Saving endangered species: New AI method counts manatee clusters in real time

FAU College of Engineering and Computer Science researchers among first to use deep learning-based crowd counting approach

Peer-Reviewed Publication

FLORIDA ATLANTIC UNIVERSITY

 

VIDEO: 

HARNESSING THE POWER OF AI, RESEARCHERS ARE AMONG THE FIRST TO USE A DEEP LEARNING-BASED CROWD COUNTING APPROACH TO AUTOMATICALLY COUNT THE NUMBER OF MANATEES IN A DESIGNATED REGION, USING IMAGES CAPTURED FROM CCTV CAMERAS, WHICH ARE READILY AVAILABLE, AS INPUT.

view more 

CREDIT: FLORIDA ATLANTIC UNIVERSITY

Manatees are endangered species volatile to the environment.  Because of their voracious appetites, they often spend up to eight hours a day grazing for food within shallow waters, making them vulnerable to environmental changes and other risks.

Accurately counting manatee aggregations within a region is not only biologically meaningful in observing their habit, but also crucial for designing safety rules for boaters and divers as well as scheduling nursing, intervention, and other plans. Nevertheless, counting manatees is challenging. 

Because manatees tend to live in herds, they often block each other when viewed from the surface. As a result, small manatees are likely to be partially or completely blocked from view. In addition, water reflections tend to make manatees invisible, and they also can be mistaken for other objects such as rocks and branches.

While aerial survey data are used in some regions to count manatees, this method is time-consuming and costly, and the accuracy depends on factors such as observer bias, weather conditions and time of day. Moreover, it is crucial to have a low-cost method that provides a real-time count to alert ecologists of threats early to enable them to act proactively to protect manatees.

Artificial intelligence is used in a wide spectrum of fields, and now, researchers from Florida Atlantic University’s College of Engineering and Computer Science have harnessed its powers to help save the beloved manatee. They are among the first to use a deep learning-based crowd counting approach to automatically count the number of manatees in a designated region, using images captured from CCTV cameras, which are readily available, as input.  

This pioneering study, published Scientific Reports, not only addresses the technical challenges of counting in complex outdoor environments but also offers potential ways to aid endangered species. 

To determine manatee densities and calculate their numbers, researchers used generic images captured from surveillance videos from the water surface. They then used a unique design matching to manatees’ shape – Anisotropic Gaussian Kernel (AGK) – to transform the images into manatee customized density maps, representing manatees’ unique body shapes.

Although many methods exist for counting, most of the existing counting methods are applied to crowds to count the number of people, due to their relevance to important applications such as urban planning and public safety.

To save labeling costs, researchers used line-label based annotation with a single straight line to mark each manatee. The goal of the study was to learn to count the number of objects within a scene and obtain labels to support counting.

Results of the study reveal that the FAU-developed method outperformed other baselines, including the traditional Gaussian kernel-based approach. Transitioning from dot to line labeling also improved wheat head counting accuracy, an important role in crop yield estimation, suggesting broader applications for convex-shaped objects in diverse contexts. This approach worked particularly well when the image had a high density of manatees in a complicated background.

By formatting manatee counting as a deep neural network density estimation learning task, this approach balanced the labeling costs vs. counting efficiency. As a result, this method delivers a simple and high throughput solution for manatee counting that requires very little labeling efforts. A direct impact is that state parks can leverage this method to understand the number of manatees in different regions, by using their existing CCTV cameras, in real time.

“There are many ways to use computational methods to help save endangered species, such as detecting the presence of the species and counting them to collect information about numbers and density,” said Xingquan (Hill) Zhu, Ph.D., senior author, an IEEE Fellow and a professor in FAU’s Department of Electrical Engineering and Computer Science. “Our method considered distortions caused by the perspective between the water space and the image plane. Since the shape of the manatee is closer to an ellipse than a circle, we used AGK to best represent the manatee contour and estimate manatee density in the scene. This allows density map to be more accurate, in terms of mean absolute errors and root mean square error, than other alternatives in estimating manatees’ numbers.” 

To validate their method and facilitate further research in this domain, the researchers developed a comprehensive manatee counting dataset, along with their source code, published through GitHub for public access at github.com/yeyimilk/deep-learning-for-manatee-counting.

“Manatees are one of the wildlife species being affected by human-related threats. Therefore, calculating their numbers and gathering patterns in real time is vital for understanding their population dynamics,” said Stella Batalama, Ph.D., dean, FAU College of Engineering and Computer Science. “The methodology developed by professor Zhu and our graduate students provides a promising trajectory for broader applications, especially for convex-shaped objects, to improve counting techniques that may foretell better ecological results from management decisions.”

Manatees can be found from Brazil to Florida and all the way around the Caribbean islands. Some species including the Florida Manatee are considered endangered by the International Union for Conservation of Nature.

Study co-authors are FAU graduate students Zhiqiang Wang; Yiran Pang; and Cihan Ulus, also a teaching assistant, all within the Department of Electrical Engineering and Computer Science. 

The research was sponsored by the United States National Science Foundation.

- FAU -

 

The left panel shows an image with dot labels of manatees and the right panel shows the density map of the image generated by applying Gaussian distributions to the labeled points.

Examples of algorithm performance with respect to different manatee densities in the scene. The first row shows original images with increasing manatee density from left to right. The second and third rows show ground-truth density map (2nd row) and predicted density map (3rd row) using dot labels. The fourth and fifth rows show ground-truth density map (4th row) and predicted density map (5th row) using line labels and generic Gaussian kernels. The sixth and seventh rows show ground-truth density map (6th row) and predicted density map (7th row) using line labels and anisotropic Gaussian kernels.

