Tyler O'Neal, Staff Editor HEALTH

Warwick physicists apply algos inspired by social networks to reveal the lifecycle of substorms, a key element of space weather

Evolution of auroral substorms revealed by physicists at University of Warwick using the same methods that link people through social media

  • Evolution of auroral substorms revealed by physicists at University of Warwick using the same methods that link people through social media
  • 'If you like this magnetometer, you might like this one too:' historical data from magnetometers used to match them with 'like-minded friends' during 41 substorms
  • This shows that a single coherent electrical current, that accompanies the Northern Lights during substorms, covers most of the Earth's night-side at high latitudes
  • It will help to validate models used to predict auroral substorms, which can disrupt electronics and power distribution systems

Space weather often manifests as substorms, where a beautiful auroral display such as the Northern Lights is accompanied by an electrical current in space which has effects on the earth that can interfere with and damage power distribution and electrical systems. Now, the lifecycle of these auroral substorms has been revealed using social media-inspired mathematical tools to analyze space weather observations across the Earth's surface.

Analysis by researchers led by the University of Warwick has revealed that these substorms manifest as global-scale electrical current systems associated with the spectacular aurora, reaching across over a third of the globe at high latitude.

 Map representing a snapshot of the community structure at onset. CREDIT Background map credit: SuperMAG

New research which involves the University of Warwick, John Hopkins University - Applied Physics Laboratory, University of Bergen, and Cranfield University, and published today in an academic journal processes data on disturbances in the Earth's magnetic field from over a hundred magnetometers in the Northern hemisphere using a new technique that enables them to find 'like-minded friends'.

Magnetometers register changes in the Earth's magnetic field. When charged particles from our Sun bombard the Earth's magnetic field, it stores up energy like a battery. Eventually, this energy is released leading to large-scale electrical currents in the ionosphere which generate disturbances of magnetic fields on the ground. At extremes, this can disrupt power lines, electronic and communications systems, and technologies such as GPS.

Using historical data from the SuperMAG collaboration of magnetometers, the researchers applied algorithms from network science to find correlations between magnetometer signals during 41 known substorms that occurred between 1997-2001. These use the same principles that allow a social networking site to recommend new friends, or to push relevant advertisements to you as you browse the internet.

Magnetometers detecting coherent signals were linked into communities, regardless of where they were located on the globe. As time progressed, they saw each substorm develop from many smaller communities into a single large correlated system or community at its peak. This led the authors to conclude that substorms are one coherent current system that extends over most of the nightside high latitude globe, rather than a number of individual small and disjointed current systems. Map representing a snapshot of the community structure at the time of maximum auroral expansion. CREDIT Background map credit: SuperMAG

Dr. Lauren Orr, who led the research as part of her Ph.D. at the University of Warwick Department of Physics and is now based at Lancaster University, said: "We used a well-established method within network science called community detection and applied it to a space weather problem. The idea is that if you have lots of little subgroups within a big group, it can pick out the subgroups.

"We applied this to space weather to pick out groups within magnetometer stations on the Earth. From that, we were trying to find out whether there was one large current system or lots of separate individual current systems.

"This is a good way of letting the data tell us what's going on, instead of trying to fit observations to what we think is occurring."

Some recent work has suggested that auroral substorms are composed of a number of smaller electrical current systems and remain so throughout their lifecycle. This new research demonstrates that while the substorm begins as lots of smaller disturbances, it quite rapidly becomes a large system over the course of around ten minutes. The lack of correlation in its early stages may also suggest that there is no single mechanism at play in how these substorms evolve.

The results have implications for models designed to predict space weather. Space weather was included in the UK National Risk Register in 2012 and updated in 2017 with a recommendation for more investment in forecasting.

Co-author Professor Sandra Chapman adds: "Our research introduces a whole new methodology for looking at this data. We've gone from a data-poor to a data-rich era in space plasma physics and space weather, so we need new tools. It's a first to show that you can take one of these tools to our field and get a really important result out of it. We've had to learn a lot to be able to do that, but in doing so it opens up a new window into the data."

University of Helsinki's image analysis based on machine learning reliably identifies hematological malignancies

Tyler O'Neal, Staff Editor HEALTH

Myelodysplastic syndrome (MDS) is a disease of the stem cells in the bone marrow, which disturbs the maturing and differentiation of blood cells. Annually, some 200 Finns are diagnosed with MDS, which can develop into acute leukemia. Globally, the incidence of MDS is 4 cases per 100,000 person-years.

To diagnose MDS, a bone marrow sample is needed to also investigate genetic changes in bone marrow cells. The syndrome is classified into groups to determine the nature of the disorder in more detail.

