RESUMEN
The ability to differentiate active from chronic valvular vegetations (VEGs) by digital image processing and by visual observation was evaluated in 18 patients with a clinical diagnosis of infective endocarditis (IE). Two-dimensional echocardiographic (2-DE) examinations were performed on all patients at diagnosis and after a mean period of 52 days. Two comparable images (active and chronic) from the same patient and in the same phase of the cardiac cycle were digitized, magnified, and displayed on a high resolution monitor. The mean pixel intensity (MPI) was 72+/-14 in the active stage and 143 +/-23 in the chronic stage (P<0.0001). The VEG size was 0.64+/- 0.15 cm(2) in the active stage and decreased to 0.46+/-0.17 cm(2) in the chronic stage (P<0.001). Two experienced echocardiographers, who were blinded to the age of the VEGs, identified each echocardiographic image as active or chronic based on visual observation of density of the VEGs. The VEGs were correctly identified as active or chronic in 17 out of the 18 patients. In summary, although digital image processing of 2-DE may be useful, the density of VEGs assessed by visual inspection will help differentiate between active and chronic VEGs of IE. The standardization procedure at the time of the initial study and use of identical gain settings in subsequent studies are key factors in making this distinction.
Asunto(s)
Ecocardiografía , Endocarditis Bacteriana/diagnóstico por imagen , Válvulas Cardíacas/diagnóstico por imagen , Enfermedad Aguda , Adulto , Enfermedad Crónica , Diagnóstico Diferencial , Endocarditis Bacteriana/complicaciones , Femenino , Enfermedades de las Válvulas Cardíacas/diagnóstico por imagen , Enfermedades de las Válvulas Cardíacas/etiología , Humanos , Procesamiento de Imagen Asistido por Computador , Masculino , Persona de Mediana Edad , Variaciones Dependientes del Observador , Estudios Prospectivos , Grabación en VideoRESUMEN
To evaluate the ability of a system developed in our laboratory to differentiate between acute and healed myocardial infarction (MI), gated 2-dimensional echocardiography was performed in 10 patients with recent MI (within 48 hours) and 10 patients with healed MI (later than 4 weeks). The 2-dimensional echocardiographic images were digitized using a Datacube VG-120 videoframe digitizer and each digitized videoframe (320 X 240 matrix) was transmitted using a high-speed serial data link to a second computer and stored on floppy disc. Five gated video frames of each patient were time-averaged to give a smoothed digitized image. This image was displayed on high-resolution color monitor connected to a color graphic subsystem. Sixty-four colors indicated pixel intensity. The exact pixel value at any given location was determined using a high-resolution light pen. Color or pixel values were not significantly different between the area of acute MI (pixel intensity 23 +/- 3, mean +/- standard deviation) identified by a regional wall motion abnormality, and the adjacent normal muscle (23 +/- 4). In each patient with healed MI, an increase in color intensity and pixel value (43 +/- 6) was seen in the area of MI (i.e., area of regional wall motion abnormality) compared with adjacent normal muscle (23 +/- 2) (p less than 0.001). These preliminary data in selected patients indicate that this technique enables differentiation between acute and healed MI. It could be of value in management of patients with suspected MI.