CREDIT

Florida Atlantic University

About FAU’s College of Engineering and Computer Science:

The FAU College of Engineering and Computer Science is internationally recognized for cutting-edge research and education in the areas of computer science and artificial intelligence (AI), computer engineering, electrical engineering, biomedical engineering, civil, environmental and geomatics engineering, mechanical engineering, and ocean engineering. Research conducted by the faculty and their teams expose students to technology innovations that push the current state-of-the art of the disciplines. The College research efforts are supported by the National Science Foundation (NSF), the National Institutes of Health (NIH), the Department of Defense (DOD), the Department of Transportation (DOT), the Department of Education (DOEd), the State of Florida, and industry. The FAU College of Engineering and Computer Science offers degrees with a modern twist that bear specializations in areas of national priority such as AI, cybersecurity, internet-of-things, transportation and supply chain management, and data science. New degree programs include Master of Science in AI (first in Florida), Master of Science and Bachelor in Data Science and Analytics, and the new Professional Master of Science and Ph.D. in computer science for working professionals. For more information about the College, please visit eng.fau.edu. 

 

About Florida Atlantic University:
Florida Atlantic University, established in 1961, officially opened its doors in 1964 as the fifth public university in Florida. Today, the University serves more than 30,000 undergraduate and graduate students across six campuses located along the southeast Florida coast. In recent years, the University has doubled its research expenditures and outpaced its peers in student achievement rates. Through the coexistence of access and excellence, FAU embodies an innovative model where traditional achievement gaps vanish. FAU is designated a Hispanic-serving institution, ranked as a top public university by U.S. News & World Report and a High Research Activity institution by the Carnegie Foundation for the Advancement of Teaching. For more information, visit www.fau.edu.

---

JOURNAL

Scientific Reports

DOI

10.1038/s41598-023-45507-3 

METHOD OF RESEARCH

Computational simulation/modeling

SUBJECT OF RESEARCH

Animals

ARTICLE TITLE

Counting manatee aggregations using deep neural networks and Anisotropic Gaussian Kernel

 

Study overturns conventional wisdom about wild turkey nesting survival

Peer-Reviewed Publication

NORTH CAROLINA STATE UNIVERSITY

A new study finds that precipitation levels during nesting season are not related to reproductive success for wild turkeys, which runs counter to the conventional wisdom regarding the role that rainfall plays in wild turkey nesting success. The findings shed new light on how climate change may affect wild turkey populations.

“We wanted to know how weather influences nesting success right now, and then use that data to assess how climate change may influence wild turkey populations in the future,” says Wesley Boone, corresponding author of a paper on the work and a postdoctoral researcher at North Carolina State University.

“Wild turkeys are fairly tolerant of a wide range of conditions, but there are a host of factors that can affect their reproductive success,” says Chris Moorman, co-author of the study and a professor of forestry and environmental resources at NC State. “This work focused on two of those conditions, precipitation and temperature, and how they may influence nest survival during the incubation period.”

For the study, researchers focused on daily nest survival, which is whether the eggs in the nest survive any given 24-hour period. Over the course of eight years, researchers monitored 715 turkey nests and collected daily precipitation and temperature data for each nest during the entire incubation period. For temperature, the researchers looked specifically at the extent to which temperatures at each nest varied from historical averages.

The researchers analyzed all of this data to determine the extent to which precipitation and temperature were associated with daily nest survival.

“The most surprising finding was that precipitation during nesting was not a good predictor of daily nest survival,” Moorman says. “It had been widely believed that particularly rainy weather made it more likely that eggs wouldn’t survive.”

“We also found that temperatures which were higher than historical averages were associated with higher rates of daily nest survival during incubation,” says Boone. “Peak nesting season is generally in April, so we’re talking about warmer than average spring weather.”

“Taken by itself, this might suggest that climate change could benefit turkey reproductive success and, by extension, turkey populations,” Moorman says. “However, we also looked at precipitation and temperature data for the months leading up to nesting season, and at the overall likelihood that a turkey nest will successfully hatch at least one egg. And when we looked at both of those datasets, things get a lot less clear.”

“For example, the data suggest that more precipitation in January – long before nesting season – is associated with greater nest survival,” Boone says. “The data also suggest that higher temperatures in January are associated with worse nesting survival. But there is so much uncertainty related to those findings that it’s not clear whether there’s a real relationship there, or if it’s an anomaly. However, it does temper any enthusiasm we might have about the likelihood that climate change will benefit turkey populations.”

The paper, “Robust assessment of associations between weather and eastern wild turkey nest success,” is published open access in The Journal of Wildlife Management. The paper was co-authored by David Moscicki, a Ph.D. student at NC State; Krishna Pacifici, an associate professor of forestry and environmental resources; Adam Terando, a research ecologist with the U.S. Geological Survey; Bret Collier, a professor of wildlife ecology at Louisiana State University; and Michael Chamberlain, the Terrell Professor of Wildlife Ecology and Management at the University of Georgia.

The research was done with support from the U.S. Geological Survey’s Southeast Climate Adaptation Science Center, which is headquartered at NC State; and from the National Institute of Food and Agriculture, under McIntire Stennis Project Number 7001494. Additional support was provided by the Georgia Department of Natural Resources-Wildlife Resources Division, the Louisiana Department of Wildlife and Fisheries, the South Carolina Department of Natural Resources, the North Carolina Wildlife Resources Commission, the National Wild Turkey Federation, the United States Department of Agriculture’s Forest Service, the Warnell School of Forestry and Natural Resources at the University of Georgia and the School of Renewable Natural Resources at Louisiana State University.

