Discriminating lung nodules as malignant or benign is still an underlying challenge. To address this challenge, radiologists need computer aided diagnosis (CAD) systems which can assist in learning discriminative imaging features corresponding to malignant and benign nodules. However, learning highly discriminative imaging features is an open problem. In this paper, our aim is to learn the most discriminative features pertaining to lung nodules by using an adversarial learning methodology. Specifically, we propose to use unsupervised learning with Deep Convolutional-Generative Adversarial Networks (DC-GANs) to generate lung nodule samples realistically. We hypothesize that imaging features of lung nodules will be discriminative if it is hard to differentiate them (fake) from real (true) nodules. To test this hypothesis, we present Visual Turing tests to two radiologists in order to evaluate the quality of the generated (fake) nodules. Extensive comparisons are performed in discerning real, generated, benign, and malignant nodules. This experimental set up allows us to validate the overall quality of the generated nodules, which can then be used to (1) improve diagnostic decisions by mining highly discriminative imaging features, (2) train radiologists for educational purposes, and (3) generate realistic samples to train deep networks with big data.
Deep learning has demonstrated tremendous revolutionary changes in the computing industry and its effects in radiology and imaging sciences have begun to dramatically change screening paradigms. Specifically, these advances have influenced the development of computer-aided detection and diagnosis (CAD) systems. These technologies have long been thought of as "second-opinion" tools for radiologists and clinicians. However, with significant improvements in deep neural networks, the diagnostic capabilities of learning algorithms are approaching levels of human expertise (radiologists, clinicians etc.), shifting the CAD paradigm from a "second opinion" tool to a more collaborative utility. This paper reviews recently developed CAD systems based on deep learning technologies for breast cancer diagnosis, explains their superiorities with respect to previously established systems, defines the methodologies behind the improved achievements including algorithmic developments, and describes remaining challenges in breast cancer screening and diagnosis. We also discuss possible future directions for new CAD models that continue to change as artificial intelligence algorithms evolve.
Myocardial fibrosis is a common endpoint in a variety of cardiac diseases and a major independent predictor of adverse cardiac outcomes. Short of histopathologic analysis, which is limited by sampling bias, most diagnostic modalities are limited in their depiction of myocardial fibrosis. Cardiac magnetic resonance (MR) imaging has the advantage of providing detailed soft-tissue characterization, and a variety of novel quantification methods have further improved its usefulness. Contrast material-enhanced cardiac MR imaging depends on differences in signal intensity between regions of scarring and adjacent normal myocardium. Diffuse myocardial fibrosis lacks these differences in signal intensity. Measurement of myocardial T1 times (T1 mapping) with gadolinium-enhanced inversion recovery-prepared sequences may depict diffuse myocardial fibrosis and has good correlation with ex vivo fibrosis content. T1 mapping calculates myocardial T1 relaxation times with image-based signal intensities and may be performed with standard cardiac MR imagers and radiologic workstations. Myocardium with diffuse fibrosis has greater retention of contrast material, resulting in T1 times that are shorter than those in normal myocardium. Early studies have suggested that diffuse myocardial fibrosis may be distinguished from normal myocardium with T1 mapping. Large multicenter studies are needed to define the role of T1 mapping in developing prognoses and therapeutic assessments. However, given its strengths as a noninvasive method for direct quantification of myocardial fibrosis, T1 mapping may eventually play an important role in the management of cardiac disease.
Anatomical and biophysical modeling of left atrium (LA) and proximal pulmonary veins (PPVs) is important for clinical management of several cardiac diseases. Magnetic resonance imaging (MRI) allows qualitative assessment of LA and PPVs through visualization. However, there is a strong need for an advanced image segmentation method to be applied to cardiac MRI for quantitative analysis of LA and PPVs. In this study, we address this unmet clinical need by exploring a new deep learning-based segmentation strategy for quantification of LA and PPVs with high accuracy and heightened efficiency. Our approach is based on a multi-view convolutional neural network (CNN) with an adaptive fusion strategy and a new loss function that allows fast and more accurate convergence of the backpropagation based optimization. After training our network from scratch by using more than 60K 2D MRI images (slices), we have evaluated our segmentation strategy to the STACOM 2013 cardiac segmentation challenge benchmark. Qualitative and quantitative evaluations, obtained from the segmentation challenge, indicate that the proposed method achieved the state-of-the-art sensitivity (90%), specificity (99%), precision (94%), and efficiency levels (10 seconds in GPU, and 7.5 minutes in CPU).
