Thursday, April 25, 2024

 

A vaccine to fight antibiotic resistance


MSU, Harvard Medical School team up to expand vaccine science’s role in the fight against MRSA and other infections



MICHIGAN STATE UNIVERSITY

MRSA 

IMAGE: 

AN ARTISTIC RECREATION OF MRSA. MICHIGAN STATE UNIVERSITY RESEARCHER XUEFEI HUANG IS DEVELOPING NEW VACCINE SCIENCE IN THE FIGHT AGAINST ANTIBIOTIC RESISTANCE.

view more 

CREDIT: JENNIFER OOSTHUIZEN, MEDICAL ILLUSTRATOR. CENTERS FOR DISEASE CONTROL AND PREVENTION PUBLIC HEALTH IMAGE LIBRARY.





Driven by the overuse of antimicrobials, pathogens are quickly building up resistances to once-successful treatments. It’s estimated that antimicrobial-resistant infections killed more than 1 million people worldwide in 2019, according to the World Health Organization.   

“There are worries that at the rate things are going, in perhaps 20 or 30 years, few of our drugs will be effective at all,” said Xuefei Huang, a Michigan State University Research Foundation Professor in the departments of Chemistry and Biomedical Engineering.  

“This would bring us back to the pre-antibiotic age.” 

Now, in a new Nature Communications study, Huang and his collaborators have reported a breakthrough that will help tackle this global threat head-on. Specifically, the team has created a promising vaccine candidate for antibiotic-resistant bacteria.  

Bacterial vaccines, along with antibiotics, are a crucial tool in the fight against deadly microbes.  

In the latest paper, Huang announced several discoveries that will help the development of a carbohydrate-based vaccine for infections caused by Staphylococcus aureus and its “superbug” relative methicillin-resistant Staphylococcus aureus, or MRSA. 

Staph aureus, or staph, and MRSA are among the most prevalent causes of bacterial infections. 

Using an innovative delivery platform created by the Huang group at MSU, the team’s preclinical vaccine formulation offered high levels of immunity from lethal levels of staph and MRSA in animal trials.  

With this work, Huang and his team have expanded the frontiers of vaccine science, equipping fellow researchers with new knowledge to improve and refine future bacterial vaccines.  

Carbohydrate hurdles 

To develop a vaccine, researchers must identify an effective antigen. This is a substance or molecule that the body flags as foreign, helping to trigger an immune response and the creation of antibodies that will fight future infection. 

While most vaccines rely on protein antigens, Huang is an expert in the chemistry and biology of carbohydrates. These are chemical compounds comprised of saccharides, or sugars.  

Developing carbohydrates to use as antigens in vaccines comes with its own unique challenges and advantages. 

“Sugar structures are very specific to certain bacteria,” Huang explained. “A vaccine that works against one bacterium might not work at all against another, even if they’re very similar.”  

This is why a single dose of a bacterial vaccine can contain many different antigens. For instance, the “20” in Pfizer’s PREVNAR 20 pediatric pneumonia vaccine refers to the 20 unique strains of bacteria it protects against. 

If researchers can develop an antigen that’s shared among many — if not all — bacteria, vaccination coverage would be greatly improved.  

Gerald Pier, professor of medicine at Harvard Medical School and Brigham and Women’s Hospital and a collaborator on the latest MSU-led paper, has studied one such antigen candidate for years.  

Polysaccharide poly-β-(1−6)-N-acetylglucosamine, or PNAG, is a carbohydrate found on the cell wall of staph, many other bacteria and even fungi. This prevalence makes it extremely useful, offering potential protection against numerous pathogens at once. 

By examining PNAG as an antigen candidate for staph, Pier, Huang and their colleagues are unlocking the secrets needed to make a more effective vaccine. 

A molecular mosaic 

Imagine creating a mosaic made from multicolored tiles. 

Arrange these tiles in a precise pattern and you’ll end up with a striking work of art. Move just a few tiles around, however, and you’ll find yourself looking at a very different image.  

PNAG — and carbohydrates in general — are kind of like mosaics. There are myriad ways to arrange their individual pieces, but only a select few have the effects that researchers desire. 

Just as changing a few tiles in a mosaic can give you a completely different image, swapping out these pieces or even changing their location within a PNAG molecule changes its performance as a potential antigen. 

