查看更多>>摘要：Objective The purpose of this study was to develop an artificial intelligence (AI)-based model to detect features of atrial fibrillation (AF) on chest radiographs. Methods This retrospective study included consecutively collected chest radiographs of patients who had echocardiography at our institution from July 2016 to May 2019. Eligible radiographs had been acquired within 30 days of the echocardiography. These radiographs were labeled as AF-positive or AF-negative based on the associated electronic medical records; then, each patient was randomly divided into training, validation, and test datasets in an 8:1:1 ratio. A deep learning-based model to classify radiographs as with or without AF was trained on the training dataset, tuned with the validation dataset, and evaluated with the test dataset. Results The training dataset included 11,105 images (5637 patients; 3145 male, mean age +/- standard deviation, 68 +/- 14 years), the validation dataset included 1388 images (704 patients, 397 male, 67 +/- 14 years), and the test dataset included 1375 images (706 patients, 395 male, 68 +/- 15 years). Applying the model to the validation and test datasets gave a respective area under the curve of 0.81 (95% confidence interval, 0.78-0.85) and 0.80 (0.76-0.84), sensitivity of 0.76 (0.70-0.81) and 0.70 (0.64-0.76), specificity of 0.75 (0.72-0.77) and 0.74 (0.72-0.77), and accuracy of 0.75 (0.72-0.77) and 0.74 (0.71-0.76). Conclusion Our AI can identify AF on chest radiographs, which provides a new way for radiologists to infer AF.
Kapsner, Lorenz A.Ohlmeyer, SabineFolle, LukasLaun, Frederik B....
11页
查看更多>>摘要：Objectives To automatically detect MRI artifacts on dynamic contrast-enhanced (DCE) maximum intensity projections (MIPs) of the breast using deep learning. Methods Women who underwent clinically indicated breast MRI between October 2015 and December 2019 were included in this IRB-approved retrospective study. We employed two convolutional neural network architectures (ResNet and DenseNet) to detect the presence of artifacts on DCE MIPs of the left and right breasts. Networks were trained on images acquired up to and including the year 2018 using a 5-fold cross-validation (CV). Ensemble classifiers were built with the resulting CV models and applied to an independent holdout test dataset, which was formed by images acquired in 2019. Results Our study sample contained 2265 examinations from 1794 patients (median age at first acquisition: 50 years [IQR: 17 years]), corresponding to 1827 examinations of 1378 individuals in the training dataset and 438 examinations of 416 individuals in the holdout test dataset with a prevalence of image-level artifacts of 53% (1951/3654 images) and 43% (381/876 images), respectively. On the holdout test dataset, the ResNet and DenseNet ensembles demonstrated an area under the ROC curve of 0.92 and 0.94, respectively. Conclusion Neural networks are able to reliably detect artifacts that may impede the diagnostic assessment of MIPs derived from DCE subtraction series in breast MRI. Future studies need to further explore the potential of such neural networks to complement quality assurance and improve the application of DCE MIPs in a clinical setting, such as abbreviated protocols.
查看更多>>摘要：Objectives Coronary computed tomography angiography (CCTA) has rapidly developed in the coronary artery disease (CAD) field. However, manual coronary artery tree segmentation and reconstruction are time-consuming and tedious. Deep learning algorithms have been successfully developed for medical image analysis to process extensive data. Thus, we aimed to develop a deep learning tool for automatic coronary artery reconstruction and an automated CAD diagnosis model based on a large, single-centre retrospective CCTA cohort. Methods Automatic CAD diagnosis consists of two subtasks. One is a segmentation task, which aims to extract the region of interest (ROI) from original images with U-Net. The second task is an identification task, which we implemented using 3DNet. The coronary artery tree images and clinical parameters were input into 3DNet, and the CAD diagnosis result was output. Results We built a coronary artery segmentation model based on CCTA images with the corresponding labelling. The segmentation model had a mean Dice value of 0.771 +/- 0.021. Based on this model, we built an automated diagnosis model (classification model) for CAD. The average accuracy and area under the receiver operating characteristic curve (AUC) were 0.750 +/- 0.056 and 0.737, respectively. Conclusion Herein, using a deep learning algorithm, we realized the rapid classification and diagnosis of CAD from CCTA images in two steps. Our deep learning model can automatically segment the coronary artery quickly and accurately and can deliver a diagnosis of >= 50% coronary artery stenosis. Artificial intelligence methods such as deep learning have the potential to elevate the efficiency in CCTA image analysis considerably.
NSTL
Springer Nature