A major obstacle when developing convolutional neural networks (CNNs) for medical imaging is the acquisition of training labels: Most current approaches rely on manual class labels from physicians, which may be challenging to obtain. Clinical biomarkers, often measured alongside medical images and used in diagnostic workup, may provide a rich set of data that can be collected retrospectively and utilized to train diagnostic models. In this work, we focused on assessing the potential of blood serum biomarkers, B-type natriuretic peptide (BNP) and NT-pro B-type natriuretic peptide (BNPP), indicative of acute heart failure (HF) and cardiogenic pulmonary edema to be used as continuously valued labels for training a radiographic deep learning algorithm. For this purpose, a CNN was trained using 27748 radiographs to automatically infer BNP and BNPP, and achieved strong performance (AUC =0.903, sensitivity =0.926, specificity =0.857, $r=0.787$ ). Also, the trained models achieved strong performance (AUC =0.801) for pulmonary edema detection when evaluated with radiologist labels. Since relevant radiographic features visible to the CNN may vary greatly based on image resolution, we also assessed the impact of image resolution on model learning and performance, comparing CNNs trained at five image sizes ( $64\times 64$ to $1024\times 1024$ ). Increasing image resolutions had diminishing but positive gains in AUC. Perhaps more importantly, experiments using three activation mapping techniques (saliency, Grad-CAM, XRAI) revealed considerably increased attention in the lungs with larger image sizes. This result emphasizes the need to utilize radiographs near native resolution for optimal CNN performance, which may not be fully captured by summary metrics like AUC.

Background Cardiac MRI is limited by long acquisition times, yet faster acquisition of smaller-matrix images reduces spatial detail. Deep learning (DL) might enable both faster acquisition and higher spatial detail via super-resolution. Purpose To explore the feasibility of using DL to enhance spatial detail from small-matrix MRI acquisitions and evaluate its performance against that of conventional image upscaling methods. Materials and Methods Short-axis cine cardiac MRI examinations performed between January 2012 and December 2018 at one institution were retrospectively collected for algorithm development and testing. Convolutional neural networks (CNNs), a form of DL, were trained to perform super resolution in image space by using synthetically generated low-resolution data. There were 70%, 20%, and 10% of examinations allocated to training, validation, and test sets, respectively. CNNs were compared against bicubic interpolation and Fourier-based zero padding by calculating the structural similarity index (SSIM) between high-resolution ground truth and each upscaling method. Means and standard deviations of the SSIM were reported, and statistical significance was determined by using the Wilcoxon signed-rank test. For evaluation of clinical performance, left ventricular volumes were measured, and statistical significance was determined by using the paired Student t test. Results For CNN training and retrospective analysis, 400 MRI scans from 367 patients (mean age, 48 years ± 18; 214 men) were included. All CNNs outperformed zero padding and bicubic interpolation at upsampling factors from two to 64 (P < .001). CNNs outperformed zero padding on more than 99.2% of slices (9828 of 9907). In addition, 10 patients (mean age, 51 years ± 22; seven men) were prospectively recruited for super-resolution MRI. Super-resolved low-resolution images yielded left ventricular volumes comparable to those from full-resolution images (P > .05), and super-resolved full-resolution images appeared to further enhance anatomic detail. Conclusion Deep learning outperformed conventional upscaling methods and recovered high-frequency spatial information. Although training was performed only on short-axis cardiac MRI examinations, the proposed strategy appeared to improve quality in other imaging planes. © RSNA, 2020 Online supplemental material is available for this article.

Purpose: Pneumonia is a common clinical diagnosis for which chest radiographs are often an important part of the diagnostic workup. Deep learning has the potential to expedite and improve the clinical interpretation of chest radiographs. While earlier approaches have emphasized the feasibility of "binary classification" to accomplish this task, alternative strategies may be possible. We explore the feasibility of a "semantic segmentation" deep learning approach to highlight the potential foci of pneumonia on frontal chest radiographs. Materials and methods: In this retrospective study, we trained a U-net convolutional neural network (CNN) to predict pixel-wise probability maps for pneumonia using a public data set provided by the Radiological Society of North America (RSNA) comprised of 22,000 radiographs and radiologist-defined bounding boxes. We reserved 3684 radiographs as an independent validation data set and assessed overall performance for localization using Dice overlap and classification performance using the area under the receiver-operator characteristic curve. Results: For classification/detection of pneumonia, area under the receiver-operator characteristic curve on frontal radiographs was 0.854 with a sensitivity of 82.8% and specificity of 72.6%. Using this strategy of neural network training, probability maps localized pneumonia to lung parenchyma for essentially all validation cases. For segmentation of pneumonia for positive cases, predicted probability maps had a mean Dice score (±SD) of 0.603±0.204, and 60.0% of these had a Dice score >0.5. Conclusions: A "semantic segmentation" deep learning approach can provide a probabilistic map to assist in the diagnosis of pneumonia. In combination with the patient's history, clinical findings and other imaging, this strategy may help expedite and improve diagnosis.