“We were very interested in this molecule and these different patterns,” Huang said. 

“We wanted to know: Was there a best combination to improve Staph aureus vaccine efficiency, and does the arrangement matter?” 

The pieces that Huang and his colleagues were most interested in were biologically active molecular components known as amines and acetyl groups that adorn PNAG’s sugary backbone. 

PNAG molecules can contain many amines. These amines can be acetylated, meaning they’re modified with an acetyl group, or they can be free and not bound to anything else.  

Currently, most researchers investigating PNAG as an antigen focus on forms of the sugar that are either fully free or fully acetylated. 

Huang and his colleagues believed there were promising opportunities in the understudied in-between space where there’s a mixture of free and acetylated amines. 

For its research, the team created a library of 32 different PNAG structures. The structures were all pentasaccharides — made from five saccharides — but they differed in how they were decorated with amines and acetyl groups.  

By screening these 32 structures with antibody studies, they made their discovery.  

“The fine pattern matters quite a bit,” Huang said. “And the impact is drastic.” 

An MSU mutant 

The team identified two PNAG combinations that were especially promising. Going a step further, the researchers attached them to a groundbreaking vaccine delivery platform. 

The platform is based on a bacteriophage, which is a virus that infects bacteria, called Qbeta, also written as Qβ (pronounced “cue beta”). Huang’s team modified the bacteriophage, giving it the power to deliver antigens for carbohydrate-based pathogens.  

PNAG and other carbohydrates typically don’t provoke strong immune responses in our bodies, but the mutant Qbeta, or mQβ, helps create an enhanced reaction. 

This breakthrough delivery platform — which also has vaccine applications for cancer and even opioid addiction — earned Huang MSU’s 2024 Technology Transfer Achievement Award. 

When coupled with mQβ, Huang and his collaborators found that the two most promising PNAG pentasaccharides offered high levels of protection in mice against staph and MRSA.  

In animal studies, the team’s new vaccine construct outperformed another PNAG-vaccine delivery system that is currently in human trials. 

The team also found their formulation had minimal impact on the biochemistry of the gut microbiome in tests.  

As the team prepares for future tests of their new vaccine candidate, Huang is looking forward to the role bacterial vaccines will play in the larger fight against antibiotic resistance.  

“Vaccines reduce the overall infection rate, which means there’s less of a need for antibiotics,” Huang said. “This reduces the chance for bacteria to develop resistance, breaking the cycle. 

“The two go hand in hand.” 

UZH acquires important Richard Wagner manuscript




UNIVERSITY OF ZURICH
Manuscript "A message to my friends" by Richard Wagner 

IMAGE: 

IMAGE OF TWO PAGES OF THE WORKING MANUSCRIPT "A MESSAGE TO MY FRIENDS" BY RICHARD WAGNER.

view more 

CREDIT: (IMAGE: ZENTRALBIBLIOTHEK ZÜRICH)




The University of Zurich has made a spectacular acquisition of an important manuscript by Richard Wagner (1812 to 1883). In Eine Mitteilung an meine Freunde (“A Message to My Friends”), the composer takes autobiographical and artistic stock and looks to the future. The manuscript, which was written in Zurich, will now undergo further scholarly analysis at UZH.

Zurich was an important stage in Richard Wagner’s life and work. After taking part in the failed May Uprising in Dresden (1849) against the Saxon King Friedrich August II, Wagner lived in exile in Zurich from 1849 to 1858. The time in Zurich marked a turning point and reorientation in his life and work. Here he not only worked on his famous tetralogy The Ring of the Nibelung, but also wrote groundbreaking works of music and drama theory.

Returning to Zurich after 170 years

Some 170 years later, the original working manuscript of one of these writings, entitled Eine Mitteilung an meine Freunde (“A Message to My Friends”), has now returned to the place where it was created. UZH acquired the manuscript at an auction at Sotheby’s. The acquisition was possible thanks to donations from UBS Culture Foundation (Zurich), the Bareva Foundation (Vaduz) and the UZH Foundation, the foundation of the University of Zurich.

“The purchase of the Wagner manuscript is of great importance for Zurich, UZH and scholarly work,” says UZH President Michael Schaepman. Musicologist Laurenz Lütteken, co-director of the University of Zurich’s Department of Musicology, is also excited about the successful acquisition: “Wagner manuscripts of such high caliber are otherwise scarcely available on the free market.” The work of Richard Wagner is one of the focal points of the Department of Musicology’s research.

