Crash report data analysis for creating scenario-wise, spatio-temporal attention guidance to support computer vision-based perception of fatal crash risks

Reducing traffic fatal crashes has been an important mission of transportation. With the rapid development of sensor and Artificial Intelligence (AI) technologies, the computer vision (CV)-based crash anticipation in the near-crash phase is receiving growing attention. The ability to perceive fatal crash risks in an early stage is of paramount importance as well because it can improve the reliability of crash anticipation. Yet this task is challenging because it requires establishing a relationship between the driving scene information that CV can recognize and the fatal crash features that CV will not get until the crash occurrence. Image data with the annotation for directly training a reliable AI model for the early visual perception of fatal crash risks are not abundant. The Fatality Analysis Reporting System (FARS) contains big data on fatal crashes, which is a reliable data source for finding fatal crash clusters and discovering their distribution patterns to tell the association between driving scene characteristics and fatal crash features. To enhance CV's ability to perceive fatal crash risks earlier, this paper develops a data analytics model from fatal crash report data, which is named scenario-wise, spatio-temporal attention guidance. First, the paper identifies five descriptive variables that are sparse and thus allow for decomposing the 5-year (2013-2017) fatal crash dataset to develop scenario-wise attention guidance. Then, an exploratory analysis of location- and time-related descriptive variables suggests dividing fatal crashes into spatially defined groups. A group's temporal distribution pattern is an indicator of the similarity of fatal crashes in the group. Hierarchical clustering and K-means clustering further merge the spatially defined groups into six clusters according to the similarity of their temporal patterns. After that, association rule mining discovers the statistical relationship between the temporal information of driving scenes with fatal crash features, such as the first harmful event and the manner of collisions, for each cluster. The paper illustrates how the developed attention guidance supports the design and implementation of a preliminary CV model that can identify agents of a possibility to involve in fatal crashes from their environmental and context information.

The STING-IRF3 pathway is involved in lipotoxic injury of pancreatic β cells in type 2 diabetes

Lipotoxic injury of pancreatic β cells is an important pathological feature in type 2 diabetes mellitus (T2DM). Stimulator of interferon genes (STING) can recognize its own DNA leaked into the cytoplasm from damaged mitochondria or nuclei of the host cell, thus activating its downstream factor interferon regulatory factor 3 (IRF3), causing inflammation and apoptosis. The STING-IRF3 signaling pathway is closely related to glycolipid metabolism, but its relationship with the lipotoxicity of pancreatic β cells has rarely been reported. Here, we investigated the role of the STING-IRF3 signaling pathway in lipotoxicity-induced inflammation, apoptosis, and dysfunction of pancreatic β cells. We examined the activation of STING and IRF3 in islets of db/db mice and identified the role of the STING-IRF3 signaling pathway in palmitic acid (PA)-induced lipotoxic injury of INS-1, a rat insulinoma cell line. STING and phosphorylated IRF3 including downstream interferon-β were upregulated in islets of db/db mice and PA-induced INS-1 cells. Gene silencing of STING or IRF3 ameliorated PA-induced INS-1 cell inflammation and apoptosis, and reversed impaired insulin synthesis. Additionally, PA induced downregulation of the phosphoinositide 3-kinase-AKT signaling pathway, and impaired high glucose-stimulated insulin secretion was reversed after knockdown of STING or IRF3. Our results suggest that activation of the STING-IRF3 pathway triggers inflammation and apoptosis of pancreatic β cells, leading to β-cell damage and dysfunction. Hence, inhibition of this signaling pathway may represent a novel approach for β-cell protection in T2DM.

Automated Measurements of Key Morphological Features of Human Embryos for IVF

A major challenge in clinical In-Vitro Fertilization (IVF) is selecting the highest quality embryo to transfer to the patient in the hopes of achieving a pregnancy. Time-lapse microscopy provides clinicians with a wealth of information for selecting embryos. However, the resulting movies of embryos are currently analyzed manually, which is time consuming and subjective. Here, we automate feature extraction of time-lapse microscopy of human embryos with a machine-learning pipeline of five convolutional neural networks (CNNs). Our pipeline consists of (1) semantic segmentation of the regions of the embryo, (2) regression predictions of fragment severity, (3) classification of the developmental stage, and object instance segmentation of (4) cells and (5) pronuclei. Our approach greatly speeds up the measurement of quantitative, biologically relevant features that may aid in embryo selection.