---

JOURNAL

Journal of Wildlife Management

DOI

10.1002/jwmg.22524 

METHOD OF RESEARCH

Observational study

SUBJECT OF RESEARCH

Animals

ARTICLE TITLE

Robust assessment of associations between weather and eastern wild turkey nest success

 

Traces of trauma in the young brain – and how to erase them

Weizmann Institute researchers reveal in mice how exposure to trauma in infancy alters the brain; they show that early treatment to reverse these changes is vital for rehabilitation

Peer-Reviewed Publication

WEIZMANN INSTITUTE OF SCIENCE

The images of Israeli child hostages being freed from Hamas captivity are heartwarming, but for most of these children, the release is just the start of a long rehabilitation process. Countless studies have shown that exposure to warfare, abuse and other traumatic events at a young age significantly raises the risk of ill health, social problems and mental health issues later in life. Now, a new study by researchers at the Weizmann Institute of Science provides a reason for optimism. In research conducted on mice, published Friday in Science Advances, a team headed by Prof. Alon Chen discovered brain mechanisms that go awry as a result of exposure to trauma in infancy and showed that these changes may be reversible if treated early. 

Our brains have a wonderful quality known as plasticity, the ability to change throughout our lives. As may be expected, in our early years, when the brain is still developing, it is at peak plasticity. This manifests in, for example, the aptitude for learning languages, but this also entails a heightened sensitivity to traumatic events, which are liable to leave a scar that only intensifies with age. Many studies provide evidence for the latter effect, but very little is known about the way that exposure to trauma at a young age affects the different kinds of brain cells and the communication between them in adulthood.

Chen’s laboratory in Weizmann’s Brain Sciences Department focuses on the molecular and behavioral aspects of the response to stress. In previous studies, Chen’s team examined how stress during pregnancy affects mouse offspring when they reach maturity. In the current research, the scientists, led by Dr. Aron Kos, studied how trauma experienced shortly after birth affects mouse pups later in life. To advance the understanding of this topic, the researchers pulled together the strengths of Chen’s lab: its expertise in exploring the brain’s molecular processes at the highest possible resolution, using genetic sequencing on the level of individual cells; the ability to use cameras to track dozens of behavioral variables in a rich social environment intended to recreate natural living conditions; and the ability to process the massive quantities of data generated in this environment, using machine learning and artificial intelligence tools.

This comprehensive behavioral mapping revealed that mice exposed after birth to a traumatic event – in the case of this study, being neglected by their mothers – displayed a variety of behaviors indicating that they found themselves at the bottom of the dominance hierarchy. “Equivalent behaviors in humans might include high levels of introversion, social anxiety and having an avoidant personality, all known to be characteristic of posttrauma,” says Dr. Juan Pablo Lopez, a former postdoctoral fellow in Chen’s joint laboratory at Weizmann and the Max Planck Institute of Psychiatry in Munich, and today head of a research group in the Department of Neuroscience at the Karolinska Institute in Stockholm. 

In the next stage of the study, the researchers exposed some of the adult mice that had experienced trauma in infancy to a stressful social situation: bullying by other mice. Ultimately, they created four groups of adult mice: those that had not been exposed to any trauma; those that had not been exposed to trauma in infancy but were subjected to bullying as adults; mice that were exposed to trauma only in infancy; and mice that were exposed to both trauma in infancy and bullying as adults. To find out how exposure to early trauma disrupts the brain and what happens as a result of this in adulthood, the researchers carried out a meticulous comparison of the four groups, using RNA sequencing at the single-cell level in the hippocampus, a brain area known to play an important role in social functioning. The comparison revealed that early trauma left a mark on different types of cells, primarily affecting gene expression in two subpopulations of neurons, those belonging to the glutamatergic excitatory system and those belonging to the GABA inhibitory system. This effect was especially strong in mice that had been exposed to both trauma in infancy and bullying as adults.

Cells in the brain communicate with one another by means of electrical signals, which can be excitatory, that is, stimulating, or inhibitory. An excitatory signal encourages communication between brain cells, whereas an inhibitory signal represses it, like the gas and brake pedals in a car. Normal brain functioning requires a balance between the excitatory and inhibitory signals, which is lacking in many psychiatric disorders. One of the ways of assessing the brain’s electrical activity and the balance between excitatory and inhibitory signals is through electrophysiological measurements. Such measurements, performed in the hippocampus of the mice by Dr. Julien Dine, a former staff scientist at the Weizmann Institute and currently a pharmaceutical electrophysiologist, supported the molecular findings: Exposure to trauma in early childhood disrupted the balance between excitatory and inhibitory signals in adulthood.

Having discovered a brain mechanism that is disrupted in adulthood as a result of early trauma – and having identified this disruption as an imbalance between the excitatory and inhibitory signals – the researchers tried to find a way to fix it. During a brief treatment window shortly after the early trauma, they gave the mice a well-known antianxiety drug – diazepam, known commercially as Valium – which affects the GABA inhibitory system. This short course of treatment led to results that were nothing less than stunning: The treated mice were able to fully or almost fully avoid the behavioral future that awaited them and were no longer at the foot of the social ladder. “Understanding the molecular and functional mechanisms allowed us to neutralize the negative behavioral impact of trauma with a drug given shortly after exposure to traumatic incidents,” Kos explains. “This certainly should not be seen as a recommendation to treat young trauma patients with drugs, but our findings do highlight the importance of early treatment for successful rehabilitation.”