Taking autobiographical and artistic stock

Eine Mitteilung an meine Freunde was written in August 1851 in the “Zum Abendstern” building in Zurich’s Enge district (roughly the location of today’s Lavaterstrasse 76), shortly before Wagner’s move to the Vordere Escherhäuser premises (Zeltweg 11). The text was published in the same year as the supplement and foreword to the libretti of the operas The Flying DutchmanTannhäuser and Lohengrin.

 

The message is a kind of autobiographical stocktaking, with reference to the works before the revolution and the great post-revolutionary Ring project. “Wagner provides an interpretation of his work to date and a succinct summary of what ‘musikdrama’ should achieve in the future,” explains Laurenz Lütteken. At the very end, there is also a sketch of the tetralogy in progress, as well as a plan for a performance “in the course of a specifically designated three-day festival plus a preceding evening.” It is therefore also the first sketch of the festival idea that Wagner later brought to fruition with the Ring of the Nibelung.

UZH investigates writing of text

The extensive autograph of this important Wagner work, which will now be kept at Zentralbibliothek Zürich, is not simply returning to its place of origin. It will continue to occupy researchers at UZH. Until now, the text has only been available in the first printing and the version published in Richard Wagner’s collected writings and poems. The handwritten version, on the other hand, reveals intensive effort. The manuscript is written in dark brown ink and contains extensive deletions, corrections, revisions and additions. The musicologists at UZH have now embarked on an extensive research project to analyze and trace the genesis of the text. This analysis will provide new findings and insights into the work, thinking and influence of Richard Wagner in Zurich.

Once the research at UZH has been completed, the manuscript will be made available by the Zentralbibliothek Zurich (ZB) library. The working manuscript of Eine Mitteilung an meine Freunde is another gem in the important collection of Wagneriana at the ZB, which includes music and text manuscripts, music prints, printed matter and letters.

 

Narcissistic CEOs appoint other narcissists to the management board



Professor from TU Dortmund University has analyzed more than 11,000 LinkedIn profiles



TU DORTMUND UNIVERSITY

Professor Lorenz Graf-Vlachy, TU Dortmund University 

IMAGE: 

PROFESSOR LORENZ GRAF-VLACHY, TU DORTMUND UNIVERSITY

view more 

CREDIT: ALIONA KARDASH/TU DORTMUND




Narcissism manifests itself in different forms in leadership situations – from self-confidence and charisma to destructive self-focus and lack of consideration for others. While this character trait in individual CEOs is already the subject of research, very little is known so far about its impact on the wider composition of senior management.

This was the starting point for the study conducted by Professor Graf-Vlachy’s team, which analyzed LinkedIn profiles of top managers. “Narcissists want to show a broader audience that they are superior. Earlier studies have found that this is also reflected in corporate press releases or letters to shareholders, among other things,” says Professor Graf-Vlachy. Elements that are indicative of narcissism include, for example, the size of an executive’s picture in the annual report or the frequency with which a manager’s name appears in a company’s press releases.

In their study, the authors used the same approach to analyze individuals’ digital presence on social media. “We have demonstrated that we can reliably measure the narcissism of managers on the basis of their LinkedIn profiles by analyzing the number of pictures of the manager, the length of the text in the ‘About’ section, and the skills, certificates, and career steps listed,” says Professor Graf-Vlachy.

Increased fluctuation can be expensive for companies

In total, the team analyzed 11,705 LinkedIn profiles of top managers of US companies. The study reveals an astonishing tendency: CEOs with a higher degree of narcissism appoint members to their management board who mirror their own character traits – in other words, who also display a narcissistic streak. An increase in the narcissism of a CEO by one standard deviation leads to an 18 percent higher level of narcissism in each newly appointed executive.

This also has an impact on the dynamics and stability of the team, as management boards with more narcissistic managers have a significantly higher turnover rate, which can mean sizeable costs for a company. “Narcissists want to dominate each other, which leads to conflicts on the board, and these in turn lead to more fluctuation in the executive team,” summarizes Professor Graf-Vlachy.

“The results of the study show that it is important for CEOs and supervisory boards to understand the dynamics in their executive teams better and to review the selection process for managers,” the researchers conclude. “This can be achieved if the personality traits of managers are also viewed in a balanced way.”