Clinical and immunologic features among COVID-19-affected mother-infant pairs: antibodies to SARS-CoV-2 detected in breast milk

The coronavirus disease 2019 (COVID-19) pandemic remains threatening to women and children, but clinical evidence regarding women during pregnancy, puerperium and lactation is limited. We assessed clinical and immunologic features of and breastfeeding advice provided to mother–infant pairs. This observational analysis was conducted in a tertiary-care centre in Wuhan, China. Pregnant patients with laboratory-confirmed COVID-19 who delivered during hospitalization were enrolled. Clinical characteristics and serial specimens of the mother–infant pairs were examined, supplemented with follow-ups regarding recovery and breastfeeding. Fourteen pregnant patients had live births and recovered well; four patients continued breastfeeding while taking precautions. No neonatal infections were observed. No infants developed COVID-19 during breastfeeding. Common maternal symptoms were fever (11/14, 78.1%) and cough (6/14, 42.9%). A pregnancy-specific symptom was abnormal foetal movement, which was noticed by three patients (21.4%). The mean virus shedding time was 9 days (standard deviation, 6 days; range, 1–22 days). The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genome was not detected in breast milk or maternal vaginal secretions. Immunologic assay revealed seroconversion of IgM on day 8 after onset and IgG on day 28. Both IgM and IgG antibodies to SARS-CoV-2 were detected in breast milk, cord blood and neonatal serum. The study results suggest that passive acquisition of antibodies against SARS-CoV-2 is available by ingesting breast milk. Breastfeeding has a low risk of transmitting SARS-CoV-2 or escalating maternal disease, so continuing breastfeeding with prudent precautions is encouraged.

Sox2 Controls Periderm and Rugae Development to Inhibit Oral Adhesions

In humans, ankyloglossia and cleft palate are common congenital craniofacial anomalies, and these are regulated by a complex gene regulatory network. Understanding the genetic underpinnings of ankyloglossia and cleft palate will be an important step toward rational treatment of these complex anomalies. We inactivated the Sry (sex-determining region Y)-box 2 (Sox2) gene in the developing oral epithelium, including the periderm, a transient structure that prevents abnormal oral adhesions during development. This resulted in ankyloglossia and cleft palate with 100% penetrance in embryos examined after embryonic day 14.5. In Sox2 conditional knockout embryos, the oral epithelium failed to differentiate, as demonstrated by the lack of keratin 6, a marker of the periderm. Further examination revealed that the adhesion of the tongue and mandible expressed the epithelial markers E-Cad and P63. The expanded epithelia are Sox9-, Pitx2-, and Tbx1-positive cells, which are markers of the dental epithelium; thus, the dental epithelium contributes to the development of oral adhesions. Furthermore, we found that Sox2 is required for palatal shelf extension, as well as for the formation of palatal rugae, which are signaling centers that regulate palatogenesis. In conclusion, the deletion of Sox2 in oral epithelium disrupts palatal shelf extension, palatal rugae formation, tooth development, and periderm formation. The periderm is required to inhibit oral adhesions and ankyloglossia, which is regulated by Sox2. In addition, oral adhesions occur through an expanded dental epithelial layer that inhibits epithelial invagination and incisor development. This process may contribute to dental anomalies due to ankyloglossia.

Risk of SARS-CoV-2 infection among contacts of individuals with COVID-19 in Hangzhou, China

Objectives

This study determined the rate of secondary infection among contacts of individuals with confirmed coronavirus disease 2019 (COVID-19) in Hangzhou according to the type of contacts, the intensity of contacts, and their relationship with the index patient.

Study design

This is a retrospective cohort study.

Methods

The analysis used the data of 2994 contacts of 144 individuals with confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. The contacts were categorized according to the information source, type of contact, location, intensity of contact, and relationship with the index patient.

