Search NASA⌕ Search

SEARCH · Search NASA

Results for “CHEST”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Domain Shift Analysis in Chest Radiographs Classification in a Veterans Healthcare Administration Population

This study aims to assess the impact of domain shift on chest X-ray classification accuracy and to analyze the influence of ground truth label quality and demographic factors such as age group, sex, and study year. We used a DenseNet121 model pre-trained MIMIC-CXR dataset for deep learning-based multi-label classification using ground truth labels from radiology reports extracted using the CheXpert and CheXbert Labeler. We compared the performance of the 14 chest X-ray labels on the MIMIC-CXR and Veterans Healthcare Administration chest X-ray dataset (VA-CXR). The validation of ground truth and the assessment of multi-label classification performance across various NLP extraction tools revealed that the VA-CXR dataset exhibited lower disagreement rates than the MIMIC-CXR datasets. Additionally, there were notable differences in AUC scores between models utilizing CheXpert and CheXbert. When evaluating multi-label classification performance across different datasets, minimal domain shift was observed in the unseen VA dataset, except for the label “Enlarged Cardiomediastinum.” The subgroup with the most significant variations in multi-label classification performance was study year. These findings underscore the importance of considering domain shift in chest X-ray classification tasks, paying particular attention to the temporality of the exam. Our study reveals the significant impact of domain shift and demographic factors on chest X-ray classification, emphasizing the need for improved transfer learning and robust model development. Addressing these challenges is crucial for advancing medical imaging research and improving patient care.

chest X-ray image classification↗
DOE OSTI2021

Computer-aided Abnormality Detection in Chest Radiographs in a Clinical Setting via Domain-adaptation

Deep learning (DL) models are being deployed at medical centers to aid radiologists for diagnosis of lung conditions from chest radiographs. Such models are often trained on a large volume of publicly available labeled radiographs. These pre-trained DL models’ ability to generalize in clinical settings is poor because of the changes in data distributions between publicly available and privately held radiographs. In chest radiographs, the heterogeneity in distributions arises from the diverse conditions in X-ray equipment and their configurations used for generating the images. In the machine learning community, the challenges posed by the heterogeneity in the data generation source is known as domain shift, which is a mode shift in the generative model. In this work, we introduce a domain-shift detection and removal method to overcome this problem. Our experimental results show the proposed method’s effectiveness in deploying a pre-trained DL model for abnormality detection in chest radiographs in a clinical setting.

Dubey, Abhishek↗
DOE OSTI2021

Systematic Screening of COVID-19 Disease Based on Chest CT and RT-PCR for Cancer Patients Undergoing Radiation Therapy in a Coronavirus French Hotspot

Patients with cancer are presumed to be more vulnerable to COVID-19. We evaluated a screening strategy combining chest computed tomography (CT) and reverse-transcription polymerase chain reaction (RT-PCR) for patients treated with radiation therapy at our cancer center located in a COVID-19 French hotspot during the first wave of the pandemic.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2023

Effect of image resolution on automated classification of chest X-rays

Deep learning (DL) models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays (CXRs). Recently available public CXR datasets include high resolution images, but state-of-the-art models are trained on reduced size images due to limitations on graphics processing unit memory and training time. As computing hardware continues to advance, it has become feasible to train deep convolutional neural networks on high-resolution images without sacrificing detail by downscaling. This study examines the effect of increased resolution on CXR classification performance. We used the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution CXR images for this study. We applied image downscaling from native resolution to 2048 × 2048 pixels, 1024 × 1024 pixels, 512 × 512 pixels, and 256 × 256 pixels and then we used the DenseNet121 and EfficientNet-B4 DL models to evaluate clinical task performance using these four downscaled image resolutions. We find that while some clinical findings are more reliably labeled using high resolutions, many other findings are actually labeled better using downscaled inputs. We qualitatively verify that tasks requiring a large receptive field are better suited to downscaled low resolution input images, by inspecting effective receptive fields and class activation maps of trained models. Lastly, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, indicating that diverse information is extracted across resolutions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

New equations of state for saturated tuff based on the CHEST model

The CHemical Equilibrium for Saturated Tuff (CHEST) model was originally presented in Ref. 1, with numerical results depicted in Figure 1 and tabulated in Appendix B of that work. The results consisted of specific internal energy as a function of pressure and specific volume, E(P, V), spanning the domain P = 10 –4 – 10 4 kbar and V = 0.17 – 752 cm 3 /g (ρ = 1.33 × 10 –3 – 5.9 g/cm 3 ).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Validating deep learning inference during chest X-ray classification for COVID-19 screening

