Both qualitative and quantitative evaluations have shown that the presented algorithm is able to accurately segment the aorta in low-dose non-contrast CT images.
Nodule classification in the context of low-resolution low-dose whole-chest CT images for the clinically relevant size range in the context of lung cancer screening is highly challenging, and results are moderate compared to what has been reported in the literature for other clinical contexts. Nodule class size distribution imbalance needs to be considered in the training and evaluation of computer-aided diagnostic systems for producing patient-relevant outcomes.
A three-dimensional (3-D) convolutional neural network (CNN) trained from scratch is presented for the classification of pulmonary nodule malignancy from low-dose chest CT scans. Recent approval of lung cancer screening in the United States provides motivation for determining the likelihood of malignancy of pulmonary nodules from the initial CT scan finding to minimize the number of follow-up actions. Classifier ensembles of different combinations of the 3-D CNN and traditional machine learning models based on handcrafted 3-D image features are also explored. The dataset consisting of 326 nodules is constructed with balanced size and class distribution with the malignancy status pathologically confirmed. The results show that both the 3-D CNN single model and the ensemble models with 3-D CNN outperform the respective counterparts constructed using only traditional models. Moreover, complementary information can be learned by the 3-D CNN and the conventional models, which together are combined to construct an ensemble model with statistically superior performance compared with the single traditional model. The performance of the 3-D CNN model demonstrates the potential for improving the lung cancer screening follow-up protocol, which currently mainly depends on the nodule size.
Inspired
by complex multifunctional leaves, in this study, we created robust
hierarchically wrinkled nanoporous polytetrafluoroethene (PTFE) surfaces
that exhibit superhydrophobic properties by combination of PTFE micellization
and spontaneous surface wrinkling on a commercially available thermoretractable
polystyrene (PS) sheet. A PTFE dispersion was coated onto the PS sheet,
followed by thermal treatment to remove the surfactants surrounding
the PTFE particles, and surface wrinkling was induced through a dynamic
thermal contraction process. Thermally induced contraction from the
PS sheet provided the driving force for developing and stabilizing
micrometer-sized wrinkle formation, whereas the nanometer-sized PTFE
particle aggregation formed a rigid nanoporous film, providing its
intrinsic hydrophobic character. By combining the hierarchical interfacial
structure and chemical composition, hierarchically wrinkled nanoporous
PTFE surfaces were fabricated, which exhibited extremely high water
repellence (water contact angle of ∼167°) and a water
rolling-off angle lower than 5°. The wrinkled patterns could
intimately bind the nanoporous PTFE layer through enhanced adhesion
from their curved surface and viscous liquid surfactants, making these
surfaces mechanically robust and offering potentially extendable alternatives
with self-cleaning, antifouling, and drag-reducing properties.
As an indispensable molecular machine universal in all living organisms, the ribosome has been selected by evolution to be the natural target of many antibiotics and small-molecule inhibitors. High-resolution structures of pathogen ribosomes are crucial for understanding the general and unique aspects of translation control in disease-causing microbes. With cryo-electron microscopy technique, we have determined structures of the cytosolic ribosomes from two human parasites, Trichomonas vaginalis and Toxoplasma gondii, at resolution of 3.2-3.4 Å. Although the ribosomal proteins from both pathogens are typical members of eukaryotic families, with a co-evolution pattern between certain species-specific insertions/extensions and neighboring ribosomal RNA (rRNA) expansion segments, the sizes of their rRNAs are sharply different. Very interestingly, rRNAs of T. vaginalis are in size comparable to prokaryotic counterparts, with nearly all the eukaryote-specific rRNA expansion segments missing. These structures facilitate the dissection of evolution path for ribosomal proteins and RNAs, and may aid in design of novel translation inhibitors.
Background
To assess the relationship between lung cancer and emphysema subtypes.
Objective
Airflow obstruction and emphysema predispose to lung cancer. Little is known, however, about the lung cancer risk associated with different emphysema phenotypes. We assessed the risk of lung cancer based on the presence, type and severity of emphysema, using visual assessment.
Methods
Seventy-two consecutive lung cancer cases were selected from a prospective cohort of 3,477 participants enrolled in the Clínica Universidad de Navarra’s lung cancer screening program. Each case was matched to three control subjects using age, sex, smoking history and body mass index as key variables. Visual assessment of emphysema and spirometry were performed. Logistic regression and interaction model analysis were used in order to investigate associations between lung cancer and emphysema subtypes.
Results
Airflow obstruction and visual emphysema were significantly associated with lung cancer (OR = 2.8, 95%CI: 1.6 to 5.2; OR = 5.9, 95%CI: 2.9 to 12.2; respectively). Emphysema severity and centrilobular subtype were associated with greater risk when adjusted for confounders (OR = 12.6, 95%CI: 1.6 to 99.9; OR = 34.3, 95%CI: 25.5 to 99.3, respectively). The risk of lung cancer decreases with the added presence of paraseptal emphysema (OR = 4.0, 95%CI: 3.6 to 34.9), losing this increased risk of lung cancer when it occurs alone (OR = 0.7, 95%CI: 0.5 to 2.6).
Conclusions
Visual scoring of emphysema predicts lung cancer risk. The centrilobular phenotype is associated with the greatest risk.
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.