Computers Math

A new collaborative effort between MBARI and other research institutions is leveraging the power of artificial intelligence and machine learning to accelerate efforts to study the ocean.
In order to manage impacts from climate change and other threats, researchers urgently need to learn more about the ocean’s inhabitants, ecosystems, and processes. As scientists and engineers develop advanced robotics that can visualize marine life and environments to monitor changes in the ocean’s health, they face a fundamental problem: The collection of images, video, and other visual data vastly exceeds researchers’ capacity for analysis.
FathomNet is an open-source image database that uses state-of-the-art data processing algorithms to help process the backlog of visual data. Using artificial intelligence and machine learning will alleviate the bottleneck for analyzing underwater imagery and accelerate important research around ocean health.
“A big ocean needs big data. Researchers are collecting large quantities of visual data to observe life in the ocean. How can we possibly process all this information without automation? Machine learning provides a pathway forwards, however these approaches rely on massive datasets for training. FathomNet has been built to fill this gap,” said MBARI Principal Engineer Kakani Katija.
Project co-founders Katija, Katy Croff Bell (Ocean Discovery League), and Ben Woodward (CVision AI), along with members of the extended FathomNet team, detailed the development of this new image database in a recent research publication in Scientific Reports.
Recent advances in machine learning enable fast, sophisticated analysis of visual data, but the use of artificial intelligence in ocean research has been limited by the lack of a standard set of existing images that could be used to train the machines to recognize and catalog underwater objects and life. FathomNet addresses this need by aggregating images from multiple sources to create a publicly available, expertly curated underwater image training database. More

The journey between identifying a potential therapeutic compound and Food and Drug Administration approval of a new drug can take well over a decade and cost upwards of a billion dollars. A research team at the CUNY Graduate Center has created an artificial intelligence model that could significantly improve the accuracy and reduce the time and cost of the drug development process. Described in a newly published paper in Nature Machine Intelligence, the new model, called CODE-AE, can screen novel drug compounds to accurately predict efficacy in humans. In tests, it was also able to theoretically identify personalized drugs for over 9,000 patients that could better treat their conditions. Researchers expect the technique to significantly accelerate drug discovery and precision medicine.
Accurate and robust prediction of patient-specific responses to a new chemical compound is critical to discover safe and effective therapeutics and select an existing drug for a specific patient. However, it is unethical and infeasible to do early efficacy testing of a drug in humans directly. Cell or tissue models are often used as a surrogate of the human body to evaluate the therapeutic effect of a drug molecule. Unfortunately, the drug effect in a disease model often does not correlate with the drug efficacy and toxicity in human patients. This knowledge gap is a major factor in the high costs and low productivity rates of drug discovery.
“Our new machine learning model can address the translational challenge from disease models to humans,” said Lei Xie, a professor of computer science, biology and biochemistry at the CUNY Graduate Center and Hunter College and the paper’s senior author. “CODE-AE uses biology-inspired design and takes advantage of several recent advances in machine learning. For example, one of its components uses similar techniques in Deepfake image generation.”
The new model can provide a workaround to the problem of having sufficient patient data to train a generalized machine learning model, said You Wu, a CUNY Graduate Center Ph.D. student and co-author of the paper. “Although many methods have been developed to utilize cell-line screens for predicting clinical responses, their performances are unreliable due to data incongruity and discrepancies,” Wu said. “CODE-AE can extract intrinsic biological signals masked by noise and confounding factors and effectively alleviated the data-discrepancy problem.”
As a result, CODE-AE significantly improves accuracy and robustness over state-of-the-art methods in predicting patient-specific drug responses purely from cell-line compound screens.
The research team’s next challenge in advancing the technology’s use in drug discovery is developing a way for CODE-AE to reliably predict the effect of a new drug’s concentration and metabolization in human bodies. The researchers also noted that the AI model could potentially be tweaked to accurately predict human side effects to drugs.
This work was supported by the National Institute of General Medical Sciences and the National Institute on Aging.
Story Source:
Materials provided by The Graduate Center, CUNY. Note: Content may be edited for style and length. More

