Ritwik Kulkarni et al. 2023
Ritwik Kulkarni et al. 2023

University of Helsinki AI methods tackle the illegal wildlife trade on the Internet

In Finland, scientists applied machine vision models and were able to deduce from the context of an image if it pertained to the sale of a live animal. These methods make it possible to flag the posts which may be selling animals illegally.

Illegal wildlife trade is estimated to be a multi-billion dollar industry where hundreds of species are traded globally. A considerable proportion of the illegal wildlife trade now uses online marketplaces to advertise and sell live animals or animal products as it can reach more buyers than previously possible. With the trade happening across the Internet, it is extremely challenging to manually search through thousands of posts, and methods for automated filtering are needed.

Compared to using computer vision to identify species from images, the identification of images related to the illegal wildlife trade of species is rendered difficult by the need to identify the context in which the species are portrayed.

In a new article published in Biological Conservation, scientists based at the Helsinki Lab of Interdisciplinary Conservation Science, University of Helsinki, Finland, have filled this gap and developed an automated algorithm using machine learning to identify such image content in the digital space.

“This is the first-time machine vision models have been applied to deduce the context of an image to identify the sale of a live animal. When a seller is advertising an animal for sale, many times the advertisement is accompanied by an image of the animal in a captive state. This differs from non-captive images, for example, a picture of an animal taken by a tourist in a national park. Using a technique called feature visualization, we demonstrated that our models could take into account both the presence of an animal in the image and the surrounding environment of the animal in the image. Thus, making it possible to flag the posts which may be selling animals illegally.” says Dr. Ritwik Kulkarni, the lead author of this study.

As part of their research, scientists trained 24 different neural-net models on a newly created dataset, under various experimental conditions. The top-performing models achieved very high accuracy and were able to discern well between natural and captive contexts. Another interesting feature of the study is that the models were also tested and performed well on data acquired from a source unrelated to training data, therefore showing the capability to work well for the identification of other content on the Internet.                    

“These methods are a game changer in our work that seeks to enhance automated identification of illegal wildlife trade content from digital sources. We are now upscaling this work to include more taxonomic groups beyond mammals and to develop new models that can identify image and text content simultaneously.”, says Associate Professor Enrico Di Minin, the other co-author who heads the Helsinki Lab of Interdisciplinary Conservation Science.

The scientists are planning to make their methods openly available for the use of the broader scientific and practitioners’ community.

Surrey's Xavier performs simulations to produce discoveries for sustainable hydrogen production

Hydrogen fuel could be a more viable alternative to traditional fossil fuels according to the University of Surrey researchers in Guildford, Surrey, England who have found that a type of metal-free catalysts could contribute to the development of cost-effective and sustainable hydrogen production technologies.

The study has shown promising results for the use of edge-decorated nano carbons as metal-free catalysts for the direct conversion of methane, which is also a powerful greenhouse gas, into hydrogen. Among the nano carbons investigated, nitrogen-doped nano carbons presented the highest level of performance for hydrogen production at high temperatures.

Crucially, the researchers also found that the nitrogen-doped and phosphorous-doped nano carbons had strong resistance to carbon poisoning, which is a common issue with catalysts in this process.

Dr. Neubi Xavier Jr, the Research Fellow who performed the material science simulations, said: "Our results suggest that using edge-decorated nano carbons as catalysts could be a game-changer for the hydrogen industry, offering a cost-effective and sustainable alternative to traditional metal catalysts. At the same time, this process gets rid of methane, which is a fossil fuel involved in global warming."

Hydrogen fuel is a clean and renewable energy source that has the potential to reduce carbon emissions and decrease our dependence on fossil fuels. When used as a fuel, hydrogen can power vehicles, generate electricity, and heat buildings. The only by-product of hydrogen fuel is water vapor, making it an environmentally friendly alternative to traditional fossil fuels.

However, the production of hydrogen fuel is currently reliant on fossil fuels, which creates carbon emissions in the process, and metal catalysts, which mining and manufacturing are energy intensive and can negatively affect the environment. Therefore, the development of sustainable hydrogen production methods and catalytic materials is crucial to realizing the full potential of hydrogen fuel as a clean energy source.

