Pneumothorax can precipitate a life-threatening emergency due to lung collapse and respiratory or circulatory distress. Pneumothorax is typically detected on chest X-ray; however, treatment is ...reliant on timely review of radiographs. Since current imaging volumes may result in long worklists of radiographs awaiting review, an automated method of prioritizing X-rays with pneumothorax may reduce time to treatment. Our objective was to create a large human-annotated dataset of chest X-rays containing pneumothorax and to train deep convolutional networks to screen for potentially emergent moderate or large pneumothorax at the time of image acquisition.
In all, 13,292 frontal chest X-rays (3,107 with pneumothorax) were visually annotated by radiologists. This dataset was used to train and evaluate multiple network architectures. Images showing large- or moderate-sized pneumothorax were considered positive, and those with trace or no pneumothorax were considered negative. Images showing small pneumothorax were excluded from training. Using an internal validation set (n = 1,993), we selected the 2 top-performing models; these models were then evaluated on a held-out internal test set based on area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV). The final internal test was performed initially on a subset with small pneumothorax excluded (as in training; n = 1,701), then on the full test set (n = 1,990), with small pneumothorax included as positive. External evaluation was performed using the National Institutes of Health (NIH) ChestX-ray14 set, a public dataset labeled for chest pathology based on text reports. All images labeled with pneumothorax were considered positive, because the NIH set does not classify pneumothorax by size. In internal testing, our "high sensitivity model" produced a sensitivity of 0.84 (95% CI 0.78-0.90), specificity of 0.90 (95% CI 0.89-0.92), and AUC of 0.94 for the test subset with small pneumothorax excluded. Our "high specificity model" showed sensitivity of 0.80 (95% CI 0.72-0.86), specificity of 0.97 (95% CI 0.96-0.98), and AUC of 0.96 for this set. PPVs were 0.45 (95% CI 0.39-0.51) and 0.71 (95% CI 0.63-0.77), respectively. Internal testing on the full set showed expected decreased performance (sensitivity 0.55, specificity 0.90, and AUC 0.82 for high sensitivity model and sensitivity 0.45, specificity 0.97, and AUC 0.86 for high specificity model). External testing using the NIH dataset showed some further performance decline (sensitivity 0.28-0.49, specificity 0.85-0.97, and AUC 0.75 for both). Due to labeling differences between internal and external datasets, these findings represent a preliminary step towards external validation.
We trained automated classifiers to detect moderate and large pneumothorax in frontal chest X-rays at high levels of performance on held-out test data. These models may provide a high specificity screening solution to detect moderate or large pneumothorax on images collected when human review might be delayed, such as overnight. They are not intended for unsupervised diagnosis of all pneumothoraces, as many small pneumothoraces (and some larger ones) are not detected by the algorithm. Implementation studies are warranted to develop appropriate, effective clinician alerts for the potentially critical finding of pneumothorax, and to assess their impact on reducing time to treatment.
The tumour-associated microbiota is an intrinsic component of the tumour microenvironment across human cancer types
. Intratumoral host-microbiota studies have so far largely relied on bulk tissue ...analysis
, which obscures the spatial distribution and localized effect of the microbiota within tumours. Here, by applying in situ spatial-profiling technologies
and single-cell RNA sequencing
to oral squamous cell carcinoma and colorectal cancer, we reveal spatial, cellular and molecular host-microbe interactions. We adapted 10x Visium spatial transcriptomics to determine the identity and in situ location of intratumoral microbial communities within patient tissues. Using GeoMx digital spatial profiling
, we show that bacterial communities populate microniches that are less vascularized, highly immuno‑suppressive and associated with malignant cells with lower levels of Ki-67 as compared to bacteria-negative tumour regions. We developed a single-cell RNA-sequencing method that we name INVADEseq (invasion-adhesion-directed expression sequencing) and, by applying this to patient tumours, identify cell-associated bacteria and the host cells with which they interact, as well as uncovering alterations in transcriptional pathways that are involved in inflammation, metastasis, cell dormancy and DNA repair. Through functional studies, we show that cancer cells that are infected with bacteria invade their surrounding environment as single cells and recruit myeloid cells to bacterial regions. Collectively, our data reveal that the distribution of the microbiota within a tumour is not random; instead, it is highly organized in microniches with immune and epithelial cell functions that promote cancer progression.
