A convolutional neural network is a deep learning neural network designed for processing structured arrays of data such as images. Such advancements in deep learning have fundamentally changed the way tasks, like architectural segmentation, are performed. However, an accurate deep convolutional neural network (DCNN) model needs a large volume of labeled training data and labeling these data can be a slow and extremely expensive manual undertaking.

To create the synthetic digital city twin data, the investigators used a 3D city model from the PLATEAU platform, which contains 3D models of most Japanese cities at an extremely high level of detail. They loaded this model into the Unity game engine and created a camera setup on a virtual car, which drove around the city and acquired the virtual data images under various lighting and weather conditions. The Google Maps API was then used to obtain real street-level images of the same study area for the experiments.

The researchers found that the digital city twin data leads to better results than purely virtual data with no real-world counterpart. Furthermore, adding synthetic data to a real dataset improves segmentation accuracy. However, most importantly, the investigators found that when a certain fraction of real data is included in the digital city twin synthetic dataset, the segmentation accuracy of the DCNN is boosted significantly. In fact, its performance becomes competitive with that of a DCNN trained on 100% real data. “These results reveal that our proposed synthetic dataset could potentially replace all the real images in the training set,” says Tomohiro Fukuda, the corresponding author of the paper.

Automatically separating out the individual building facades that appear in an image is useful for construction management and architecture design, large-scale measurements for retrofits and energy analysis, and even visualizing building facades that have been demolished. The system was tested on multiple cities, demonstrating the proposed framework’s transferability. The hybrid dataset of real and synthetic data yields promising prediction results for most modern architectural styles. This makes it a promising approach for training DCNNs for architectural segmentation tasks in the future — without the need for costly manual data annotation.

Story Source:

Materials provided by Osaka University. Note: Content may be edited for style and length.

The measurements and analyses of coronavirus levels in the wastewater of Gothenburg attracted a great deal of attention during the pandemic. The weekly reports have shown both how widespread SARS-CoV-2 infection is in the community and its distribution among variants of the virus.

Beginning in February 2020, the virus measurements taken in the monitoring rapidly became a useful indicator for forecasting load peaks in health care. High concentrations of SARS-CoV-2 in the wastewater were followed by rising numbers of people with COVID-19 needing hospitalization.

The association emerges with striking precision in the study now published in the scientific journal iScience. Each of the four pandemic waves in 2020-2022 exhibits a pattern in which, within a couple of weeks after SARS-CoV-2 peaked in the wastewater, a rise in the number of newly admitted hospital patients with COVID-19 ensued.

Increased pressure on 1177 Vårdguiden

The virus peaks in the wastewater were followed not only by heavier burdens on inpatient care, but also by predictable increases in pressure on the 1177 Vårdguiden e-service. One to two weeks after a wastewater virus peak, more calls were coming in about acute breathlessness in adults.

“The study shows that virus monitoring in wastewater can predict how a pandemic is going to develop and its burden on several parts of the health system,” says Hao Wang, postdoctoral researcher at Sahlgrenska Academy’s Department of Infectious Diseases, University of Gothenburg, and the study’s first author.

The study provides an account of the degree of testing capacity in Sweden during various phases of the pandemic. The scientists say that the national statistics on the number of confirmed cases did not reflect the actual spread of infection.

However, even when their symptoms were mild, all those infected had the virus in their urine and feces. This enabled detection of the virus in wastewater, in the form of RNA (genetic material). The weekly reports were based on daily samples of wastewater collected by Gryaab, the municipal water treatment company in Gothenburg.

Lasting benefit to community

Heléne Norder, research leader at the University of Gothenburg’s Sahlgrenska Academy and the last author of the study, sees the results as proof of how virus surveillance during the pandemic has benefited the community. She also identifies the distinct advantages of the method used.

“The method we use in Gothenburg also enables monitoring of other viruses, which gives us unique scope for quickly spotting ongoing outbreaks. In the area, besides SARS-CoV-2, we’ve also been able to demonstrate the presence of viruses that spread through food or water and are excreted in the feces. Some of those caused outbreaks — norovirus, the ‘winter vomiting bug’, and astrovirus among children, for instance — during the pandemic.

“So, during that time,” she continues, “we found changes in the virus levels that were directly relatable to the number of people infected in Gothenburg. Our research group intends to keep analyzing different viruses in the wastewater as long as we have research funding for it.”

The planning also includes further development of the technique to enable its adoption by all the laboratories that wish to carry out continuous virus monitoring.

“This might result in monitoring of more viruses in more regions for rapid identification of future outbreaks and efficient, proactive nationwide surveillance of ongoing ones,” Norder concludes.

Story Source:

Materials provided by University of Gothenburg. Note: Content may be edited for style and length.

Animals, especially endangered and threatened species, are often exploited and sold for their pelts or used as medicine, aphrodisiacs, curios, food and spiritual artifacts.

In a new study published in Global Change Biology, Arizona State University researchers estimate that more than 1.1 million sea turtles have been illegally killed and, in some cases, trafficked between 1990 and 2020. Even with existing laws prohibiting their capture and use, as many as 44,000 sea turtles were exploited each year over the past decade in 65 countries or territories and in 44 of the world’s 58 major sea turtle populations.