The research was conducted by a team led by Dr. Marco Sacchi from the University of Surrey, an expert in the field of sustainable energy and computational chemistry, who combined quantum chemistry, thermodynamics, and chemical kinetics to determine the most efficient edge decoration for hydrogen production.

Dr. Sacchi said: "One of the biggest challenges with catalysts for hydrogen production is that they can get poisoned by carbon. But our study found that nitrogen and phosphorous-doped nano carbons are pretty resistant to this problem. This is a huge step forward for sustainable hydrogen production."

Credit: Carlos Padilla, NRAO/AUI/NSF
Credit: Carlos Padilla, NRAO/AUI/NSF

ALMA gets supercomputing system upgrades that increase data production capacity by a factor of 200-400

The Board of the Atacama Large Millimeter/submillimeter Array (ALMA)— an international collaboration in which the National Science Foundation’s National Radio Astronomy Observatory (NRAO) is a partner— has approved multi-million dollar upgrades for the development of a second-generation correlator and a digital transmission system (DTS). As part of the ALMA2030 Wideband Sensitivity Upgrade, these projects aim to double and eventually quadruple the correlated bandwidth of the array.

Central to the ALMA2030 upgrades, the Second Generation ALMA Correlator— the “brain” of the array— is a type of supercomputer that combines the individual signals from each antenna to create exquisite images of astronomical objects. The new correlator will improve the current one’s already the highly refined ability to process and combine data and increase the sensitivity of astronomical images and the flexibility of making them.

“While ALMA’s current correlators are already some of the fastest supercomputing signal processors in the world, the Second Generation Correlator will be capable of producing 200, and ultimately 400 times more data per second along with an increased sensitivity equivalent to adding more than 1000 hours of observing time per year,” said Crystal Brogan, ALMA-North America Program Scientist and the ALMA Development Program Coordinator at NRAO. “The initial expansion in system bandwidth by a factor of two, and eventually four, will enhance the science throughput for all areas of ALMA science from the most distant galaxies to our Solar System. The Second Generation ALMA Correlator will also enable high spectral resolution at wide bandwidth for the first time – affording an unprecedented view of the kinematics and chemistry of star and planet formation.”

The $36 million project will take approximately six years to complete and combines the hardware and firmware expertise of scientists and engineers at the National Research Council of Canada (NRC) and the software expertise of NRAO’s Data Management and Software Department. Additionally, experts at the Massachusetts Institute of Technology’s Haystack Observatory will be assisting with the implementation and testing of the Phased Array aspects of the new correlator. The project is led by the NRAO’s North American ALMA Department.

“The new correlator provides the foundation for the rest of the Wideband Sensitivity Upgrade (WSU).  With the project’s approval, the WSU has moved from plans to construction. The international ALMA collaboration will work together to deliver this project, and by the end of this decade, we’ll see the results in amazing new science,” said Phil Jewell, Director for ALMA-North America.

The upgraded Digital Transmission System (DTS)— a collaboration between NRAO’s Central Development Laboratory (CDL) and the National Astronomical Observatory of Japan (NAOJ), also a partner in ALMA— will act as an expanded information highway, increasing the amount of data that can travel from each of ALMA’s upgraded receivers to the upgraded correlator by a factor of eight.

“The DTS is an exciting collaboration with our colleagues at NAOJ and will provide a higher-capacity digital path for data from the upgraded receivers to the ALMA Talon Central Signal Processor. The project leverages our expertise in photonics and digital signal processing and will be built using state-of-the-art hardware, enabling a wide range of improvements,” said Bert Hawkins, Director of CDL.

Alvaro Gonzalez, East Asia ALMA Program Manager at NAOJ added, “The new ALMA2030 DTS will be based on the latest high-speed data-transmission standards and use commercially available technology as much as possible. As a collaboration between NAOJ and NRAO, we will combine the best aspects of technology and know-how from the two partners. The DTS will be designed to support the goal of a 4 times increase of instantaneous bandwidth of ALMA receivers and also the eventual increase of the distance between antennas, up to around 75 km, for improved angular resolution.”

Phase 1 of the DTS upgrade— approved for ~US$800,000— aims to produce a prototype of the new end-to-end system by 2026 and will be followed by a Phase 2 production proposal.