In the study conducted at the University of Helsinki, microscopic images of MDS patients' bone marrow samples were examined utilizing an image analysis technique based on machine learning. The samples were stained with hematoxylin and eosin (H&E staining), a procedure that is part of the routine diagnostics for the disease. The slides were digitized and analyzed with the help of computational deep learning models.

The study was published in the Blood Cancer Discovery, a journal of the American Association for Cancer Research, and the results can also be explored with an interactive tool: http://hruh-20.it.helsinki.fi/mds_visualization/.

By employing machine learning, the digital image dataset could be analyzed to accurately identify the most common genetic mutations affecting the progression of the syndrome, such as acquired mutations and chromosomal aberrations. The higher the number of aberrant cells in the samples, the higher the reliability of the results generated by the prognostic models.

REFERENCE: BLOOD CANCER DISCOVERY, A JOURNAL OF THE AMERICAN ASSOCIATION FOR CANCER RESEARCH

Diagnosis supported by data analysis

One of the greatest challenges of utilizing neural network models is understanding the criteria on which they base their conclusions drawn from data, such as information contained in images. The recently released study succeeded in determining what deep learning models see in tissue samples when they have been taught to look for, for example, genetic mutations related to MDS. The technique provides new information on the effects of complex diseases on bone marrow cells and the surrounding tissues.

"The study confirms that computational analysis helps to identify features that elude the human eye. Moreover, data analysis helps to collect quantitative data on cellular changes and their relevance to the patient's prognosis," says Professor Satu Mustjoki.

Part of the analytics carried out in the study was implemented using the Helsinki University Hospital (HUS) data lake environment, which enables the efficient collection and analysis of extensive clinical datasets.

"We've developed solutions to structure and analyze data stored in the HUS data lake. Image analysis helps us analyze large quantities of biopsies and rapidly produce diverse information on disease progression. The techniques developed in the project are suited to other projects as well, and they are perfect examples of digitalizing medical science," says doctoral student Oscar Bruck.

"[This] study provides new insights into the pathobiology of MDS and paves the way for increased use of artificial intelligence for the assessment and diagnosis of hematological malignancies," says Ph.D. Olivier Elemento from the Caryl and Israel Englander Institute for Precision Medicine in his commentary to the article in Blood Cancer Discovery, a journal of the American Association for Cancer Research.

CyrusOne deploys second net-positive water data center

Tyler O'Neal, Staff Editor HEALTH

CyrusOne is collaborating with Bonneville Environmental Foundation and Trout Unlimited to promote water conservation and sustainability

CyrusOne has announced its second net-positive water data center at the Company’s Carrollton, TX location. The Carrollton data center is the single largest CyrusOne data center in the United States and is located in a high-water stress region as designated by the World Resources Institute. 

“Data center operators across the world are becoming more aware of the significant water usage at these facilities,” said Kyle Myers, Senior Director of Environmental Health, Safety and Sustainability at CyrusOne. “This is a great way to celebrate World Water Day on March 22 and a milestone moment for CyrusOne to further our goals on our path to a sustainable future.” 

Over the past year, water efficiency projects completed at the Carrollton location have reduced onsite water consumption by 67%. By leveraging our zero water-consumption cooling technology, CyrusOne will save millions of gallons of water annually in a drought-prone region. 

The collaboration with the Bonneville Environmental Foundation (BEF) and Trout Unlimited will restore to regional watersheds 20% more water than CyrusOne consumed at the site in 2020. BEF was CyrusOne’s partner in the Company’s first net-positive water facility in Chandler, Arizona, and they are pleased to work together again on the Carrollton project. Trout Unlimited partners with agencies, farmers, and water managers to flexibly manage, store, and deliver water during critical times of the years to increase flows in waterways and facilitate groundwater recharge. By leasing water, exchanging water at critical times, and shifting the timing of water delivery, project partners have been able to increase habitat for fish and provide important economic and community benefits for residents in the region. 

“Addressing water scarcity will require us to not only reduce consumptive water use but also restore water to nature,” said Val Fishman, Chief Development Officer at BEF. “We’re excited to have CyrusOne on board as a leader and partner helping to keep water in rivers during times of water stress. Through Water Restoration Credits, CyrusOne is supporting innovative projects on the Upper Rio Grande region that help farmers, agencies, and water managers restore water flows.” 

This project exemplifies CyrusOne’s commitment to minimize impacts to the local environment and communities in which we operate and also deliver benefits when possible. The Carrollton project is the next step toward our effort to become net water positive in high water stress regions across the company’s global data center portfolio. This is in addition to the announcement earlier this year that the facility will begin incorporating renewable energy in 2021, in support of its Zero Carbon by 2040 target.