Intense, ongoing stress can, at any age, contribute to disease, from psychiatric disorders to obesity and diabetes. But in the first years of life – and also in the womb – such stress can have dramatic ramifications. “The wars in Israel, Ukraine, Sudan and elsewhere, and the unprecedented global refugee crisis that is caused, in part, by climate change, alongside an increased understanding of the long-term harm caused by exposure to war and violence at a young age – all these highlight the need for better rehabilitation capabilities,” says Chen. “Our new study identifies a key brain mechanism that is especially sensitive to childhood trauma. But the most exciting part is the prospect of using the plasticity of the young brain to help it recover, avoiding the toll this trauma can exact in adulthood.”

Also participating in the study were Dr. Joeri Bordes, Carlo de Donno, Dr. Elena Brivio, Stoyo Karamihalev, Dr. Alec Dick, Lucas Miranda, Rainer Stoffel, Cornelia Flachskamm and Dr. Mathias V. Schmidt from the Max Planck Institute of Psychiatry; Dr. Malte D. Luecken, Dr. Maren Büttner and Prof. Fabian J. Theis from the Helmholtz Zentrum München; Dr. Suellen Almeida-Correa from Weizmann’s Brain Sciences Department; and Serena Gasperoni from the Karolinska Institute.

Prof. Alon Chen holds the Vera and John Schwartz Professorial Chair in Neurobiology. His research is supported by the Ruhman Family Laboratory for Research on the Neurobiology of Stress; and by the Licht Family.

 

 

---

JOURNAL

Science Advances

DOI

10.6084/m9.figshare.24420802.v1 

ARTICLE TITLE

Early life adversity shapes social subordination and cell type–specific transcriptomic patterning in the ventral hippocampus

 U$A

Job outlook for college graduates to hold steady in 2024

Reports and Proceedings

DREXEL UNIVERSITY

The job market for college graduates in 2024 will be as dynamic as it is promising, according to a new report from job market analysts from Drexel University and Michigan State University. The 53rd Annual College Hiring Outlook report, produced by Drexel’s LeBow College of Business, Michigan State’s Collegiate Employment Research Institute, with the employment agency ROCS Grad Staffing, sheds light on the trends and challenges shaping this year's hiring projections.

Over 1,100 organizations across the U.S. responded to a survey about college student recruitment and hiring practices. The resulting report indicates a resilient job market, especially for those with bachelor's degrees, with a 5.3% increase in hiring opportunities compared to the previous year. The demand for advanced degrees, like MBAs, is experiencing a slight downturn, suggesting a shift in employer priorities, according to the report.

"2024 is a year of significant opportunities for college graduates," said Murugan Anandarajan, PhD, senior associate dean for Academic Programs and Faculty Affairs and professor in LeBow College, who was a coauthor of the report with Michigan State’s Phil Gardner, PhD. "The report highlights the growing emphasis on practical skills and experience over traditional academic qualifications. This trend opens new pathways for graduates, encouraging a more diverse skill set and adaptability in their career journey."

One of the most notable trends is the increased demand for candidates with specific credentials and certificates, which is seeing a 4.8% increase in hiring. This shift toward specialized skills indicates a broader change in the employment landscape.

Regional hiring patterns also offer interesting insights, according to the analysts. While there is a consistent demand for bachelor-level candidates across the U.S., graduates with advanced degrees are finding a more restrained market. The industries leading the charge in hiring include construction, education and health services. Sectors like finance and insurance are cautiously approaching recruitment.

The report also explored the burgeoning role of artificial intelligence in recruitment, with a focus on ChatGPT. The findings suggest a cautious approach toward AI adoption in hiring processes. Nearly two-thirds of organizations are not considering the use of ChatGPT at present, with only a quarter contemplating its adoption. 

"While AI presents remarkable hiring efficiencies, its integration into recruitment processes is progressing slower than anticipated,” Anandarajan said. “This hesitancy reflects a cautious approach towards balancing technology with the human aspects of recruitment. However, as we move forward, the adoption of AI in hiring will become more pronounced, reshaping how companies identify and attract talent."

As the hiring landscape for college graduates evolves, the 2024 outlook offers both challenges and opportunities. Graduates are advised to remain agile, leveraging tools like AI judiciously while staying attuned to the changing economic and employment dynamics.

Anandarajan encourages readers to use the report as one of many sources to understand the labor market for graduates. "Our report, alongside others like the NACE outlook, provides a comparative perspective on hiring trends, especially from large employers," he said.

This report also covers the outlook for college talent hiring by region and academic discipline. Starting salary expectations and internships/co-ops/early talent programs are also discussed. The final section provides employer feedback about college recruiting strategies and recruiting platforms. To read the full report, click here.

 

Inequity in U.S. wildfire emergency response

Reports and Proceedings

SOCIETY FOR RISK ANALYSIS

WASHINGTON, DC, 2023 – Recent U.S. wildfire events -- including the 2023 Maui wildfire in Hawaii, the 2022 Hermit’s Peak/Calf Canyon fire in New Mexico, and the 2020 Cameron Peak Fire in Colorado -- are tragic examples of how disadvantaged communities can suffer most during and after a wildfire. While all three fires had a devastating impact on an entire community, they disproportionately affected low-income populations who were left without adequate insurance or the financial means to rebuild their homes. 