Results

The incidence of infection differed significantly according to contact type. Of the contacts, 186 (6.2%) developed symptoms, and 71 (2.4%) had confirmed SARS-CoV-2 infection. The main symptoms were cough and fever. Compared with those who had brief contact with the index case, those who had dined with the index case had 2.6 times higher risk of acquiring infection; those who had shared transport with, had visited, or had contact with the index case in a medical institution had 3.6 times higher risk of acquiring infection; and household contacts had 41.7 times higher risk of acquiring infection. Family members had 31.6 times higher risk of acquiring infection than healthcare providers or other patients exposed to an index case.

Conclusions

The form and frequency of contact are the main factors affecting the risk of infection among contacts of individuals with COVID-19. Centralized isolation and observation of close contacts of individuals with confirmed SARS-CoV-2 infection, in addition to population-based control measures, can reduce the risk of secondary infections and curb the spread of the infection.

•

The incidence of infection differed significantly according to contact type.

•

Contact form and frequency are important to the risk of COVID-19.

•

Centralized isolation and observation of contacts curbed the spread of the infection.

Using IL-2R/lymphocytes for predicting the clinical progression of patients with COVID-19

Effective laboratory markers for the estimation of disease severity and predicting the clinical progression of coronavirus disease-2019 (COVID-19) is urgently needed. Laboratory tests, including blood routine, cytokine profiles and infection markers, were collected from 389 confirmed COVID-19 patients. The included patients were classified into mild (n = 168), severe (n = 169) and critical groups (n = 52). The leukocytes, neutrophils, infection biomarkers [such as C-reactive protein (CRP), procalcitonin (PCT) and ferritin] and the concentrations of cytokines [interleukin (IL)-2R, IL-6, IL-8, IL-10 and tumor necrosis factor (TNF)-α] were significantly increased, while lymphocytes were significantly decreased with increased severity of illness. The amount of IL-2R was positively correlated with the other cytokines and negatively correlated with lymphocyte number. The ratio of IL-2R to lymphocytes was found to be remarkably increased in severe and critical patients. IL-2R/lymphocytes were superior compared with other markers for the identification of COVID-19 with critical illness, not only from mild but also from severe illness. Moreover, the cytokine profiles and IL-2R/lymphocytes were significantly decreased in recovered patients, but further increased in disease-deteriorated patients, which might be correlated with the outcome of COVID-19. Lymphopenia and increased levels of cytokines were closely associated with disease severity. The IL-2R/lymphocyte was a prominent biomarker for early identification of severe COVID-19 and predicting the clinical progression of the disease.

Engineering CRISPR interference system to enhance the production of pyrroloquinoline quinone in Klebsiella pneumonia

Pyrroloquinoline quinone (PQQ) is a cofactor of glucose dehydrogenase (GDH) and thus participates in glucose utilization. In Klebsiella pneumoniae, glucose utilization involves PQQ-dependent direct oxidation pathway (DOP) and phosphoenolpyruvate-dependent transport system (PTS). It is challenging to overproduce PQQ, as its biosynthesis remains unclear. Here, we report that PQQ production can be enhanced by stimulating the metabolic demand for it. First, we developed CRISPR interference (CRISPRi) system to block PTS and thereby intensify DOP. In shake-flask cultivation, the strain with CRISPRi system (simultaneously inhibiting four PTS-related genes) produced 225·65 nmol l^-1 PQQ, which was 2·14 times that of wild type. In parallel, an exogenous soluble glucose dehydrogenase (sGDH) was overexpressed in K. pneumoniae. In the shake-flask cultivation, this sGDH-overexpressing strain accumulated 140·05 nmol l^-1 PQQ, which was 1·33 times that of wild type. To combine the above two strategies, we engineered a strain harbouring both CRISPRi vector and sGDH-overexpressing vector. In the shake-flask cultivation, this two-plasmid strain generated 287·01 nmol l^-1 PQQ, which was 2·72 times that of wild type. In bioreactor cultivation, this two-plasmid strain produced 2206·1 nmol l^-1 PQQ in 57 h, which was 7·69 times that in shake-flask cultivation. These results indicate that PQQ production can be enhanced by intensifying DOP, as the apo-enzyme GDH is intrinsically coupled with cofactor PQQ. This study provides a strategy for the production of cofactors whose biosynthesis mechanisms remain ambiguous. SIGNIFICANCE AND IMPACT OF THE STUDY: Pyrroloquinoline quinone (PQQ) is an economically important chemical, which typically serves as a cofactor of glucose dehydrogenase (GDH) and thus participates in glucose metabolism. Klebsiella pneumoniae can naturally synthesize PQQ, but current yield constrains its commercialization. In this study, the PQQ level was improved by stimulating metabolic demand for PQQ, instead of overexpressing PQQ synthetic genes, as the synthetic mechanism remains ambiguous.

