RESUMO
PURPOSE: To propose an automatic approach based on a convolutional neural network (CNN) to evaluate the quality of T2-weighted liver magnetic resonance (MR) images as nondiagnostic (ND) or diagnostic (D). MATERIALS AND METHODS: We included 150 T2-weighted liver MR imaging examinations in this retrospective study. Each slice of liver image was annotated with a label D or ND by two radiologists with seven and six years of experience, respectively. Additionally, the radiologists segmented the liver region manually as the ground truth for liver segmentation. A CNN was trained to segment the liver region and another CNN was used to classify the qualities of patches extracted from the liver region. The quality of an image was obtained from the percentage of nondiagnostic patches in all liver patches in the image. Treating nondiagnostic as positive, the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), area under the receiver operating characteristic curve (AUC), and confusion matrix were used to evaluate our model. A Mann-Whitney U test was performed with the statistical significance set at 0.05. RESULTS: Our model achieved good performance with an accuracy of 88.3 %, sensitivity of 86.0 %, specificity of 89.4 %, PPV of 78.6 %, NPV of 93.4 %, and AUC of 0.911 (95 % confidence interval: 0.882-0.939, p < 0.05). The confusion matrix of our model indicated good concordance with that of the radiologists. CONCLUSIONS: The proposed two-step patch-based model achieved excellent performance when assessing the quality of liver MR images.
Assuntos
Interpretação de Imagem Assistida por Computador/métodos , Hepatopatias/diagnóstico por imagem , Imageamento por Ressonância Magnética/métodos , Redes Neurais de Computação , Humanos , Fígado/diagnóstico por imagem , Curva ROC , Reprodutibilidade dos Testes , Estudos Retrospectivos , Sensibilidade e EspecificidadeRESUMO
Schizotypy is associated with anhedonia. However, previous findings on the neural substrates of anhedonia in schizotypy are mixed. In the present study, we measured the neural substrates associated with reward anticipation and consummation in positive and negative schizotypy using functional MRI. The Monetary Incentive Delay task was administered to 33 individuals with schizotypy (18 positive schizotypy (PS),15 negative schizotypy (NS)) and 22 healthy controls. Comparison between schizotypy individuals and controls were performed using two-sample T tests for contrast images involving gain versus non-gain anticipation condition and gain versus non-gain consummation condition. Multiple comparisons were corrected using Monte Carlo Simulation correction of p<.05. The results showed no significant difference in brain activity between controls and schizotypy individuals as a whole during gain anticipation or consummation. However, during the consummatory phase, NS individuals rather than PS individuals showed diminished left amygdala and left putamen activity compared with controls. We observed significantly weaker activation at the left ventral striatum during gain anticipation in NS individuals compared with controls. PS individuals, however, exhibited enhanced right ventral lateral prefrontal activity. These findings suggest that different dimensions of schizotypy may be underlied by different neural dysfunctions in reward anticipation and consummation.