Study links poor diet to 14 million cases of type 2 diabetes globally

Researchers from Tufts University estimate 7 out of 10 cases of type 2 diabetes worldwide in 2018 linked to food choices

Peer-Reviewed Publication

TUFTS UNIVERSITY

 

IMAGE: RESEARCHERS ESTIMATE 7 OUT OF 10 CASES OF TYPE 2 DIABETES WORLDWIDE IN 2018 LINKED TO FOOD CHOICES view more 

CREDIT: ALONSO NICHOLS/TUFTS UNIVERSITY

A research model of dietary intake in 184 countries, developed by researchers at the Friedman School of Nutrition Science and Policy at Tufts University, estimates that poor diet contributed to over 14.1 million cases of type 2 diabetes in 2018, representing over 70% of new diagnoses globally. The analysis, which looked at data from 1990 and 2018, provides valuable insight into which dietary factors are driving type 2 diabetes burden by world region. The study was published April 17 in the journal Nature Medicine.

Of the 11 dietary factors considered, three had an outsized contribution to the rising global incidence of type 2 diabetes: Insufficient intake of whole grains, excesses of refined rice and wheat, and the overconsumption of processed meat. Factors such as drinking too much fruit juice and not eating enough non-starchy vegetables, nuts, or seeds, had less of an impact on new cases of the disease.

“Our study suggests poor carbohydrate quality is a leading driver of diet-attributable type 2 diabetes globally, and with important variation by nation and over time,” says senior author Dariush Mozaffarian, Jean Mayer Professor of Nutrition and dean for policy at the Friedman School. “These new findings reveal critical areas for national and global focus to improve nutrition and reduce devastating burdens of diabetes.”

Type 2 diabetes is characterized by the resistance of the body’s cells to insulin. Of the 184 countries included in the Nature Medicine study, all saw an increase in type 2 diabetes cases between 1990 and 2018, representing a growing burden on individuals, families, and healthcare systems.

The research team based their model on information from the Global Dietary Database, along with population demographics from multiple sources, global type 2 diabetes incidence estimates, and data on how food choices impact people living with obesity and type 2 diabetes from multiple published papers.  

The analysis revealed that poor diet is causing a larger proportion of total type 2 diabetes incidence in men versus women, in younger versus older adults, and in urban versus rural residents at the global level.

Regionally, Central and Eastern Europe and Central Asia —particularly in Poland and Russia, where diets tend to be rich in red meat, processed meat, and potatoes —had the greatest number of type 2 diabetes cases linked to diet. Incidence was also high in Latin America and the Caribbean, especially in Colombia and Mexico, which was credited to high consumption of sugary drinks, processed meat, and low intake of whole grains.

Regions where diet had less of an impact on type 2 diabetes cases included South Asia and Sub-Sharan Africa —though the largest increases in type 2 diabetes due to poor diet between 1990 and 2018 were observed in Sub-Saharan Africa. Of the 30 most populated countries studied, India, Nigeria, and Ethiopia had the fewest case of type 2 diabetes related to unhealthy eating.

“Left unchecked and with incidence only projected to rise, type 2 diabetes will continue to impact population health, economic productivity, health care system capacity, and drive heath inequities worldwide,” says first author Meghan O’Hearn. She conducted this research while a PhD candidate at the Friedman School and currently works as Impact Director for Food Systems for the Future, a non-profit institute and for-profit fund that enables innovative food and agriculture enterprises to measurably improve nutrition outcomes for underserved and low-income communities. “These findings can help inform nutritional priorities for clinicians, policymakers, and private sector actors as they encourage healthier dietary choices that address this global epidemic.”

Other recent studies have estimated that 40% of type 2 diabetes cases globally are attributed to suboptimal diet, lower than the 70% reported in the Nature Medicine paper. The research team attributes this to the new information in their analysis, such as the first ever inclusion of refined grains, which was one of the top contributors to diabetes burdens; and updated data on dietary habits based on national individual-level dietary surveys, rather than agricultural estimates. The investigators also note that they presented the uncertainty of these new estimates, which can continue to be refined as new data emerges.

 

Research reported in this article was supported by the Bill and Melinda Gates Foundation. Complete information on authors, funders, methodology, and conflicts of interest is available in the published paper. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders.