NRAO’s and NA ALMA’s central role in the ALMA2030 upgrades extends beyond the correlator and DTS and includes the conversion of the Operations Support Facility to house and operate the new correlator, additional infrastructure and support systems, and receiver upgrades. CDL has already commenced work to upgrade ALMA’s 1.3mm (Band 6) receivers after receiving approval and Phase 1 funding in late 2021. The Band 6v2 receiver prototype is expected in 2025, allowing for the build-out of an entirely upgraded set of Band 6 receivers for ALMA that will increase the quantity and quality of science measured in wavelengths between 1.4mm and 1.1mm.

Upon completion, ALMA2030 will realize upgrades to most ALMA receivers resulting in increased bandwidth and sensitivity, complete replacement of the ALMA digital signal chain— digitizer, digital transmission system, and correlator— and installation of new fiber cables connecting ALMA’s Operations Site to its Operations Support Facility, and develop associated control, data acquisition, and data processing software.

“An already immensely powerful observatory, ALMA has uncovered the secrets of protoplanetary disks and the unseen gas and dust that drives the formation of stars, planets, and galaxies. These upgrades will help us see further than ever before and process this information faster and more clearly,” said NRAO Director Tony Beasley. “With each upgrade, we are quite literally building the future of radio astronomy.”

ALMA Director Sean Dougherty added, “This is a very exciting moment for ALMA. The approval of these two major components of the Wide-Band Sensitivity Upgrade— a new data transmission system and future-forward correlator— will extend the science capabilities of ALMA enormously across all fields of science.”

“This exciting project ensures ALMA continues to operate and provide fantastic observations,” says Joe Pesce, NSF Program Officer for ALMA.  “Improved capabilities enabled by the upgraded correlator will lead to discoveries about our universe and advancement of science.”

“This project will significantly improve the sensitivity, flexibility, and efficiency of the telescope,” said Brent Carlson, Research Officer at the NRC’s Herzberg Astronomy and Astrophysics Centre and the NRC’s Principal Investigator for the correlator project. “The Second Generation ALMA Correlator will allow much more spectral information from radio sources to be imaged instantaneously, giving scientists access to a colossal amount of new data. The ability to do spectral scans efficiently at such high resolution is unprecedented and will keep ALMA at the forefront of scientific discovery.”

The North American ALMA Development Program is funded by the NSF and the National Research Council of Canada.

The National Radio Astronomy Observatory (NRAO) is a facility of the National Science Foundation, operated under a cooperative agreement by Associated Universities, Inc.

Prof. Dr. Bernhard (Image: FAU/Georg Pöhlein)
Prof. Dr. Bernhard (Image: FAU/Georg Pöhlein)

FAU prof Kainz wins 2 million euros for ML in medical imaging diagnostics

It is a feather in the cap for outstanding research at the Department of Artificial Intelligence in Biomedical Engineering at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) in Bavaria, Germany: The European Research Council (ERC) has awarded a Consolidator Grant to Prof. Dr. Bernhard Kainz. The professor of Image Data Exploration and Analysis has received the most prestigious research funding award available throughout the whole of Europe for a project focusing on automated medical image analysis. The two million euros in funding over five years is to be used to train computer tools based on artificial intelligence to reliably recognize healthy human tissue based on image material. The aim is to use these tools for medical diagnosis to support and ease the workload of experts working in the field, for example during health screening programs.

Improving medical care

Imaging is playing an increasingly significant role in medicine, but analyzing images is both time-consuming and costly. It not only ties up existing resources and leads to an increase in costs, but it also causes long waiting times for patients. With their machine learning project, Professor Kainz and his team hope to train computer programs to recognize healthy tissue structures. Artificial intelligence would then be able to pre-sort the images obtained during the diagnosis process into “probably healthy” or “possibly sick”. The final decision is taken by medical experts. With the support of the machines, however, medical staff would gain valuable time that they could then use to investigate any images deviating from the norm more thoroughly. As a knock-on effect, more patients could be treated, and patients would not have to wait so long to find out whether the images indicated that there was a problem with their health.

According to Bernhard Kainz, he is driven by the “conviction that everyone deserves the same quality of medical care, no matter where they live or how much they earn. That is why our working group is working on developing methods that make high-quality medical imaging analyses widely available and scalable.”

