVWM, and this ANN was then tested in the remaining subjects. Excellent concordance between expert and ANN analysis was shown by ROC curve analysis, with measured area under the curve of 0.975. An excellent correlation was also obtained for global LV segmental WM index by expert and ANN analysis (R² = 0.99). In conclusion, ANN showed high accuracy for automated semi-quantitative grading of WM based on CK images. This technique can be an important aid, improving diagnostic accuracy and reducing inter-observer variability in scoring segmental LVWM.
Assuntos
Humanos , Masculino , Feminino , Adulto , Pessoa de Meia-Idade , Ecocardiografia/métodos , Hipertrofia Ventricular Esquerda , Interpretação de Imagem Assistida por Computador/métodos , Redes Neurais de Computação , Disfunção Ventricular Esquerda , Estudos de Casos e Controles , Modelos Lineares , Curva ROC , Índice de Gravidade de DoençaRESUMO
OBJECTIVE: This study used image processing techniques to quantify the upper eyelid contour of patients with Graves upper eyelid retraction and congenital blepharoptosis. DESIGN: The study design was a cross-sectional study. PARTICIPANTS: A total of 29 patients with Graves disease, 22 patients with congenital blepharoptosis, and 50 patients with no history of eye disease participated. INTERVENTION: The images of the palpebral fissure of all participants were transferred to a personal computer and processed with NIH Image 1.55 software. MAIN OUTCOME MEASURES: The following parameters were analyzed: the curvature of the upper eyelid contour, the position of the contour peak relative to the midline, and the ratio between the temporal and nasal upper quadrant areas of the palpebral fissure. RESULTS: All upper eyelid contours could be fitted with second-degree polynomial functions. The mean temporal/nasal area ratio was 1.33 mm in patients with Graves disease, 0.92 mm in patients with blepharoptosis, and 1.04 mm in control subjects. The peak of the upper eyelid contour was found to be lateral to the midline in control subjects (1.05 mm) and in patients with Graves disease (2.09 mm). In patients with blepharoptosis, the peak was 0.69 mm medial to the midline. Overall, the distance between the midpupil and the upper eyelid margin was correlated with several factors: the degree of curvature, the position of the peak of the eyelid contour, and the temporal/nasal area ratio. CONCLUSIONS: In Graves eyelid retraction, the curvature of the upper eyelid is enhanced, the peak of the contour is displaced laterally, and the temporal upper quadrant area is increased. Conversely, in congenital blepharoptosis, the eyelid is almost flat, the peak of the contour is displaced medially, and the upper quadrant area is diminished. The lateral segment of the upper eyelid is more involved than the nasal segment in both Graves upper eyelid retraction and congenital blepharoptosis.