To study inequities in U.S. wildfire management, scientists at the State University of New York at Buffalo are conducting a data-driven assessment of how socioeconomic variables affect the allocation of resources during a wildfire incident. Specifically, they are investigating how socioeconomic variables (such as income and racial demographics) impact the number of personnel dispatched to a community post-wildfire and the dollars spent on extinguishing the fire.   

They will present their preliminary results at the 2023 Society for Risk Analysis Annual Conference in Washington, D.C.  

In their ongoing study, Industrial and Systems Engineering professor Sayanti Mukherjee and M.S. student Fatima Umar have collected data from the U.S. Census Bureau on county-level socioeconomic indicators like average income and racial demographic composition and obtained detailed information on more than 230,000 wildfire events nationwide from 2014 to 2022 (provided by the National Interagency Fire Center). 

“Our results show a pronounced trend in which counties with higher percentages of lower-income and black populations receive less personnel and funding,” says Mukherjee. “Conversely, counties with higher proportions of high-income and white people are more likely to secure significant amounts of these resources.” 

The researchers have also observed that the percentage of households with annual income over $200,000 positively correlates with the number of personnel and the estimated cost of putting out a fire. “This indicates that high-income neighborhoods receive more attention in the wildfire disaster response and recovery phase,” says Umar. 

Based on their preliminary findings, Mukherjee and Umar suggest that equity-informed risk analysis should be incorporated into disaster response planning to provide a more accurate assessment of the wildfire-induced risk in a community. “This would enable government officials to work towards reducing disparities in wildfire management and response,” says Mukherjee. “By considering marginalized communities’ unique needs and vulnerabilities, disaster response efforts can be more equitable and just.” 

Their research is funded by the National Science Foundation (NSF)’s Strengthening American Infrastructure (SAI) program (NSF Award # 2324616). 

### 

Presentations are at the Westin Washington DC.  

Data-Driven Analysis of Equity in Wildfire Resource Allocation – Wednesday, December 13, 9:10-9:30 a.m.  

About SRA   

The Society for Risk Analysis is a multidisciplinary, interdisciplinary, scholarly, international society that provides an open forum for all those interested in risk analysis. SRA was established in 1980. Since 1982, it has continuously published Risk Analysis: An International Journal, the leading scholarly journal in the field. For more information, visit www.sra.org.   

---

JOURNAL

Risk Analysis

 

Helping more people get to safety in a wildfire

Scientists have developed a web-based tool to help communities design an optimal wildfire evacuation plan

Reports and Proceedings

SOCIETY FOR RISK ANALYSIS

WASHINGTON, DC, Dec. 13, 2023 – Wildfires pose an increasing threat to communities at the wildland-urban interface (WUI) – where dry, flammable vegetation borders back yards, often in remote locations. Despite the well-known danger, many communities at highest risk do not have a strong wildfire evacuation plan in place. (One of these was the town of Lahaina on Maui, where wind-driven wildfires killed nearly 100 people in August 2023.) 

Researchers from UCLA’s John Garrick Institute for the Risk Sciences have built a new web-based software platform that allows emergency planners to design custom-made evacuation plans for their communities that will help more people safely escape an oncoming wildfire. The results of a first test of the tool, using the deadly 2018 Camp Fire in California as an example, will be presented by Mohammad Pishahang at the annual meeting of the Society for Risk Analysis (SRA), Dec. 11-13 in Washington, D.C. 

 

The Wildfire Safe Egress planning (WISE) simulation platform is a visual, Geographic Information Systems (GIS) environment that allows users to simulate several wildfire scenarios on a map and calculate the likelihood of the safe evacuation of a population. It incorporates data on the demographics and road network of a community and allows users to input specific factors such as fire dynamics, community boundaries, warning system parameters, and shelter locations. It is the first wildfire evacuation planning tool to also include several factors affecting human behavior in emergency situations such as age, income level, English proficiency, disabilities, and vehicle ownership (all of which impact a household’s decision to evacuate).  

The Deadly Camp Fire of 2018 

To demonstrate WiSE and its capabilities, Pishahang and his colleagues recently simulated an evacuation of the entire city of Paradise during California’s deadliest wildfire: the Camp Fire of 2018, which killed 85 people.  

 

For the simulation, they set the “awareness trigger time” (one of the data inputs) to just one hour before the fire entered the city, and they assumed that the warning system completely failed. “This is consistent with the real-life events in which the official evacuation warnings failed to reach the population successfully,” says Pishahang.  

 

In this simulation, the researchers found that only 16 percent of the at-risk community population had the opportunity to safely exit the danger zone. Consequently, 84 percent of the population evacuated while already facing the fire and being surrounded by it (not considered to be a safe evacuation).  These results were qualitatively validated by the firefighters who served in the Camp Fire disaster. 

When they shortened the pre-evacuation time (i.e. indicating faster decision-making and action by the at-risk people) and accounted for a better (working) warning system in a second simulation, they discovered that: 

• if the community had 2 hours to exit the area in the same situation, more than 40 percent of the population could have left the danger zone before the fire reached the borders of the city; and 

• if only 20 percent of the at-risk people had been made aware of the need to evacuate by an official warning system one hour before the fire reached the city borders, still more than 44 percent of the residents would have had the opportunity to safely exit the danger zone. 

“These findings demonstrate how WiSE can help emergency managers design the best possible evacuation plan and indicate the best evacuation preparation strategies for their community that are based on data and maximize the number of people who can safely escape an oncoming wildfire,” says Pishahang. 