Abstract The new coronavirus unleashed a worldwide pandemic in early 2020, and a fatality rate several times that of the flu. As the number of infections soared, and capabilities for testing lagged behind, chest X-ray (CXR) imaging became more relevant in the early diagnosis and treatment planning for patients with suspected or confirmed COVID-19 infection. In a few weeks, proposed new methods for lung screening using deep learning rapidly appeared, while quality assurance discussions lagged behind. This paper proposes a set of protocols to validate deep learning algorithms, including our ROI Hide-and-Seek protocol, which emphasizes or hides key regions of interest from CXR data. Our protocol allows assessing the classification performance for anomaly detection and its correlation to radiological signatures, an important issue overlooked in several deep learning approaches proposed so far. By running a set of systematic tests over CXR representations using public image datasets, we demonstrate the weaknesses of current techniques and offer perspectives on the advantages and limitations of automated radiography analysis when using heterogeneous data sources.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Feature‐shared adaptive‐boost deep learning for invasiveness classification of pulmonary subsolid nodules in CT images

Purpose In clinical practice, invasiveness is an important reference indicator for differentiating the malignant degree of subsolid pulmonary nodules. These nodules can be classified as atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA), or invasive adenocarcinoma (IAC). The automatic determination of a nodule's invasiveness based on chest CT scans can guide treatment planning. However, it is challenging, owing to the insufficiency of training data and their interclass similarity and intraclass variation. To address these challenges, we propose a two‐stage deep learning strategy for this task: prior‐feature learning followed by adaptive‐boost deep learning. Methods The adaptive‐boost deep learning is proposed to train a strong classifier for invasiveness classification of subsolid nodules in chest CT images, using multiple 3D convolutional neural network (CNN)‐based weak classifiers. Because ensembles of multiple deep 3D CNN models have a huge number of parameters and require large computing resources along with more training and testing time, the prior‐feature learning is proposed to reduce the computations by sharing the CNN layers between all weak classifiers. Using this strategy, all weak classifiers can be integrated into a single network. Results Tenfold cross validation of binary classification was conducted on a total of 1357 nodules, including 765 noninvasive (AAH and AIS) and 592 invasive nodules (MIA and IAC). Ablation experimental results indicated that the proposed binary classifier achieved an accuracy of with an AUC of 81.3 . These results are superior compared to those achieved by three experienced chest imaging specialists who achieved an accuracy of , , and , respectively. About 200 additional nodules were also collected. These nodules covered 50 cases for each category (AAH, AIS, MIA, and IAC, respectively). Both binary and multiple classifications were performed on these data and the results demonstrated that the proposed method definitely achieves better performance than the performance achieved by nonensemble deep learning methods. Conclusions It can be concluded that the proposed adaptive‐boost deep learning can significantly improve the performance of invasiveness classification of pulmonary subsolid nodules in CT images, while the prior‐feature learning can significantly reduce the total size of deep models. The promising results on clinical data show that the trained models can be used as an effective lung cancer screening tool in hospitals. Moreover, the proposed strategy can be easily extended to other similar classification tasks in 3D medical images.

Wang, Jun↗
DOE OSTI2019

Outcome of Primary Desmoid Tumors at All Anatomic Locations Initially Managed with Active Surveillance

Background: The behavior of desmoid tumors is unpredictable and varies from spontaneous remission to symptomatic and radiologic progression. This study aimed to evaluate the radiologic and symptomatic course of the disease in patients initially managed with active surveillance. Methods: Patients with a primary desmoid tumor at any anatomic location diagnosed between 1998 and 2016 were identified in a prospectively maintained database from a single sarcoma reference center in the United Kingdom. Inverse univariate Cox proportional hazard regression analyses were conducted to evaluate the course of the disease and indications for initiating treatment. Results: The study identified 168 patients with a primary desmoid tumor initially managed with active surveillance. The tumors were located in the abdominal wall (n = 61, 36%), an extremity (n = 51, 30%), chest wall (n = 30, 18%), intra-abdominal site (n = 15, 9%), or elsewhere (n = 11, 6%). Of all the patients, 36% experienced radiologic progressive disease, 36% had stable disease, and 27% regressed. The patients younger than 50 years were more likely to progress (p = 0.046), whereas the patients with chest wall or upper-extremity tumors reported significantly more pain (p = 0.01). Eventually, 46% of the patients proceeded to treatment. The median time to start of treatment after initial surveillance was 31 months, whereas the median follow-up time for the patients not receiving any treatment was 40.5 months. The indications for initiation of treatment were pain (32%), progression (31%), or both (13%). Conclusions: Patients with desmoid tumors can be managed with initial active surveillance, although almost half of patients may eventually need treatment. Pain, tumor progression, or both are the most common indications for the initiation of treatment.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2021