Recognize healthy tissue

Why should AI recognize healthy tissue? Why is it not being trained to diagnose diseases? Prof. Kainz has a clear answer: “Training machine learning tools using hundreds and hundreds of examples of every possible disease would be extremely costly in terms of time and manpower. Medical experts who are already overstretched would have to provide and comment on vast amounts of images of pathological structures.” In his opinion, it makes much more sense to “feed” the AI with images of healthy tissue structures. That is time-consuming enough, as healthy tissue differs depending on age and other characteristics such as gender.

The team is therefore working on providing self-learning machine diagnostic tools that recognize what healthy anatomy should look like at a certain point in time. The computer tools are to be trained to recognize normal physiological traits and any unusual changes over a certain period in individual patients. In addition, they should match patient information provided by physicians (for example laboratory results) to the available images. Bernhard Kainz’s goal is that the tools will recognize any deviations that require a more detailed medical investigation, at the same time as avoiding any unnecessary examinations, for the benefit of the patients.

CHARON CHASMAS
CHARON CHASMAS

Rhoden's models explain canyons on Pluto moon

In 2015, when NASA’s New Horizons spacecraft encountered the Pluto-Charon system, the Southwest Research Institute-led science team discovered interesting, geologically active objects instead of the inert icy orbs previously envisioned. A SwRI scientist has revisited the data to explore the source of cryovolcanic flows and an obvious belt of fractures on Pluto’s large moon Charon. These new models suggest that when the moon’s internal ocean froze, it may have formed the deep, elongated depressions along its girth but was less likely to lead to cryovolcanoes erupting with ice, water, and other materials in its northern hemisphere.

“A combination of geological interpretations and thermal-orbital evolution models implies that Charon had a subsurface liquid ocean that eventually froze,” said SwRI’s Dr. Alyssa Rhoden, a specialist in the geophysics of icy satellites, particularly those containing oceans, and the evolution of giant planet satellite systems. She authored a new paper on the source of Charon’s surface features in Icarus. “When an internal ocean freezes, it expands, creating large stresses in its icy shell and pressurizing the water below. We suspected this was the source of Charon’s large canyons and cryovolcanic flows.”

New ice forming on the inner layer of the existing ice shell can also stress the surface structure. To better understand the evolution of the moon’s interior and surface, Rhoden modeled how fractures formed in Charon’s ice shell as the ocean beneath it froze. The team modeled oceans of water, ammonia, or a mixture of the two based on questions about the makeup. Ammonia can act as antifreeze and prolong the life of the ocean; however, results did not differ substantially.

When fractures penetrate the entire ice shell and tap the subsurface ocean, the liquid, pressurized by the increase in the volume of the newly frozen ice, can be pushed through the fractures to erupt onto the surface. Models sought to identify the conditions that could create fractures that fully penetrate Charon’s icy shell, linking its surface and subsurface water to allow ocean-sourced cryovolcanism. However, based on current models of Charon’s interior evolution, ice shells were far too thick to be fully cracked by the stresses associated with ocean freezing.

The timing of the ocean freeze is also important. The synchronous and circular orbits of Pluto and Charon stabilized relatively early, so tidal heating only occurred during the first million years.

“Either Charon’s ice shell was less than 6 miles (10 km) thick when the flows occurred, as opposed to the more than 60 miles or 100 km indicated, or the surface was not in direct communication with the ocean as part of the eruptive process,” Rhoden said. “If Charon’s ice shell had been thin enough to be fully cracked, it would imply substantially more ocean freezing than is indicated by the canyons identified on Charon’s encounter hemisphere.”

Fractures in the ice shell may be the initiation points of these canyons along the global tectonic belt of ridges that traverse the face of Charon, separating the northern and southern geological regions of the moon. If additional large extensional features were identified on the hemisphere not imaged by New Horizons, or compositional analysis could prove that Charon’s cryovolcanism originated from the ocean, it would support the idea that its ocean was substantially thicker than expected.

“Ocean freezing also predicts a sequence of geologic activity, in which ocean-sourced cryovolcanism ceases before strain-created tectonism,” Rhoden said. “A more detailed analysis of Charon’s geologic record could help determine whether such a scenario is viable.”