RESUMEN
OBJECTIVES: The Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is a promising tool for the evaluation of stroke expansion to determine suitability for reperfusion therapy. The aim of this study was to validate deep learning-based ASPECTS calculation software that utilizes a three-dimensional fully convolutional network-based brain hemisphere comparison algorithm (3D-BHCA). MATERIALS AND METHODS: We retrospectively collected head non-contrast computed tomography (CT) data from 71 patients with acute ischemic stroke and 80 non-stroke patients. The results for ASPECTS on CT assessed by 5 stroke neurologists and by the 3D-BHCA model were compared with the ground truth by means of region-based and score-based analyses. RESULTS: In total, 151 patients and 3020 (151 × 20) ASPECTS regions were investigated. Median time from onset to CT was 195 min in the stroke patients. In region-based analysis, the sensitivity (0.80), specificity (0.97), and accuracy (0.96) of the 3D-BHCA model were superior to those of stroke neurologists. The sensitivity (0.98), specificity (0.92), and accuracy (0.97) of dichotomized ASPECTS > 5 analysis and the intraclass correlation coefficient (0.90) in total score-based analysis of the 3D-BHCA model were superior to those of stroke neurologists overall. When patients with stroke were stratified by onset-to-CT time, the 3D-BHCA model exhibited the highest performance to calculate ASPECTS, even in the earliest time period. CONCLUSIONS: The automated ASPECTS calculation software we developed using a deep learning-based algorithm was superior or equal to stroke neurologists in performing ASPECTS calculation in patients with acute stroke and non-stroke patients.
Asunto(s)
Aprendizaje Profundo , Interpretación de Imagen Radiográfica Asistida por Computador , Accidente Cerebrovascular/diagnóstico por imagen , Tomografía Computarizada por Rayos X , Anciano , Anciano de 80 o más Años , Toma de Decisiones Clínicas , Femenino , Humanos , Imagenología Tridimensional , Masculino , Selección de Paciente , Valor Predictivo de las Pruebas , Pronóstico , Reproducibilidad de los Resultados , Estudios Retrospectivos , Accidente Cerebrovascular/fisiopatología , Accidente Cerebrovascular/terapia , Trombectomía , Terapia TrombolíticaRESUMEN
A 52-year-old man was admitted with a brain abscess in the left basal ganglia. He had a paradoxical brain embolic mechanism owing to a pulmonary arteriovenous fistula (PAVF) and was diagnosed as having a right-to-left shunt by transesophageal echocardiography (TEE) and transcranial color Doppler (TCD) with saline contrast medium. We determined that the brain abscess was caused by the PAVFs without Rendu-Osler-Weber disease. TEE and TCD with saline contrast medium were very useful for identifying the presence of the right-to-left shunt that caused the brain abscess.
Asunto(s)
Fístula Arteriovenosa/complicaciones , Absceso Encefálico/etiología , Arteria Pulmonar/anomalías , Venas Pulmonares/anomalías , Telangiectasia Hemorrágica Hereditaria/diagnóstico , Angiografía , Fístula Arteriovenosa/diagnóstico , Absceso Encefálico/diagnóstico , Diagnóstico Diferencial , Ecocardiografía Transesofágica , Estudios de Seguimiento , Humanos , Masculino , Persona de Mediana Edad , Ultrasonografía Doppler TranscranealRESUMEN
OBJECTIVE: The right-to-left shunt (RLS) is diagnosed by contrast-enhanced transesophageal echocardiographic monitoring of the bilateral atria (cTEE-BA). However, the procedure is often disturbed by nonsmoke spontaneous individual contrast (NSSIC) with fast motion, which appears in the left atrium after respiratory maneuvers without administration of a contrast medium and moves past in several seconds. We attempted to perform cTEE monitoring of the aortic arch (cTEE-AA) for evaluation of the RLS and compared the findings with those of cTEE-BA. METHODS: Both cTEE-BA and cTEE-AA were performed in 168 patients with ischemic stroke (133 men and 35 women; mean age +/- SD, 62.0 +/- 14.4 years). The frequency of NSSIC in the left atrium was compared with that in the aortic arch during the respiratory maneuver. When contrast much brighter than the NSSIC was visualized in the left atrium and the aortic arch during the respiratory maneuver with administration of the contrast medium, we considered the RLS to be positive in the cTEE-BA and cTEE-AA, respectively. Findings were then compared between the 2 examinations. RESULTS: Nonsmoke spontaneous individual contrast was more frequently observed in the left atrium than the aortic arch (61.3% versus 14.9%; chi2 test, P < .0001). The RLS was positive in 34 patients in the cTEE-BA and in 39 patients in the cTEE-AA. The sensitivity and specificity of the cTEE-AA for the cTEE-BA were 100% and 96.3%, respectively. CONCLUSIONS: The cTEE-AA may be an alternative method for detection of an RLS, especially in patients with a large amount of NSSIC in the left atrium.