---

JOURNAL

Nature Medicine

DOI

10.1038/s41591-023-02278-8 

ARTICLE TITLE

Incident type 2 diabetes attributable to suboptimal diet in 184 countries

ARTICLE PUBLICATION DATE

17-Apr-2023

New genetic target for male contraception identified

Peer-Reviewed Publication

WASHINGTON STATE UNIVERSITY

PULLMAN, Wash. – Discovery of a gene in multiple mammalian species could pave the way for a highly effective, reversible and non-hormonal male contraceptive for humans and animals.

Washington State University researchers identified expression of the gene, Arrdc5, in the testicular tissue of mice, pigs, cattle and humans. When they knocked out the gene in mice, it created infertility only in the males, impacting their sperm count, movement and shape. The researchers detailed their findings in the journal Nature Communications.

“The study identifies this gene for the first time as being expressed only in testicular tissue, nowhere else in the body, and it’s expressed by multiple mammalian species,” said Jon Oatley, senior author and professor in WSU’s School of Molecular Biosciences. “When this gene is inactivated or inhibited in males, they make sperm that cannot fertilize an egg, and that’s a prime target for male contraceptive development.”

While other molecular targets have been identified for potential male contraceptive development, the Arrdc5 gene is specific to the male testes and found in multiple species. Importantly, lack of the gene also causes significant infertility creating a condition called oligoasthenoteratospermia or OAT. This condition, the most common diagnosis for human male infertility, shows a decrease in the amount of sperm produced, slowed mobility and distorted shape so that the sperm are unable to fuse with an egg.

In the WSU study, the male mice lacking this gene produced 28% less sperm that moved 2.8 times slower than in normal mice – and about 98% of their sperm had abnormal heads and mid-pieces.

The study indicates that the protein encoded by this gene is required for normal sperm production. Oatley’s team will next work on designing a drug that would inhibit production or function of that protein.

Disrupting this protein wouldn’t require any hormonal interference, a key hurdle in male contraception since testosterone plays other roles beyond sperm production in men including building bone mass and muscle strength as well as red blood cell production. Designing a drug to target this protein would also make it easily reversible as a contraceptive.

“You don't want to wipe out the ability to ever make sperm – just stop the sperm that are being made from being made correctly,” he said. “Then, in theory, you could remove the drug and the sperm would start being built normally again.”

Oatley and study first author Mariana Giassetti have filed a provisional patent for the development of a male contraceptive based on this gene and the protein it encodes.

Because the gene is found across mammalian species, this knowledge also holds promise for use in animals, Oatley said. The team analyzed available biological data on DNA and protein sequences in mammals and found the gene in almost every known mammal species. This opens the potential to develop male contraception for use in livestock, perhaps replacing castration in some instances as a way to control reproduction, and in wildlife when managers seek to limit overpopulation of a species.

The initial focus, however, is on giving humans more control over their own reproduction. While there are many forms of birth control for women, they are not always effective or widely available, and more than half of pregnancies worldwide are still unintended, according to the United Nations.

“Developing a way to curb population growth and stop unwanted pregnancies is really important for the future of the human race,” said Oatley. “Right now, we don't really have anything on the male side for contraception other than surgery and only a small percentage of men choose vasectomies. If we can develop this discovery into a solution for contraception, it could have far-ranging impacts.”

This study received support from the National Institutes of Health and WSU’s Functional Genomics Initiative, a multi-year university investment to support development of genetic technology research.  

---

JOURNAL

Nature Communications

DOI

10.1038/s41467-023-37735-y 

ARTICLE PUBLICATION DATE

17-Apr-2023

UCF scientist publishes book on emergence of new pathogens

Climate change may be linked to an increase in the emergence of new pandemics, according to a new book published by an internationally recognized College of Medicine microbiologist

Book Announcement

UNIVERSITY OF CENTRAL FLORIDA

 

IMAGE: DR. SALVADOR ALMAGRO-MORENO HOLDS HIS RECENTLY PUBLISHED BOOK VIBRIO SPP. INFECTIONS. view more 

CREDIT: UNIVERSITY OF CENTRAL FLORIDA

Climate change may be linked to an increase in the emergence of new pandemics, according to a new book published by an internationally recognized College of Medicine microbiologist. 

Dr. Salvador Almagro-Moreno has teamed with fellow molecular biologist Dr. Stefan Pukatzki of City University New York – CUNY, to author the book titled Vibro spp. Infections, recently published by Springer Nature and includes the latest scientific research articles in this field from experts worldwide.

The book examines the factors associated with the increases in infections caused by pathogenic members of the Vibrionaceae family of aquatic bacteria. A group that encompasses species that cause cholera, flesh-eating disease and diarrhea in humans and can also devastate aquaculture farms.