Despite the seemingly large number of poached turtles, the study shows that the reported illegal exploitation of sea turtles declined by approximately 28% over the last decade — something that surprised the researchers. They initially expected to see an overall increase in reported poaching.

“The decline over the past decade could be due to increased protective legislation and enhanced conservation efforts, coupled with an increase in awareness of the problem or changing local norms and traditions,” says Kayla Burgher, co-first author of the study and a doctoral student in ASU’s environmental life sciences program in the School of Life Sciences.

In addition to the slight decline, the researchers found that most of the reported illegal exploitation over the past decade occurred in large, stable and genetically diverse sea turtle populations.

Jesse Senko, co-first author of the study and an assistant research professor with the ASU School for the Future of Innovation in Society says this discovery may be a silver lining to the high number of turtles illegally exploited. “What this means is that most of these sea turtles came from healthy, low-risk populations, which suggests that, with a few exceptions, current levels of illegal exploitation are likely not having a major detrimental impact on most major sea turtle populations throughout the world’s oceans.”

Senko adds, however, the results should be cautiously considered. “Assessing any illegal activity is difficult, and the take and trade of sea turtles is no exception, especially when it becomes organized or connected to crime syndicates. Our assessment also did not include eggs or turtle products, such as bracelets or earrings made from sea turtle shells that could not be easily attributed to individual turtles,” says Senko.

In the study, the researchers reviewed data from peer-reviewed journal articles, archived media reports, NGO reports, and online questionnaires to determine a comprehensive look at existing information on exploited sea turtles. The study revealed additional patterns and trends that may assist in determining conservation management priorities. For example, Vietnam was the most common country of origin for illegal sea turtle trafficking, while China and Japan served as destinations for nearly all trafficked sea turtle products. Similarly, Vietnam to China was the most common trade route across all three decades.

Across the 30-year study period, 95% of poached sea turtles came from two species — green and hawksbill turtles — both of which are listed under the U.S. Endangered Species Act. Also, Southeast Asia and Madagascar emerged as major hotspots for illegal sea turtle take and trade, particularly for critically endangered hawksbills, which are prized in the illicit wildlife trade for their beautiful shells.

“Our assessment is an important foundation for future research and outreach efforts regarding illegal sea turtle exploitation. We believe this study can help conservation practitioners and legislators prioritize conservation efforts and allocate their resources to best help protect sea turtle populations from harmful levels of exploitation worldwide,” says Burgher.

The research team says much more needs to be done to sustain global biodiversity.

“Increased support for governments lacking the resources to protect sea turtles is needed, along with support for communities to sustain human well-being in the face of restrictions or bans on sea turtle exploitation. We must develop conservation strategies that benefit both people and turtles,” says Senko.

The ADNP gene affects brain formation, development, and function, and the protein produced from it helps control the expression of other genes. ADNP mutations are one of the most common single-gene causes of autism. Ketamine was approved in the United States in 1970 and is used for anesthesia and pain management, and more recently as a treatment for depression. Studies in animal models suggest that low-dose ketamine may be neuroprotective and increase expression of the ADNP gene.

“We were intrigued by the preclinical evidence suggesting that low-dose ketamine may increase levels of the ADNP protein and compensate for its loss in ADNP syndrome, so we designed this study to evaluate the safety, tolerability, and behavioral outcomes of low-dose ketamine in children with the syndrome,” says Alexander Kolevzon, MD, Clinical Director of the Seaver Autism Center. “We also sought to explore the feasibility of using electrophysiological biomarkers and computerized eye-tracking to assess sensitivity to treatment.”

In order to evaluate the effect of ketamine, the Mount Sinai research team used a single-dose (0.5mg/kg), open-label design, with ketamine infused intravenously over 40 minutes. Ten children with ADNP syndrome, ages six to 12 years, were enrolled. They found ketamine was generally well-tolerated, and there were no serious adverse events. The most common adverse events were elation/silliness (50 percent), fatigue (40 percent), and increased aggression (40 percent). Using parent-report instruments to assess treatment effects, ketamine was associated with improvements in a wide array of domains, including social behavior, attention deficit and hyperactivity, restricted and repetitive behaviors, and sensory sensitivities, a week after administration.

Results from the clinician-rated assessments indicated improvement based on the Clinical Global Impression-Improvement scale, a seven-point scale commonly used by clinicians to assess how much a patient’s illness has improved or worsened relative to a baseline state at the beginning of an intervention. Importantly, results across clinician-rated and caregiver-rated assessments were largely consistent. The results also highlight the potential of assessing early changes in social attention with computerized eye-tracking and the electrophysiological measurement of a listening task known as auditory steady-state response.

“We are encouraged by these findings, which provide preliminary support for ketamine to help reduce negative effects of this devastating syndrome,” Dr. Kolevzon says. “Future studies using a placebo-controlled design and studying the effects of repeated dosing over a longer duration of time and in a larger cohort of participants are needed before ketamine is used clinically, but our study is a promising first step in that process.”

Ongoing studies are using RNA sequencing to measure change in ADNP expression and other genes, as well as DNA methylation analysis, which has been previously described as relevant in ADNP syndrome. DNA methylation regulates and silences the expression of genes and is vital for embryonic development. Increased occurrence of rare and extreme DNA methylation levels have been linked with neurodevelopmental disorders and congenital anomalies.