Narcissism manifests itself in different forms in leadership situations – from self-confidence and charisma to destructive self-focus and lack of consideration for others. While this character trait in individual CEOs is already the subject of research, very little is known so far about its impact on the wider composition of senior management.

This was the starting point for the study conducted by Professor Graf-Vlachy’s team, which analyzed LinkedIn profiles of top managers. “Narcissists want to show a broader audience that they are superior. Earlier studies have found that this is also reflected in corporate press releases or letters to shareholders, among other things,” says Professor Graf-Vlachy. Elements that are indicative of narcissism include, for example, the size of an executive’s picture in the annual report or the frequency with which a manager’s name appears in a company’s press releases.

In their study, the authors used the same approach to analyze individuals’ digital presence on social media. “We have demonstrated that we can reliably measure the narcissism of managers on the basis of their LinkedIn profiles by analyzing the number of pictures of the manager, the length of the text in the ‘About’ section, and the skills, certificates, and career steps listed,” says Professor Graf-Vlachy.

In total, the team analyzed 11,705 LinkedIn profiles of top managers of US companies. The study reveals an astonishing tendency: CEOs with a higher degree of narcissism appoint members to their management board who mirror their own character traits – in other words, who also display a narcissistic streak. An increase in the narcissism of a CEO by one standard deviation leads to an 18 percent higher level of narcissism in each newly appointed executive.

This also has an impact on the dynamics and stability of the team, as management boards with more narcissistic managers have a significantly higher turnover rate, which can mean sizeable costs for a company. “Narcissists want to dominate each other, which leads to conflicts on the board, and these in turn lead to more fluctuation in the executive team,” summarizes Professor Graf-Vlachy.

“The results of the study show that it is important for CEOs and supervisory boards to understand the dynamics in their executive teams better and to review the selection process for managers,” the researchers conclude. “This can be achieved if the personality traits of managers are also viewed in a balanced way.”

 


 

Holographic displays offer a glimpse into an immersive future




PRINCETON UNIVERSITY, ENGINEERING SCHOOL
Illustration of a holographic display with the new optical device 

IMAGE: 

RESEARCHERS AT PRINCETON AND META HAVE CREATED A TINY OPTICAL DEVICE THAT MAKES HOLOGRAPHIC IMAGES LARGER AND CLEARER. SMALL ENOUGH TO FIT ON A PAIR OF EYEGLASSES, THE DEVICE COULD ENABLE A NEW KIND OF IMMERSIVE VIRTUAL REALITY DISPLAY.

view more 

CREDIT: ILLUSTRATION BY LIZ SABOL, PHOTO BY NATHAN MATSUDA




Setting the stage for a new era of immersive displays, researchers are one step closer to mixing the real and virtual worlds in an ordinary pair of eyeglasses using high-definition 3D holographic images, according to a study led by Princeton University researchers.

Holographic images have real depth because they are three dimensional, whereas monitors merely simulate depth on a 2D screen. Because we see in three dimensions, holographic images could be integrated seamlessly into our normal view of the everyday world.

The result is a virtual and augmented reality display that has the potential to be truly immersive, the kind where you can move your head normally and never lose the holographic images from view. “To get a similar experience using a monitor, you would need to sit right in front of a cinema screen,” said Felix Heide, assistant professor of computer science and senior author on a paper published April 22 in Nature Communications.

And you wouldn’t need to wear a screen in front of your eyes to get this immersive experience. Optical elements required to create these images are tiny and could potentially fit on a regular pair of glasses. Virtual reality displays that use a monitor, as current displays do, require a full headset. And they tend to be bulky because they need to accommodate a screen and the hardware necessary to operate it.

“Holography could make virtual and augmented reality displays easily usable, wearable and ultrathin,” said Heide. They could transform how we interact with our environments, everything from getting directions while driving, to monitoring a patient during surgery, to accessing plumbing instructions while doing a home repair.

One of the most important challenges is quality. Holographic images are created by a small chip-like device called a spatial light modulator. Until now, these modulators could only create images that are either small and clear or large and fuzzy. This tradeoff between image size and clarity results in a narrow field of view, too narrow to give the user an immersive experience. “If you look towards the corners of the display, the whole image may disappear,” said Nathan Matsuda, research scientist at Meta and co-author on the paper.

