Automatic Prediction of Atrial Fibrillation Based on Convolutional Neural Network Using a Short-term Normal Electrocardiogram Signal

BACKGROUND: In this study, we propose a method for automatically predicting atrial fibrillation (AF) based on convolutional neural network (CNN) using a short-term normal electrocardiogram (ECG) signal. METHODS: We designed a CNN model and optimized it by dropout and normalization. One-dimensional convolution, max-pooling, and fully-connected multiple perceptron were used to analyze the short-term normal ECG. The ECG signal was preprocessed and segmented to train and evaluate the proposed CNN model. The training and test sets consisted of the two AF and one normal dataset from the MIT-BIH database. RESULTS: The proposed CNN model for the automatic prediction of AF achieved a high performance with a sensitivity of 98.6%, a specificity of 98.7%, and an accuracy of 98.7%. CONCLUSION: The results show the possibility of automatically predicting AF based on the CNN model using a short-term normal ECG signal. The proposed CNN model for the automatic prediction of AF can be a helpful tool for the early diagnosis of AF in healthcare fields.

Clinical impact of fat clearing technique in nodal staging of rectal cancer after preoperative chemoradiotherapy / 대한외과학회지

PURPOSE: This study was designed to evaluate the efficacy of a fat clearing technique for accurate nodal staging of rectal cancer patients after preoperative chemoradiotherapy (CRT). METHODS: A total of 19 patients with rectal cancer within 10 cm from anal verge were divided into two groups: non-CRT group (n = 10) and CRT group (n = 9). For pathologic assessment, lymph node (LN) harvest was performed using conventional manual dissection followed by a fat clearing technique. RESULTS: A median of 3.0 additional LNs in non-CRT group and 3.8 LNs in CRT group were identified by the fat clearing technique. When subanalysis was performed in patients with fewer than 12 retrieved LNs, a median of 4.0 extra LNs in non-CRT group and 3.5 extra LNs in CRT group were identified after the fat clearing technique. None of additionally identified nodes were metastatic. In both groups, the median size of retrieved LNs following the fat clearing technique was smaller than that obtained by manual dissection (2.0 mm vs. 3.0 mm, P < 0.001). CONCLUSION: The fat clearing technique allowed detection of additional LNs that were missed by the manual method, but these detected LNs were not proven to be metastatic.