For each wildfire scenario, the program identifies main bottlenecks and traffic congestion points so planners can address these issues before a natural disaster occurs. 

“Many existing WUI communities do not sufficiently meet evacuation-related travel needs, including suburbs built with only one road in and out,” says Pishahang. “WiSE can help planners see how to decrease a community’s vulnerability by increasing roads traffic capacity in addition to improving the reliability of communication and warning systems and educating residents on how to follow evacuation orders in event of an emergency.”  

### 

Presentations are located at The Westin Washington, D.C. 

WiSE: Wildfire Safe Evacuation Planning and Management – Wednesday, December 13, 8:30-8:50 a.m. 

About SRA
The Society for Risk Analysis is a multidisciplinary, interdisciplinary, scholarly, international society that provides an open forum for all those interested in risk analysis. SRA was established in 1980. Since 1982, it has continuously published Risk Analysis: An International Journal, the leading scholarly journal in the field. For more information, visit www.sra.org.   

---

JOURNAL

Risk Analysis

 

Amsterdam UMC to use AI to increase the accessibility of medical imaging technology

The Dutch Research Council have awarded Amsterdam UMC with six million euros, to be used to increase the accessibility of imaging technology through the use of AI

Grant and Award Announcement

AMSTERDAM UNIVERSITY MEDICAL CENTERS

The demand for acquiring and interpreting medical images is increasing faster than number of medical experts required to operate the medical imaging device and interpret their output. This is leading to an increase in the expert workload and extending waiting lists. An Amsterdam UMC-led consortium wants to tackle this problem by making imaging technology more accessible. With the help of artificial intelligence, they want to enable less specialized experts to acquire and analyse medical images. To support them in this goal, the Dutch Research Council (NWO) has awarded the AI4AI project a grant of 6.1 million euros. 

Medical images are mainly captured in the hospitals, using expensive imaging devices such as CT or MRI-scanners. Specialists are required operate these devices as well as analyse their output. "Our aim is to use artificial intelligence to develop technologies that allow the use of affordable and/or portable devices such as ultrasound and ultra-low-field MRI," says Amsterdam UMC Professor of Artificial Intelligence and Medical Imaging Ivana Išgum. Išgum is the coordinator of the national consortium implementing the AI4AI project. "Also, our aim is to allow the use of imaging devices by e.g. general practitioners, sonographers and specialist nurses to reduce the need for very highly specialized experts." 

Less pressure on staff 

AI technology can reduce the need for highly specialized experts for operation of medical devices and analysis of medical images, which has the potential to greatly reduce the pressure on personnel and the associated costs" says Išgum, who together with Professor Clarisa Sánchez, leads the interfaculty research group qurAI that focuses on the responsible development of AI in healthcare. 

Shortening waiting lists  

The increased demand for medical images means that the workload for radiologists and other specialists is increasing enormously. This can lead to burnout symptoms, which in turn affects both the sustainability of care and waiting lists. This means that patients might need to travel further to receive the necessary care. "With this project, we want to contribute to bringing medical imaging closer to patients’ living environment and make it more accessible for patients. In addition, hospital care in developing countries may not always be accessible to everyone. There may also be fewer highly skilled experts available. We also hope to contribute to more accessible healthcare for people in these countries," says Išgum. 

Amsterdam UMC Radiologist Nils Planken adds that many fellow radiologists and other medical specialists welcome the support of technology. "AI technology that can support the creation, interpretation and reporting of medical imaging studies has the potential to shorten waiting lists and reduce workload and perhaps also improve quality. The correct use of diagnostics outside the hospital has the potential to prevent patients from being sent to the hospital, or to sending patients to the hospital in an even more targeted way," says Planken. 

Applications  

AI4AI focuses on many diseases and specialties, such as analysis of stroke and brain tumours, visualization and interpretation of organ tissue perfusion in surgery, quantification of foetal biomarkers to spot abnormalities in the pregnancy, identification of patients requiring invasive coronary artery treatment, identification of patients with heart disease, improvement of the workflow in image-guided radiotherapy, referrals for urgent care, screening and triage of threatening visual disorders, selection of patients eligible for immunotherapy, and improvement of imaging workflow to assess orthopaedic implants. 

 

When parents drink during Super Bowl, kids get harsh discipline

Alcohol use affected both mothers and fathers, study finds

Peer-Reviewed Publication

OHIO STATE UNIVERSITY

COLUMBUS, Ohio – Parents who drank alcohol while watching the Super Bowl were more likely than those who abstained to use aggressive discipline on their children during the game, a new study shows.

 

Most of the parents in the study – more than 90% - were mothers, which is significant, said Bridget Freisthler, lead author of the study and professor of social work at The Ohio State University.

 

“The links between alcohol use, aggression and watching violent sports have been studied almost exclusively among men,” Freisthler said.

 

“This is the first study we’re aware of that shows women may also be affected to act more aggressively by the combination of alcohol and watching violent sports like pro football.”

 

The study also looked at use of alcohol among parents during another special occasion – Valentine’s Day. Results showed that parents who drank on Valentine’s Day were actually less likely to use aggressive discipline on their children that day than were parents who didn’t drink.

 

The study was published today (Dec. 13, 2023) in the journal Alcoholism: Clinical and Experimental Research.

 

Most studies that examine alcohol use tend to focus on people’s typical drinking habits. But special occasions like holidays, weddings and big sporting events may be anything but typical, Freisthler said.