[Genomic analysis of a 2019-novel coronavirus (2019-nCoV) strain in the first COVID-19 patient found in Hangzhou]

Objective: To understand the viral genomic characteristics of a 2019-novel coronavirus (2019-nCoV) strain in the first COVID-19 patient found in Hangzhou, China. Methods: Viral RNA was extracted in throat swab and sputum sample of the patient and was performed real-time reverse transcription PCR detection and obtained viral genome by high-throughput sequencing method. Phylogenetic analysis was conducted using 29 2019-nCoV genomes and 30 β-coronavirus genomes deposited in NCBI GenBank. Fifteen genomes from Wuhan were grouped by mutation sites and others were identified by Wuhan's or specific mutation sites. Results: A 29 833 bp length genome of the first 2019-nCoV strain in Hangzhou was obtained, covering full length of the coding regions of coronavirus. Phylogenetic analysis showed that the genome was closest to the genome of a bat SARS-like coronavirus strain RaTG13 with an identity of 96.11% (28 666/29 826). Among the genes between two genomes, E genes were highly conserved (99.56%), while S genes had lowest identity (92.87%). The genome sequence similarities among 29 strains from China (Hangzhou, Wuhan, and Shenzhen), Japan, USA, and Finland, were all more than 99.9%; however, some single nucleotide polymorphisms were identified in some strains. Conclusion: The genome of Hangzhou 2019-nCoV strain was very close to the genomes of strains from other cities in China and overseas collected at early epidemic phase. The 2019-nCoV genome sequencing method used in this paper provides an useful tool for monitoring variation of viral genes.

[The impact of carbapenem-resistance Pseudomonas aeruginosa infections on mortality of patients with hematological disorders]

Objective: To assess the risk factors for mortality and clinical outcome of carbapenem-resistant Pseudomonas aeruginosa (CRPA) infections in patients with hematological disorders. Methods: The data of in-patients with hematological disorders infected by CRPA or carbapenem-susceptible Pseudomonas aeruginosa (CSPA) were recorded in a seven-year retrospective cohort study. Risk factors for CRPA infections and impact of on mortality were identified. The primary end point was 30-day all-cause mortality. Results: A total of 81 patients with PA infections were included in the study, including 58 CSPA and 23 CRPA. Most of the primary diseases were acute leukemia or lymphoma (79.0%, 64/81). The median absolute neutrophil count at infection onset was 0.24×10(9)/L. Independent risk factors associated with carbapenem-resistance included longer duration of hospital stay (P=0.013, OR=1.045) and carbapenem exposure one month prior to infections (P=0.005, OR=8.132). The 30-day all-cause mortality of the whole cohort was 29.6%(24/81), and 30-day attributable mortality was 13.6%(11/81). Pulmonary infection was the leading cause of death, accounting for 41.7%(10/24). The adjusted 30-day mortality rate was significantly higher in patients with CRPA compared with CSPA [60.9%(14/23) vs. 17.2%(10/58), P<0.001, respectively]. CRPA infection was an independent prognostic factor for 30-day mortality(P=0.011, OR=5.427). Other factors included old age, longer duration of neutropenia and poor functional performance. Conclusions: Patients with hematological disorders have high mortality rate and poor prognosis caused by CRPA infections, which mainly develop in lungs.