Non-invasive detection of cardiovascular disorders from radiology scans requires quantitative image analysis of the heart and its substructures. There are well-established measurements that radiologists use for diseases assessment such as ejection fraction, volume of four chambers, and myocardium mass. These measurements are derived as outcomes of precise segmentation of the heart and its substructures. The aim of this paper is to provide such measurements through an accurate image segmentation algorithm that automatically delineates seven substructures of the heart from MRI and/or CT scans. Our proposed method is based on multi-planar deep convolutional neural networks (CNN) with an adaptive fusion strategy where we automatically utilize complementary information from different planes of the 3D scans for improved delineations. For CT and MRI, we have separately designed three CNNs (the same architectural configuration) for three planes, and have trained the networks from scratch for voxel-wise labeling for the following cardiac structures: myocardium of left ventricle (Myo), left atrium (LA), left ventricle (LV), right atrium (RA), right ventricle (RV), ascending aorta (Ao), and main pulmonary artery (PA). We have evaluated the proposed method with 4-fold-cross-validation on the multi-modality whole heart segmentation challenge (MM-WHS 2017) dataset. The precision and dice index of 0.93 and 0.90, and 0.87 and 0.85 were achieved for CT and MR images, respectively. While a CT volume was segmented about 50 seconds, an MRI scan was segmented around 17 seconds with the GPUs/CUDA implementation.
Objectives To identify the incremental value and optimal role of Cardiac Magnetic Resonance (CMR) imaging in arrhythmic risk stratification of Arrhythmogenic Right Ventricular Dysplasia/Cardiomyopathy (ARVD/C) associated desmosomal mutation carriers without a prior history of sustained ventricular arrhythmia. Background Risk stratification of ARVD/C mutation carriers is challenging. Methods We included 69 patients (age 27.0 ± 15.3 years, 42% male) harboring ARVD/C associated pathogenic mutations (83% PKP-2) without prior sustained ventricular arrhythmias. Electrocardiography (ECG) and 24-hours Holter monitoring closest to presentation were analyzed for electrical abnormalities as per revised Task Force Criteria. CMR studies were done to identify abnormal cardiac structure and function according to the revised Task Force Criteria. Results Overall, 42 (61%) patients presented with electrical abnormalities based on their ECG and Holter monitor, of whom 20 (48%) had an abnormal CMR. Only 1 (4%) of 27 patients without electrical abnormalities at initial evaluation had an abnormal CMR. Over a mean follow-up of 5.8 ± 4.4 years, 11 (16%) patients experienced a sustained ventricular arrhythmia, exclusively in patients with both electrical abnormalities (ECG and/or Holter) and abnormal CMR. Conclusion Our results suggest that electrical abnormalities on ECG and Holter monitoring precede detectable structural abnormalities in ARVD/C mutation carriers. Therefore, evaluation of cardiac structure and function using CMR is probably not necessary in the absence of baseline electrical abnormalities. Among ARVD/C mutation carriers, the presence of both electrical and CMR abnormalities identifies patients at high risk of events and thus patients who might benefit from prophylactic implantable cardioverter-defibrillator implantation.
Background: With the widespread use of cardiac multidetector row computed tomography (MDCT), the issue of incidental findings is receiving increasing attention. Our objectives were to evaluate the prevalence of incidental findings discovered during cardiac MDCT scanning and to identify clinical variables associated with incidental findings. Methods: This cross-sectional analysis involved a population-based sample recruited from an integrated health care delivery system in Northern California as part of the Atherosclerotic Disease, Vascular Function and Genetic Epidemiology (ADVANCE) Study. Healthy men and women aged 60 to 69 years without diagnosed cardiovascular disease underwent cardiac MDCT for the detection and quantification of coronary artery calcification. The images were prospectively evaluated for incidental findings. Results: A total of 459 participants underwent MDCT scanning, and the overall prevalence of any incidental finding was 41%. Of the 459 participants, 105 (23%) had at least 1 incidental finding that was recommended for clinical or radiological follow-up examination, the most common of which was single or multiple pulmonary nodules (18%). Participants with and without incidental findings had comparable baseline demographics and selected clinical variables, although there were significantly fewer men and a significantly lower prevalence of the metabolic syndrome in those with incidental findings. Conclusions: Incidental findings, especially pulmonary nodules, are common in cardiac MDCT performed to assess coronary artery calcification in older healthy adults. The net risks and benefits of looking for noncardiac abnormalities during cardiac MDCT should be rigorously evaluated.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.