 

“When you ask about typical drinking behavior, people may say they usually only have one beer a day,” she said.  “But on the Fourth of July they may say they have four beers.  That could be a big difference.”

 

This study was part of a larger project that asked parents in central Ohio to report on their alcohol use and parenting techniques three times a day for 14 days. For this study, the researchers included parents who participated during the Super Bowl in February 2021 (255 participants) and one week later on Valentine’s Day (184 participants).

 

All the parents in the study had a child who was 2 to 12 years old at the time.

 

During the three times on Super Bowl Sunday and Valentine’s Day (and on the rest of the 14 days in the study) parents were asked whether they were drinking alcohol and whether they were using punitive or aggressive discipline with their child.

 

That could include discipline such as spanking or shaking the child, or shouting or yelling at them – not just to stop a behavior but to call them a name or shame them. This includes behaviors that are less severe than official child abuse or neglect, yet more frequent.

 

The study didn’t find any relationship between alcohol use and aggressive discipline on most days studied – but the results of the Super Bowl and Valentine’s Day stood out, according to Freisthler.

 

“That’s why we think drinking on special occasions deserves more attention,” she said.

 

There are many reasons why the use of alcohol during the Super Bowl may make parents – particularly mothers – more likely to use aggressive discipline, Freisthler said.

 

“When you add stress and alcohol, that is not a good combination. There’s the stress of the game, particularly if you’re invested in one of the teams. If mothers are hosting a Super Bowl party, that’s another level of stress,” she said.

 

“And for mothers, if their husbands are invested in the game, they may feel it is their job to keep the kids quiet and out of the way of the TV.”

 

The reason that Valentine’s Day resulted in less use of aggressive discipline, despite the alcohol use, may be because of the different nature of that holiday compared to the Super Bowl, according to Freisthler.

 

On Valentine’s Day, the parents may be more likely to be drinking at a restaurant, away from their children. They may feel less stress than normal because of the nature of the romantic holiday, and alcohol use could enhance the good feelings – all of which could lead to less harsh discipline.

 

But Freisthler noted that there are probably more special occasions throughout the year that resemble the Super Bowl than Valentine’s Day.  Holidays like Christmas and the Fourth of July often involve events at home with large groups of people and children around, and also involve more alcohol use than usual.

 

“We need to understand how much parents are drinking on special occasions, how that differs from their normal drinking behaviors, and how is that related to their parenting,” she said. “That’s what we are trying to get at in this study.”

 

Knowing that drinking on special occasions affects levels of drinking – and parenting – means that parents can take steps beforehand to minimize negative impacts, she said.

 

For example, during events like the Super Bowl, parents could hire babysitters if they are able. They could have a special room set up for the kids with fun activities. Or maybe some people who are not interested in the game, like possibly grandparents, could entertain the children during the game.

 

“Parents need to create environments that are most conducive to positive parenting and reduce the risk of harsh parenting,” she said.

 

Freisthler conducted the study with Joselyn Sarabia, a doctoral student in social work at Ohio State, and Jennifer Price Wolf, associate professor of social work at San Jose State University.

---

JOURNAL

Alcoholism Clinical and Experimental Research

METHOD OF RESEARCH

Observational study

SUBJECT OF RESEARCH

People

ARTICLE TITLE

Using ecological momentary assessments to understand how drinking during special occasions relates to parenting behaviors

ARTICLE PUBLICATION DATE

13-Dec-2023

U$A

Reported drug use among adolescents continued to hold below pre-pandemic levels in 2023

New data show relatively low use of illicit substances, yet overdose death rates among teens have risen in recent years

Reports and Proceedings

NIH/NATIONAL INSTITUTE ON DRUG ABUSE

The percentage of adolescents reporting they used any illicit substances in 2023 continued to hold steady below the pre-pandemic levels reported in 2020, with 10.9% of eighth graders, 19.8% of 10th graders, and 31.2% of 12th graders reporting any illicit drug use in the past year, according to the latest results from the Monitoring the Future survey. Reported use for almost all substances decreased dramatically between 2020 and 2021, after the onset of the COVID-19 pandemic and related changes like school closures and social distancing. In 2022, most reported substance use among adolescents held steady at these lowered levels, and these latest data show that this trend has continued into 2023.

The Monitoring the Future survey is conducted by researchers at the University of Michigan, Ann Arbor, and funded by the National Institute on Drug Abuse (NIDA), part of the National Institutes of Health. 

The 2023 data continue to document stable or declining trends in the use of illicit drugs among young people over many years. However, importantly, other research has reported a dramatic rise in overdose deaths among teens between 2010 to 2021, which remained elevated well into 2022 according to a NIDA analysis of CDC and Census data. This increase is largely attributed to illicit fentanyl, a potent synthetic drug, contaminating the supply of counterfeit pills made to resemble prescription medications. Taken together, these data suggest that while drug use is not becoming more common among young people, it is becoming more dangerous.

“Research has shown that delaying the start of substance use among young people, even by one year, can decrease substance use for the rest of their lives. We may be seeing this play out in real time,” said Nora Volkow, M.D., NIDA director. “This trend is reassuring. Though, it remains crucial to continue to educate young people about the risks and harms of substance use in an open and honest way, emphasizing that illicit pills and other substances may contain deadly fentanyl.”