Heide, Matsuda and Ethan Tseng, doctoral student in computer science, have created a device to improve image quality and potentially solve this problem. Along with their collaborators, they built a second optical element to work in tandem with the spatial light modulator. Their device filters the light from the spatial light modulator to expand the field of view while preserving the stability and fidelity of the image. It creates a larger image with only a minimal drop in quality.

Image quality has been a core challenge preventing the practical applications of holographic displays, said Matsuda. “The research brings us one step closer to resolving this challenge,” he said.

The new optical element is like a very small custom-built piece of frosted glass, said Heide. The pattern etched into the frosted glass is the key. Designed using AI and optical techniques, the etched surface scatters light created by the spatial light modulator in a very precise way, pushing some elements of an image into frequency bands that are not easily perceived by the human eye. This improves the quality of the holographic image and expands the field of view.

Still, hurdles to making a working holographic display remain. The image quality isn’t yet perfect, said Heide, and the fabrication process for the optical elements needs to be improved. “A lot of technology has to come together to make this feasible,” said Heide. “But this research shows a path forward.”

The paper, “Neural Etendue Expander for Ultra-Wide-Angle High-Fidelity Holographic Display” was published April 22 in Nature Communications. In addition to Heide and Tseng, co-authors from Princeton include Seung-Hwan Baek and Praneeth Chakravarthula. In addition to Matsuda, co-authors from Meta Research are Grace Kuo, Andrew Maimone, Florian Schiffers, and Douglas Lanman. Qiang Fu and Wolfgang Heidrich from the Visual Computing Center at King Abdullah University of Science and Technology in Saudi Arabia also contributed. The work was supported by Princeton University’s Imaging and Analysis Center and the King Abdullah University of Science and Technology’s Nanofabrication Core Lab.

 

Crises like pandemics or financial crashes could stall progress on gender diversity in boardrooms


UNIVERSITY OF SURREY



The study revealed a concerning trend: major external disruptions like the 2008 Global Financial Crisis (GFC) triggered a significant drop in the number of women holding boardroom seats. This decline wasn't specific to certain board positions or industries, and it impacted companies across the financial performance spectrum. Furthermore, the research suggests that even strong leadership from female CEOs or existing institutional efforts promoting gender diversity, such as quotas, weren't effective in mitigating this decrease during crisis periods. 

Sorin Krammer, Professor of Strategy and International Business and corresponding author of the study from the University of Surrey, said: 

“Our findings highlight the vulnerability of the gains made towards gender diversity on boards when businesses face a major crisis. During these times, companies prioritise short-term survival and often demote diversity initiatives. 

While female leadership is crucial for advancing gender equality, our results suggest that female CEOs are not able to fully protect diversity efforts during periods of extreme turbulence. Similarly, existing initiatives, such as gender quotas or corporate governance codes, are not robust enough to withstand the negative effects of significant disruptions on gender diversity in top management teams.” 

The research encourages companies to develop long-term diversity strategies that are more resilient in the face of unforeseen, major challenges. Researchers suggest that policymakers can focus on strengthening institutional frameworks to promote and enforce gender diversity on boards. More research is needed to examine if there are any specific conditions or factors that would be able to mitigate the negative effects of crises when it comes to gender diversity. 

Professor Krammer continued: 

“These findings highlight the need for companies to develop long-term diversity strategies that can withstand unforeseen challenges. Moreover, policymakers should ensure that current institutional frameworks are adapted to cope with such exogenous shocks and be able to still deliver progress towards gender diversity on boards even in the wake of major disruptions.” 

 

This study has been published in The Leadership Quarterly

 

### 

Note to editors 

 

New study reveals how parasites shape complex food webs




QUEEN MARY UNIVERSITY OF LONDON





A new study published today in Proceedings of the Royal Society B sheds light on how parasites influence the intricate relationships between predator and prey populations. Researchers from the Faculty of Science and Engineering at Queen Mary University of London have developed a novel analysis to explore how factors like parasite virulence and infection probability affect the coexistence of species in a complex predator-prey-parasite system. 

"Food webs are intricate networks where species interact with each other," says Dr Weini Huang Reader in Mathematical Biology. "Parasites, a massive but often overlooked component of these webs, can significantly impact their stability by affecting both predator and prey populations." 