Omnipose, a deep learning software, is helping to solve the challenge of identifying varied and miniscule bacteria in microscopy images. It has gone beyond this initial goal to identify several other types of tiny objects in micrographs.
The UW Medicine microbiology lab of Joseph Mougous and the University of Washington physics and bioengineering lab of Paul A. Wiggins tested the tool. It was developed by University of Washington physics graduate student Kevin J. Cutler and his team.
Mougous said that Cutler, as a physics student, “demonstrated an unusual interest in immersing himself in a biology environment so that he could learn first-hand about problems in need of solution in this field. He came over to my lab and quickly found one that he solved in spectacular fashion.”
Their results are reported in the Oct. 17 edition of Nature Methods.
The scientists found that Omnipose, trained on a large database of bacterial images, performed well in characterizing and quantifying the myriad of bacteria in mixed microbial cultures and eliminated some of the errors that can occur in its predecessor, Cellpose.
Moreover, the software wasn’t easily fooled by extreme changes in a cell’s shape due to antibiotic treatment or antagonism by chemicals produced during interbacterial aggression. In fact, the program showed that it could even detect cell intoxication in a trial using E. coli.
In addition, Omnipose did well in overcoming recognition problems due to differences in the optical characteristics across diverse bacteria. More

Researchers from North Carolina State University have developed a dynamic computational tool to help improve user access to electric vehicle (EV) charging stations, with the goal of making EVs more attractive for drivers.
“We already know that there is a need for EV charging networks that are flexible, in order to support the adoption of EVs,” says Leila Hajibabai, corresponding author of a paper on the work and an assistant professor in NC State’s Fitts Department of Industrial and Systems Engineering. “That’s because there is tremendous variability in when and where people want to charge their vehicles, how much time they can spend at a charging station, how long it takes to charge their vehicles, and so on.
“The fundamental question we wanted to address with this work is: What is the best way to manage existing charging station infrastructure in order to best meet the demands of electric vehicle users?”
To answer that question, the researchers wanted to take the user’s perspective, so focused on questions that are important to EV drivers. How long will it take me to reach a charging station? What is the cost of using the charging station? How long might I have to wait to access a charging station? And what sort of fines are there if I stay at a charging station beyond the time limit?
The researchers developed a technique that accounts for all of these factors in a complex computational model that makes use of a game theory framework.
The technique does two things. First, it helps users find the nearest charging facility that meets their needs. Second, it has a dynamic system that charging station operators can use to determine how long vehicles can spend at a charging station before they need to make way for the next vehicle.
“These outcomes are themselves dynamic — they evolve as additional data comes in about how users are making use of charging facilities,” Hajibabai says.
For example, a user’s nearest available charging facility may change, depending on whether any spaces are available. And the amount of time users can spend at a charging station may change from day to day to reflect the reality of how people are using different charging facilities.
“There’s no clear real-world benchmark that we can use to assess the extent to which our technique would improve user access to charging facilities,” Hajibabai says. “But in simulations, the technique did improve user access. The simulations also suggest that flexibility in when charging station slots are available was a key predictor of which stations users would visit.
“A next step would be to work with existing charging station networks to pilot the technique and assess its performance in a real-world setting.”
Story Source:
Materials provided by North Carolina State University. Original written by Matt Shipman. Note: Content may be edited for style and length. More

Too much screen time can slow children’s recovery from concussions, but new research from UBC and the University of Calgary suggests that banning screen time is not the answer.
The researchers looked for links between the self-reported screen time of more than 700 children aged 8-16 in the first 7-10 days following an injury, and symptoms reported by them and their caregivers over the following six months.
The children whose concussion symptoms cleared up the fastest had engaged in a moderate amount of screen time. “We’ve been calling this the ‘Goldilocks’ group, because it appears that spending too little or too much time on screens isn’t ideal for concussion recovery,” said Dr. Molly Cairncross, an assistant professor at Simon Fraser University who conducted the research while a postdoctoral fellow working with associate professor Dr. Noah Silverberg in UBC’s psychology department. “Our findings show that the common recommendation to avoid smartphones, computers and televisions as much as possible may not be what’s best for kids.”
The study was part of a larger concussion project called Advancing Concussion Assessment in Pediatrics (A-CAP) led by psychology professor Dr. Keith Yeates at the University of Calgary and funded by the Canadian Institutes of Health Research. The data came from participants aged 8-16 who had suffered either a concussion or an orthopaedic injury, such as a sprained ankle or broken arm, and sought care at one of five emergency departments in Canada.
The purpose of including children who had orthopaedic injuries was to compare their recoveries with the group who had concussions.
Patients in the concussion group generally had relatively worse symptoms than their counterparts with orthopaedic injuries, but within the concussion group it was not simply a matter of symptoms worsening with more screen time. Children with minimal screen time recovered more slowly, too. More

