BackgroundPneumothorax 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.Methods and findingsIn 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.ConclusionsWe 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 pneum...

BackgroundHealthy individuals on the lower end of the insulin sensitivity spectrum also have a reduced gene expression response to exercise for specific genes. The goal of this study was to determine the relationship between insulin sensitivity and exercise-induced gene expression in an unbiased, global manner.Methods and FindingsEuglycemic clamps were used to measure insulin sensitivity and muscle biopsies were done at rest and 30 minutes after a single acute exercise bout in 14 healthy participants. Changes in mRNA expression were assessed using microarrays, and miRNA analysis was performed in a subset of 6 of the participants using sequencing techniques. Following exercise, 215 mRNAs were changed at the probe level (Bonferroni-corrected P<0.00000115). Pathway and Gene Ontology analysis showed enrichment in MAP kinase signaling, transcriptional regulation and DNA binding. Changes in several transcription factor mRNAs were correlated with insulin sensitivity, including MYC, r=0.71; SNF1LK, r=0.69; and ATF3, r= 0.61 (5 corrected for false discovery rate). Enrichment in the 5’-UTRs of exercise-responsive genes suggested regulation by common transcription factors, especially EGR1. miRNA species of interest that changed after exercise included miR-378, which is located in an intron of the PPARGC1B gene.ConclusionsThese results indicate that transcription factor gene expression responses to exercise depend highly on insulin sensitivity in healthy people. The overall pattern suggests a coordinated cycle by which exercise and insulin sensitivity regulate gene expression in muscle.

Affect sharing and prosocial motivation are integral parts of empathy that are conceptually and mechanistically distinct. We used a neurodegenerative disease (NDG) lesion model to examine the neural correlates of these two aspects of real-world empathic responding. The study enrolled 275 participants, including 44 healthy older controls and 231 patients diagnosed with one of five neurodegenerative diseases (75 Alzheimer's disease, 58 behavioral variant frontotemporal dementia (bvFTD), 42 semantic variant primary progressive aphasia (svPPA), 28 progressive supranuclear palsy, and 28 non-fluent variant primary progressive aphasia (nfvPPA). Informants completed the Revised Self-Monitoring Scale's Sensitivity to the Expressive Behavior of Others (RSMS-EX) subscale and the Interpersonal Reactivity Index's Empathic Concern (IRI-EC) subscale describing the typical empathic behavior of the participants in daily life. Using regression modeling of the voxel based morphometry of T1 brain scans prepared using SPM8 DARTEL-based preprocessing, we isolated the variance independently contributed by the affect sharing and the prosocial motivation elements of empathy as differentially measured by the two scales. We found that the affect sharing component uniquely correlated with volume in right > left medial and lateral temporal lobe structures, including the amygdala and insula, that support emotion recognition, emotion generation, and emotional awareness. Prosocial motivation, in contrast, involved structures such as the nucleus accumbens (NaCC), caudate head, and inferior frontal gyrus (IFG), which suggests that an individual must maintain the capacity to experience reward, to resolve ambiguity, and to inhibit their own emotional experience in order to effectively engage in spontaneous altruism as a component of their empathic response to others.