The study aimed to determine if computer vision techniques rooted in deep learning can use a small set of radiographs to perform clinically relevant image classification with high fidelity. One ...thousand eight hundred eighty-five chest radiographs on 909 patients obtained between January 2013 and July 2015 at our institution were retrieved and anonymized. The source images were manually annotated as frontal or lateral and randomly divided into training, validation, and test sets. Training and validation sets were augmented to over 150,000 images using standard image manipulations. We then pre-trained a series of deep convolutional networks based on the open-source GoogLeNet with various transformations of the open-source ImageNet (non-radiology) images. These trained networks were then fine-tuned using the original and augmented radiology images. The model with highest validation accuracy was applied to our institutional test set and a publicly available set. Accuracy was assessed by using the Youden Index to set a binary cutoff for frontal or lateral classification. This retrospective study was IRB approved prior to initiation. A network pre-trained on 1.2 million greyscale ImageNet images and fine-tuned on augmented radiographs was chosen. The binary classification method correctly classified 100 % (95 % CI 99.73–100 %) of both our test set and the publicly available images. Classification was rapid, at 38 images per second. A deep convolutional neural network created using non-radiological images, and an augmented set of radiographs is effective in highly accurate classification of chest radiograph view type and is a feasible, rapid method for high-throughput annotation.
The first step of RAF activation involves binding to active RAS, resulting in the recruitment of RAF to the plasma membrane. To understand the molecular details of RAS-RAF interaction, we present ...crystal structures of wild-type and oncogenic mutants of KRAS complexed with the RAS-binding domain (RBD) and the membrane-interacting cysteine-rich domain (CRD) from the N-terminal regulatory region of RAF1. Our structures reveal that RBD and CRD interact with each other to form one structural entity in which both RBD and CRD interact extensively with KRAS. Mutations at the KRAS-CRD interface result in a significant reduction in RAF1 activation despite only a modest decrease in binding affinity. Combining our structures and published data, we provide a model of RAS-RAF complexation at the membrane, and molecular insights into RAS-RAF interaction during the process of RAS-mediated RAF activation.
Many human cells can sense the presence of exogenous DNA during infection though the cytosolic DNA receptor cyclic GMP-AMP synthase (cGAS), which produces the second messenger cyclic GMP-AMP (cGAMP). ...Other putative DNA receptors have been described, but whether their functions are redundant, tissue-specific or integrated in the cGAS-cGAMP pathway is unclear. Here we show that interferon-γ inducible protein 16 (IFI16) cooperates with cGAS during DNA sensing in human keratinocytes, as both cGAS and IFI16 are required for the full activation of an innate immune response to exogenous DNA and DNA viruses. IFI16 is also required for the cGAMP-induced activation of STING, and interacts with STING to promote STING phosphorylation and translocation. We propose that the two DNA sensors IFI16 and cGAS cooperate to prevent the spurious activation of the type I interferon response.
Lenvatinib, an oral inhibitor of vascular endothelial growth factor receptors 1, 2, and 3, fibroblast growth factor receptors 1 through 4, platelet-derived growth factor receptor α, RET, and KIT, ...showed clinical activity in a phase 2 study involving patients with differentiated thyroid cancer that was refractory to radioiodine (iodine-131).
In our phase 3, randomized, double-blind, multicenter study involving patients with progressive thyroid cancer that was refractory to iodine-131, we randomly assigned 261 patients to receive lenvatinib (at a daily dose of 24 mg per day in 28-day cycles) and 131 patients to receive placebo. At the time of disease progression, patients in the placebo group could receive open-label lenvatinib. The primary end point was progression-free survival. Secondary end points included the response rate, overall survival, and safety.