The Monitoring the Future survey is given annually to students in eighth, 10th, and 12th grades who self-report their substance use behaviors over various time periods, such as past 30 days, past 12 months, and lifetime. The survey also documents students’ perceptions of harm, disapproval of use, and perceived availability of drugs. The survey results are released the same year the data are collected. From February through June 2023, the Monitoring the Future investigators collected 22,318 surveys from students enrolled across 235 public and private schools in the United States.

When breaking down the data by specific drugs, the survey found that adolescents most commonly reported use of alcohol, nicotine vaping, and cannabis in the past year, and levels generally declined from or held steady with the lowered use reported in 2022. Compared to levels reported in 2022, data reported in 2023 show: 

• Alcohol use remained stable for eighth and 10th graders, with 15.1% and 30.6% reporting use in the past year respectively, and declined for 12th graders, with 45.7% reporting use in the past year (compared to 51.9% in the previous year).
• Nicotine vaping remained stable for eighth graders, with 11.4% reporting vaping nicotine in the past year. It declined in the older grades, from 20.5% to 17.6% in 10th grade and from 27.3% to 23.2% in 12th grade. 
• Cannabis use remained stable for all three grades surveyed, with 8.3% of eighth graders, 17.8% of 10th graders, and 29.0% of 12th graders reporting cannabis use in the past year. Of note, 6.5% of eighth graders, 13.1% of 10th graders, and 19.6% of 12th graders reported vaping cannabis within the past year, reflecting a stable trend among all three grades.
• Delta-8-THC (a psychoactive substance found in the Cannabis sativa plant, of which marijuana and hemp are two varieties) use was measured for the first time in 2023, with 11.4% of 12th graders reporting use in the past year. Beginning in 2024, eighth and 10th graders will also be asked about Delta-8 use.
• Any illicit drug use other than marijuana also remained stable for all three grades surveyed, with 4.6% of eighth graders, 5.1% of 10th graders, and 7.4% of 12th graders reporting any illicit drug use other than marijuana in the past year. These data build on long-term trends documenting low and fairly steady use of illicit substances reported among teenagers – including past-year use of cocaine, heroin, and misuse of prescription drugs, generally.
• Use of narcotics other than heroin (including Vicodin, OxyContin, Percocet, etc.) decreased among 12th graders, with 1.0% reporting use within the past year (matching the all-time low reported in 2021 and down from a high of 9.5% in 2004).
• Abstaining, or not using, marijuana, alcohol, and nicotine increased for 12th graders, with 62.6% reporting abstaining from any use of these substances over the past month. This percentage remained stable for eighth and 10th graders, with 87.0% and 76.9% reporting abstaining from any use of marijuana, alcohol, and nicotine over the past month.

“For 49 consecutive years, the Monitoring the Future survey has closely monitored shifts in substance use trends among young people, helping to identify and track emerging trends with public health relevance,” said Richard A. Miech, Ph.D., team lead of the Monitoring the Future study at the University of Michigan. “This year’s addition of a measure to assess use of delta-8 is one example of this, and we were surprised to see use levels this high among 12th graders. Moving forward, as policies and access to this drug change, it will be important to continue to monitor use of this drug among teens.”

Though the sample size of 22,318 respondents in 2023 was lower than the sample size of a typical pre-pandemic year’s data collection, the results were gathered from a nationally representative sample, and the data were statistically weighted to provide national numbers. This year, 13% of students who took the survey identified as Black or African American, 1% as American Indian or Alaska Native, 3% as Asian, 25% as Hispanic, 1% as Middle Eastern, 41% as white, and 16% as more than one of the preceding categories. The survey also asks respondents to identify as male, female, other, or prefer not to answer. For the 2023 survey, 48% of students identified as male, 47% identified as female, 1% identified as other, and 4% selected the “prefer not to answer” option. 

The Monitoring the Future investigators note that schools opt-in to participate in the survey, and some schools that had historically participated opted-out in the years following the onset of the COVID-19 pandemic. All participating students took the survey via the web – either on tablets or on a computer – with 98% of respondents taking the survey in-person in school in 2023. Therefore, students with less engagement in school – a known risk factor for drug use – may have been less likely to participate in the survey.

The 2023 Monitoring the Future data tables highlighting the survey results are available online from the University of Michigan. 

For more information on substance and mental health treatment programs in your area, call the free and confidential National Helpline 1-800-662-HELP (4357) or visit www.FindTreatment.gov. 

###

About the National Institute on Drug Abuse (NIDA): NIDA is a component of the National Institutes of Health, U.S. Department of Health and Human Services. NIDA supports most of the world’s research on the health aspects of drug use and addiction. The Institute carries out a large variety of programs to inform policy, improve practice, and advance addiction science. For more information about NIDA and its programs, visit www.nida.nih.gov.

About the National Institutes of Health (NIH): NIH, the nation’s medical research agency, includes 27 Institutes and Centers and is a component of the U.S. Department of Health and Human Services. NIH is the primary federal agency conducting and supporting basic, clinical, and translational medical research, and is investigating the causes, treatments, and cures for both common and rare diseases. For more information about NIH and its programs, visit www.nih.gov.

About substance use disorders: Substance use disorders are chronic, treatable conditions from which people can recover. In 2022, nearly 49 million people in the United States had at least one substance use disorder. Substance use disorders are defined in part by continued use of substances despite negative consequences. They are also relapsing conditions, in which periods of abstinence (not using substances) can be followed by a return to use. Stigma can make individuals with substance use disorders less likely to seek treatment. Using preferred language can help accurately report on substance use and addiction. View NIDA’s online guide.

NIH…Turning Discovery Into Health®