“Recent decades have seen a steady increase in Vibrio spp. infections originating in aquatic and marine habitats, driven by higher human population densities, warming of polluted oceans, natural and human-made disasters and mass seafood production,” said Dr. Almagro-Moreno.

The book addresses how climate change affects the spread of these bacteria and how pollution and warmer ocean temperatures may be impacting emerging pathogenic strains. It also discusses the current status of vaccines to treat these bacteria and novel therapeutics that may help treat their infections. It lays out the foundation for future studies and covers timely issues in the post- COVID-19 era that can help researchers understand and better predict new emerging pathogens.

Dr. Almagro-Moreno started his research career understanding the spread of the cholera bacterium and then began exploring other Vibrionaceae that inhabit the Florida coastlines such as the deadly flesh-eating bacterium Vibrio vulnificus. His goal is to identify the molecular and environmental factors that allow these bacteria to survive, proliferate, and in some cases, turn deadly to humans. 

Dr. Griff Parks, associate dean for research and director of the Burnett School of Biomedical Sciences, said Dr. Almagro-Moreno’s expertise is key as the world seeks to understand emergent pathogens like COVID-19 and better prepare for future pandemics.

“Dr. Almagro-Moreno continues to carry out highly impactful research on topics that we do not understand very well,” he said, “including the critical question of what environmental and social factors direct the emergence of human pathogens.”

BY SUHTLING WONG 

Webb captures the spectacular galactic merger Arp 220

Reports and Proceedings

NASA/GODDARD SPACE FLIGHT CENTER

 

IMAGE: A STUNNING SMASH-UP OF TWO SPIRAL GALAXIES SHINES IN INFRARED WITH THE LIGHT OF MORE THAN A TRILLION SUNS. COLLECTIVELY CALLED ARP 220, THE COLLIDING GALAXIES IGNITED A TREMENDOUS BURST OF STAR BIRTH. EACH OF THE COMBINING GALACTIC CORES IS ENCIRCLED BY A ROTATING, STAR-FORMING RING BLASTING OUT THE GLARING LIGHT THAT WEBB CAPTURED IN INFRARED. THIS BRILLIANT LIGHT CREATES A PROMINENT, SPIKED, STARBURST FEATURE. view more 

CREDIT: CREDITS: NASA, ESA, CSA, STSCI, ALYSSA PAGAN (STSCI)

Shining like a brilliant beacon amidst a sea of galaxies, Arp 220 lights up the night sky in this view from NASA’s James Webb Space Telescope. Actually two spiral galaxies in the process of merging, Arp 220 glows brightest in infrared light, making it an ideal target for Webb. It is an ultra-luminous infrared galaxy (ULIRG) with a luminosity of more than a trillion suns. In comparison, our Milky Way galaxy has a much more modest luminosity of about ten billion suns.

 

Located 250 million light-years away in the constellation of Serpens, the Serpent, Arp 220 is the 220th object in Halton Arp’s Atlas of Peculiar Galaxies. It is the nearest ULIRG and the brightest of the three galactic mergers closest to Earth.

 

The collision of the two spiral galaxies began about 700 million years ago. It sparked an enormous burst of star formation. About 200 huge star clusters reside in a packed, dusty region about 5,000 light-years across (about 5 percent of the Milky Way's diameter). The amount of gas in this tiny region is equal to all of the gas in the entire Milky Way galaxy.

 

Previous radio telescope observations revealed about 100 supernova remnants in an area of less than 500 light-years. NASA’s Hubble Space Telescope uncovered the cores of the parent galaxies 1,200 light-years apart. Each of the cores has a rotating, star-forming ring blasting out the dazzling infrared light so apparent in this Webb view. This glaring light creates diffraction spikes — the starburst feature that dominates this image.

On the outskirts of this merger, Webb reveals faint tidal tails, or material drawn off the galaxies by gravity, represented in blue — evidence of the galactic dance that is occurring. Organic material represented in reddish-orange appears in streams and filaments across Arp 220.

 

Webb viewed Arp 220 with its Near-Infrared Camera (NIRCam) and Mid-Infrared Instrument (MIRI).

 

The James Webb Space Telescope is the world’s premier space science observatory. Webb will solve mysteries in our solar system, look beyond to distant worlds around other stars, and probe the mysterious structures and origins of our universe and our place in it. Webb is an international program led by NASA with its partners, ESA (European Space Agency) and the Canadian Space Agency.