The median progression-free survival was 18.3 months in the lenvatinib group and 3.6 months in the placebo group (hazard ratio for progression or death, 0.21; 99% confidence interval, 0.14 to 0.31; P<0.001). A progression-free survival benefit associated with lenvatinib was observed in all prespecified subgroups. The response rate was 64.8% in the lenvatinib group (4 complete responses and 165 partial responses) and 1.5% in the placebo group (P<0.001). The median overall survival was not reached in either group. Treatment-related adverse effects of any grade, which occurred in more than 40% of patients in the lenvatinib group, were hypertension (in 67.8% of the patients), diarrhea (in 59.4%), fatigue or asthenia (in 59.0%), decreased appetite (in 50.2%), decreased weight (in 46.4%), and nausea (in 41.0%). Discontinuations of the study drug because of adverse effects occurred in 37 patients who received lenvatinib (14.2%) and 3 patients who received placebo (2.3%). In the lenvatinib group, 6 of 20 deaths that occurred during the treatment period were considered to be drug-related.
Lenvatinib, as compared with placebo, was associated with significant improvements in progression-free survival and the response rate among patients with iodine-131-refractory thyroid cancer. Patients who received lenvatinib had more adverse effects. (Funded by Eisai; SELECT ClinicalTrials.gov number, NCT01321554.).
Whole-genome sequencing projects are increasingly populating the tree of life and characterizing biodiversity
. Sparse taxon sampling has previously been proposed to confound phylogenetic inference
, ...and captures only a fraction of the genomic diversity. Here we report a substantial step towards the dense representation of avian phylogenetic and molecular diversity, by analysing 363 genomes from 92.4% of bird families-including 267 newly sequenced genomes produced for phase II of the Bird 10,000 Genomes (B10K) Project. We use this comparative genome dataset in combination with a pipeline that leverages a reference-free whole-genome alignment to identify orthologous regions in greater numbers than has previously been possible and to recognize genomic novelties in particular bird lineages. The densely sampled alignment provides a single-base-pair map of selection, has more than doubled the fraction of bases that are confidently predicted to be under conservation and reveals extensive patterns of weak selection in predominantly non-coding DNA. Our results demonstrate that increasing the diversity of genomes used in comparative studies can reveal more shared and lineage-specific variation, and improve the investigation of genomic characteristics. We anticipate that this genomic resource will offer new perspectives on evolutionary processes in cross-species comparative analyses and assist in efforts to conserve species.
Given predicted increases in urbanization in tropical and subtropical regions, understanding the processes shaping urban coral reefs may be essential for anticipating future conservation challenges. ...We used a case study approach to identify unifying patterns of urban coral reefs and clarify the effects of urbanization on hard coral assemblages. Data were compiled from 11 cities throughout East and Southeast Asia, with particular focus on Singapore, Jakarta, Hong Kong, and Naha (Okinawa). Our review highlights several key characteristics of urban coral reefs, including “reef compression” (a decline in bathymetric range with increasing turbidity and decreasing water clarity over time and relative to shore), dominance by domed coral growth forms and low reef complexity, variable city-specific inshore-offshore gradients, early declines in coral cover with recent fluctuating periods of acute impacts and rapid recovery, and colonization of urban infrastructure by hard corals. We present hypotheses for urban reef community dynamics and discuss potential of ecological engineering for corals in urban areas.
Display omitted
•Corals and coral reefs near coastal cities are subject to a suite of intense stressors.•Sediment pollution constrains urban reefs to shallow depths and limits complexity.•Reefs in major cities declined early, and coral cover now fluctuates frequently.•Artificial structures near cities provide novel habitats that are colonized by coral.•Whether urban coral reefs constitute a unique ecosystem type needs to be tested.