Imaging and dosimetric characteristics of 67 Cu

In recent years the use of beta-emitting radiopharmaceuticals for cancer therapy has expanded rapidly following development of therapeutics for neuroendocrine tumors, prostate cancer, and other oncologic malignancies. One emerging beta-emitting radioisotope of interest for therapy is 67 Cu (t 1/2 : 2.6 d) due to its chemical equivalency with the widely-established positron-emitting isotope 64 Cu (t 1/2 : 12.7 h). Here we evaluate both the imaging and dosimetric characteristics of 67 Cu, as well as producing the first report of SPECT/CT imaging using 67 Cu. To this end, 67 Cu was produced by photon-induced reactions on isotopically-enriched 68 Zn at the Low-Energy Accelerator Facility (LEAF) of Argonne National Laboratory, followed by bulk separation of metallic 68 Zn by sublimation and radiochemical purification by column chromatography. Gamma spectrometry was performed by efficiency-calibrated high-purity germanium (HPGe) analysis to verify absolute activity calibration and establish radionuclidic purity. Absolute activity measurements corroborated manufacturer-recommended dose-calibrator settings and no radionuclidic impurities were observed. Using the Clinical Trials Network anthropomorphic chest phantom, SPECT/CT images were acquired. Medium Energy (ME) SPECT collimation was found to provide the best image quality from the primary 185 keV gamma emission of 67 Cu. Reconstructed images of 67 Cu were similar in quality to images acquired using 177 Lu. Recovery coefficients were calculated and compared against quantitative images of 99m Tc, 177 Lu, and 64 Cu within the same anthropomorphic chest phantom. Production and clinical imaging of 67 Cu appears feasible, and future studies investigating the therapeutic efficacy of 67 Cu-based radiopharmaceuticals are warranted.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2020

Lung cancer mortality among construction workers: implications for early detection

Objectives This study examined predictors of lung cancer mortality, beyond age and smoking, among construction workers employed at US Department of Energy (DOE) sites to better define eligibility for low-dose CT (LDCT) lung cancer screening. Methods Predictive models were based on 17 069 workers and 352 lung cancer deaths. Risk factors included age, gender, race/ethnicity, cigarette smoking, years of trade or DOE work, body mass index (BMI), chest X-ray results, spirometry results, respiratory symptoms, beryllium sensitisation and personal history of cancer. Competing risk Cox models were used to obtain HRs and to predict 5-year risks. Results Factors beyond age and smoking included in the final predictive model were chest X-ray changes, abnormal lung function, chronic obstructive pulmonary disease (COPD), respiratory symptoms, BMI, personal history of cancer and having worked 5 or more years at a DOE site or in construction. Risk-based LDCT eligibility demonstrated improved sensitivity, specificity and positive predictive value compared with current US Preventive Services Task Force guidelines. The risk of lung cancer death from 5 years of work in the construction industry or at a DOE site was comparable with the risk from a personal cancer history, a family history of cancer or a diagnosis of COPD. LDCT eligibility criteria used for DOE construction workers, which includes factors beyond age and smoking, identified 86% of participants who eventually would die from lung cancer compared with 51% based on age and smoking alone. Conclusions Results support inclusion of risk from occupational exposures and non-malignant respiratory clinical findings in LDCT clinical guidelines.

Public, Environmental & Occupational Health↗
DOE OSTI2022

Factors linked to participant attrition in a longitudinal occupational health surveillance program

For occupational medical screening programs focused on long-term health surveillance, participant attrition is a significant barrier to success. Here we investigate demographic, medical history, and clinical data from National Supplemental Screening Program (NSSP) examinees for association with likelihood of return for a second exam (rescreening). A total of 15,733 individuals completed at least one NSSP exam before December 31, 2016; of those, 4832 also completed a second exam on or before December 31, 2019. Stepwise logistic regression models were used to identify variables associated with whether a participant was rescreened in the NSSP. Individuals were less likely to return for rescreening if they had a history of any cancer; cardiovascular problems; diabetes or kidney disease; or if they used insulin. Age at time of first exam and job site category significantly influenced likelihood of return. Workers categorized as “guests” were more likely to return. Participants were less likely to return if they had an abnormal urinalysis, abnormal pulmonary function, pneumoconiosis, aortic atherosclerosis, or hearing loss at their initial exam. Participants who received a chest X-ray at their initial screening were more likely to return. The presence of health problems is strongly linked to screening program attrition. Participants who are older at the time of their initial screening exam are less likely to return. The discovery of several strong demographic, medical, and job associations reveals the importance for medical screening programs to understand and address factors that influence participant retention and, consequently, the effectiveness of long-term health surveillance activities.

60 APPLIED LIFE SCIENCES↗