The research team is a new interdisciplinary collaboration between the group of Professor Christophe Eizaguirre, Professor of Evolutionary and Conservation Genetics in the school of Biological and Behavioural Sciences and Dr Weini Huang, Reader in Mathematical Biology in the School of Mathematical Sciences. Together with two earlier career researchers, Ana C. Hijar Islas and Amy Milne, they thoughtfully investigated this complex system through mathematical analysis and stochastic simulations, which considers microscopic events like reproduction, death, competition, infection, and predation at the individual level.  

The study revealed that stochasticity, random fluctuations in population sizes, plays a significant role in determining whether species coexist or go extinct, particularly at the boundary between these states. 

Dr Weini Huang and Amy Milne played a crucial role in developing the model's mathematical framework. Dr Huang explains: "We found that the relative abundance of infected and uninfected individuals can be reversed between the prey and predator populations. This counterintuitive finding suggests that the interplay of direct and indirect parasite effects plays a significant role in shaping infection prevalence throughout the food web." 

Professor Eizaguirre emphasises the importance of this research: "Understanding how parasites influence food web dynamics is crucial for predicting the impact of environmental changes on ecosystem health but also risks of spread of invasive diseases. Our findings provide a valuable framework for exploring risks of certain parasites to become invasive as they are move with their hosts" 

“Our findings provide a valuable foundation for understanding how these systems evolve over time,” concludes Dr Weini Huang. The team have developed a further study built upon this framework by incorporating the evolution of key parameters, such as reproduction costs and infection probability, under the combined influence of ecological and evolutionary pressures.” 

This research has the potential to inform conservation efforts by providing a deeper understanding of how parasites can influence the resilience of ecosystems. By incorporating parasite dynamics into ecological models, conservation biologists can develop more effective strategies to protect vulnerable species and maintain healthy ecosystems. 

BAD A.I.

On the trail of deepfakes, Drexel researchers identify ‘fingerprints’ of AI-generated video



Machine-learning approach could be key to mitigating AI-driven misinformation




DREXEL UNIVERSITY

In February, OpenAI released videos created by its generative artificial intelligence program Sora. The strikingly realistic content, produced via simple text prompts, is the latest breakthrough for companies demonstrating the capabilities of AI technology. It also raised concerns about generative AI’s potential to enable the creation of misleading and deceiving content on a massive scale. According to new research from Drexel University, current methods for detecting manipulated digital media will not be effective against AI-generated video; but a machine-learning approach could be the key to unmasking these synthetic creations.

In a paper accepted for presentation at the IEEE Computer Vision and Pattern Recognition Conference in June, researchers from Multimedia and Information Security Lab in Drexel’s College of Engineering explained that while existing synthetic image detection technology has failed thus far at spotting AI-generated video, they’ve had success with a machine learning algorithm that can be trained to extract and recognize digital “fingerprints” of many different video generators, such as Stable Video Diffusion, Video-Crafter and Cog-Video. Additionally, they have shown that this algorithm can learn to detect new AI generators after studying just a few examples of their videos.

“It’s more than a bit unnerving that this video technology could be released before there is a good system for detecting fakes created by bad actors,” said Matthew Stamm, PhD, an associate professor in Drexel’s College of Engineering and director of the MISL. “Responsible companies will do their best to embed identifiers and watermarks, but once the technology is publicly available, people who want to use it for deception will find a way. That’s why we’re working to stay ahead of them by developing the technology to identify synthetic videos from patterns and traits that are endemic to the media.”

Deepfake Detectives

Stamm’s lab has been active in efforts to flag digitally manipulated images and videos for more than a decade, but the group has been particularly busy in the last year, as editing technology is being used to spread political misinformation.

Until recently, these manipulations have been the product of photo and video editing programs that add, remove or shift pixels; or slow, speed up or clip out video frames. Each of these edits leaves a unique digital breadcrumb trail and Stamm’s lab has developed a suite of tools calibrated to find and follow them.

The lab’s tools use a sophisticated machine learning program called a constrained neural network. This algorithm can learn, in ways similar to the human brain, what is “normal” and what is “unusual” at the sub-pixel level of images and videos, rather than searching for specific predetermined identifiers of manipulation from the outset. This makes the program adept at both identifying deepfakes from known sources, as well as spotting those created by a previously unknown program.