Researchers have developed an automatic drawing machine that uses pens and pencils to draw metamaterials onto paper. They demonstrated the new approach by using it to make three metamaterials that can be used to manipulate the microwave region of the electromagnetic spectrum.
Metamaterials are artificially engineered composite materials that derive their properties from patterned microstructures, rather than the chemical composition of the materials themselves. The exact shape, geometry, size, orientation and arrangement of the structures can be used to manipulate electromagnetic waves in ways that aren’t possible with conventional materials.
“Metamaterials, especially those used as absorbers, generally need to be thin, lightweight, wide and strong, but it isn’t easy to create thin and lightweight devices using traditional substrates,” said research team leader Junming Zhao from Nanjing University in China. “Using paper as the substrate can help meet these requirements while also lending itself to metasurfaces that conform to a surface or that are mechanically reconfigurable.”
In the journal Optical Materials Express, the researchers describe their new technique, which uses aballpoint pen with conductive ink to draw conductors and mechanical pencils to draw resistors and resistive films. They incorporated this process into a computer-controlled drawing machine to make it more automatic and accurate.
“Although paper-based metamaterials have been made previously using inkjet printing technology, our drawing technique is lower cost, simpler and more flexible,” said Zhao. “Our method could be useful for making reconfigurable antennas and metalenses as well as metamaterial devices that absorb incident electromagnetic energy from cell phones or other sources.”
Automated drawing
The new drawing machine uses pens with ink containing conductive material or normal mechanical pencils with varying graphite content. It has three stepper motors, two of which control the movement of the pen or pencil in the horizontal plane, while the other lifts or drops the writing instrument in the vertical plane. The parameters of the drawing machine, such as the movement speed, are controlled by a computer. More

Using existing experimental and computational resources, a multi-institutional team has developed an effective method for measuring high-dimensional qudits encoded in quantum frequency combs, which are a type of photon source, on a single optical chip.
Although the word “qudit” might look like a typo, this lesser-known cousin of the qubit, or quantum bit, can carry more information and is more resistant to noise — both of which are key qualities needed to improve the performance of quantum networks, quantum key distribution systems and, eventually, the quantum internet.
Classical computer bits categorize data as ones or zeroes, whereas qubits can hold values of one, zero orboth — simultaneously — owing to superposition, which is a phenomenon that allows multiple quantum states to exist at the same time. The “d” in qudit stands for the number of different levels or values that can be encoded on a photon. Traditional qubits have two levels, but adding more levels transforms them into qudits.
Recently, researchers from the U.S. Department of Energy’s Oak Ridge National Laboratory, Purdue University and the Swiss Federal Institute of Technology Lausanne, or EPFL, fully characterized an entangled pair of eight-level qudits, which formed a 64-dimensional quantum space — quadrupling the previous record for discrete frequency modes. These results were published in Nature Communications.
“We’ve always known that it’s possible to encode 10- or 20-level qudits or even higher using the colors of photons, or optical frequencies, but the problem is that measuring these particles is very difficult,” said Hsuan-Hao Lu, a postdoctoral research associate at ORNL. “That’s the value of this paper — we found an efficient and novel technique that is relatively easy to do on the experimental side.”
Qudits are even more difficult to measure when they are entangled, meaning they share nonclassical correlations regardless of the physical distance between them. Despite these challenges, frequency-bin pairs — two qudits in the form of photons that are entangled in their frequencies — are well suited to carrying quantum information because they can follow a prescribed path through optical fiber without being significantly modified by their environment. More

Inspired by living things from trees to shellfish, researchers at The University of Texas at Austin set out to create a plastic much like many life forms that are hard and rigid in some places and soft and stretchy in others. Their success — a first, using only light and a catalyst to change properties such as hardness and elasticity in molecules of the same type — has brought about a new material that is 10 times as tough as natural rubber and could lead to more flexible electronics and robotics.
The findings are published today in the journal Science.
“This is the first material of its type,” said Zachariah Page, assistant professor of chemistry and corresponding author on the paper. “The ability to control crystallization, and therefore the physical properties of the material, with the application of light is potentially transformative for wearable electronics or actuators in soft robotics.”
Scientists have long sought to mimic the properties of living structures, like skin and muscle, with synthetic materials. In living organisms, structures often combine attributes such as strength and flexibility with ease. When using a mix of different synthetic materials to mimic these attributes, materials often fail, coming apart and ripping at the junctures between different materials.
Oftentimes, when bringing materials together, particularly if they have very different mechanical properties, they want to come apart,” Page said. Page and his team were able to control and change the structure of a plastic-like material, using light to alter how firm or stretchy the material would be.
Chemists started with a monomer, a small molecule that binds with others like it to form the building blocks for larger structures called polymers that were similar to the polymer found in the most commonly used plastic. After testing a dozen catalysts, they found one that, when added to their monomer and shown visible light, resulted in a semicrystalline polymer similar to those found in existing synthetic rubber. A harder and more rigid material was formed in the areas the light touched, while the unlit areas retained their soft, stretchy properties. More