The neural network is typically trained on hundreds or thousands of examples to get a very good feel for the difference between unedited media and something that has been manipulated — this can be anything from variation between adjacent pixels, to the order of spacing of frames in a video, to the size and compression of the files themselves.

A New Challenge

“When you make an image, the physical and algorithmic processing in your camera introduces relationships between various pixel values that are very different than the pixel values if you photoshop or AI-generate an image,” Stamm said. “But recently we’ve seen text-to video generators, like Sora, that can make some pretty impressive videos. And those pose a completely new challenge because they have not been produced by a camera or photoshopped.”

Last year a campaign ad circulating in support of Florida Gov. Ron DeSantis appeared to show former President Donald Trump embracing and kissing Antony Fauci was the first to use generative-AI technology. This means the video was not edited or spliced together from others, rather it was created whole-cloth by an AI program.

And if there is no editing, Stamm notes, then the standard clues do not exist — which poses a unique problem for detection.

“Until now, forensic detection programs have been effective against edited videos by simply treating them as a series of images and applying the same detection process,” Stamm said. “But with AI-generated video, there is no evidence of image manipulation frame-to-frame, so for a detection program to be effective it will need to be able to identify new traces left behind by the way generative-AI programs construct their videos.”

In the study, the team tested 11 publicly available synthetic image detectors. Each of these programs was highly effective — at least 90% accuracy — at identifying manipulated images. But their performance dropped by 20-30% when faced with discerning videos created by publicly available AI-generators, Luma, VideoCrafter-v1, CogVideo and Stable Diffusion Video.

“These results clearly show that synthetic image detectors experience substantial difficulty detecting synthetic videos,” they wrote. “This finding holds consistent across multiple different detector architectures, as well as when detectors are pretrained by others or retrained using our dataset.”

A Trusted Approach

The team speculated that convolutional neural network-based detectors, like its MISLnet algorithm, could be successful against synthetic video because the program is designed to constantly shift its learning as it encounters new examples. By doing this, it’s possible to recognize new forensic traces as they evolve. Over the last several years, the team has demonstrated MISLnet’s acuity at spotting images that had been manipulated using new editing programs, including AI tools — so testing it against synthetic video was a natural step.

“We’ve used CNN algorithms to detect manipulated images and video and audio deepfakes with reliable success,” said Tai D. Nguyen, a doctoral student in MISL, who was a coauthor of the paper. “Due to their ability to adapt with small amounts of new information we thought they could be an effective solution for identifying AI-generated synthetic videos as well.”

For the test, the group trained eight CNN detectors, including MISLnet, with the same test dataset used to train the image detectors, which including real videos and AI-generated videos produced by the four publicly available programs. Then they tested the program against a set of videos that included a number created by generative AI programs that are not yet publicly available: Sora, Pika and VideoCrafter-v2.

By analyzing a small portion — a patch — from a single frame from each video, the CNN detectors were able to learn what a synthetic video looks like at a granular level and apply that knowledge to the new set of videos. Each program was more than 93% effective at identify the synthetic videos, with MISLnet performing the best, at 98.3%.

The programs were slightly more effective when conducting an analysis of the entire video, by pulling out a random sampling of a few dozen patches from various frames of the video and using those as a mini training set to learn the characteristics of the new video. Using a set of 80 patches, the programs were between 95-98% accurate.

With a bit of additional training, the programs were also more than 90% accurate at identifying the program that was used to create the videos, which the team suggests is because of the unique, proprietary approach each program uses to produce a video.

“Videos are generated using a wide variety of strategies and generator architectures,” the researchers wrote. “Since each technique imparts significant traces, this makes it much easier for networks to accurately discriminate between each generator.”

A Quick Study

While the programs struggled when faced with the challenge of detecting a completely new generator without previously being exposed to at least a small amount of video from it, with a small amount of fine tuning MISLnet could quickly learn to make the identification at 98% accuracy. This strategy, called “few-shot learning” is an important capability because new AI technology is being created every day, so detection programs must be agile enough to adapt with minimal training.

“We’ve already seen AI-generated video being used to create misinformation,” Stamm said. “As these programs become more ubiquitous and easier to use, we can reasonably expect to be inundated with synthetic videos. While detection programs shouldn’t be the only line of defense against misinformation — information literacy efforts are key — having the technological ability to verify the authenticity of digital media is certainly an important step.”