James Webb Space Telescope keeps findings galaxies that shouldn’t exist, scientist warns

Story by Andrew Griffin • Yesterday



ngc1433.png© NASA, ESA, CSA, and J. Lee (NOIRLab). Image processing: A. Pagan (STScI)

The James Webb Space Telescope keeps finding galaxies that shouldn’t exist, a scientist has warned.

Six of the earliest and most massive galaxies that Nasa’s breakthrough telescope has seen so far appear to be bigger and more mature than they should be given where they are in the universe, researchers have warned.

The new findings build on previous research where scientists reported that despite coming from the very beginnings of the universe, the galaxies were as mature as our own Milky Way.

Now a new paper has appeared to confirm those findings, by “stress testing” the galaxies to better understand how they formed.

It suggests that, if scientists have not made a mistake, we may be missing some fundamental information about the universe.

“If the masses are right, then we are in uncharted territory,” said Mike Boylan-Kolchin, from the University.of Texas at Austin, and the author of a new paper examining the unsual galaxies. “We’ll require something very new about galaxy formation or a modification to cosmology. One of the most extreme possibilities is that the universe was expanding faster shortly after the Big Bang than we predict, which might require new forces and particles.”

It suggests that the information from the JWST proposes a profound dilemma for scientists. The data indicates that there mighttbe somehitn wrong with the dark energy and cold dark matter paradigm, or ΛCDM, that has been guiding cosmology for decades.

Usually, galaxies convert around 10 per cent of their gas into stars. But the newly discovered galaxies would have to be converting almost the entirety of it into stars.

That is theoretically possible. But it is a departure from what scientists would ever have expected.

Further observation of the galaxies should better clarify their ages and masses. It might show that the observations are incorrect: that supermassive black holes at their centre are heating the galaxies up, so they look more massive than they are, or that they are actually from a later time than expected but look older because of imaging problems.

But if they are confirmed, then astronomers may have to change their understanding of the cosmos and how galaxies grow, to adjust their model to account for the unusually large and mature galaxies.

Mission Ax-2 set to launch stem cells to space

Upcoming missions will be the first time induced pluripotent stem cells will be produced in microgravity by astronauts

Business Announcement

CEDARS-SINAI MEDICAL CENTER

 

VIDEO: IN EARLY MAY, A SPACEX FALCON 9 ROCKET WILL LAUNCH THE AX-2 CREW ABOARD A DRAGON SPACECRAFT TO THE INTERNATIONAL SPACE STATION FROM LAUNCH COMPLEX 39A AT NASA’S KENNEDY SPACE CENTER IN FLORIDA. view more 

CREDIT: CEDARS-SINAI

How music can prevent cognitive decline

A team from UNIGE, HES-SO Geneva and EPFL shows the positive impacts of musical activities to counteract brain ageing.

Peer-Reviewed Publication

UNIVERSITÉ DE GENÈVE

 

IMAGE: SIDE VIEW OF A BRAIN. IN BLUE, THE AREAS AFFECTED BY THE INCREASE IN GREY MATTER IN THE ELDERLY AS A RESULT OF MUSIC PRACTICE. view more 

CREDIT: © UNIGE - DAMIEN MARIE

Normal ageing is associated with progressive cognitive decline. But can we train our brain to delay this process? A team from the University of Geneva (UNIGE), HES-SO Geneva and EPFL has discovered that practicing and listening to music can alter cognitive decline in healthy seniors by stimulating the production of grey matter. To achieve these results, the researchers followed over 100 retired people who had never practiced music before. They were enrolled in piano and music awareness training for six months. These results open new prospects for the support of healthy ageing. They are reported in NeuroImage: Reports.

Throughout our lives, our brain remodels itself. Brain morphology and connections change according to the environment and the experiences, for instance when we learn new skills or overcome the consequences of a stroke. However, as we age, this ‘‘brain plasticity’’ decreases. The brain also loses grey matter, where our precious neurons are located. This is known as ‘‘brain atrophy’’.

Gradually, a cognitive decline appears. Working memory, at the core of many cognitive processes, is one of the cognitive functions suffering the most. Working memory is defined as the process in which we briefly retain and manipulate information in order to achieve a goal, such as remembering a telephone number long enough to write it down or translating a sentence from a foreign language.

A study led by the UNIGE, HES-SO Geneva, and EPFL revealed that music practice and active listening could prevent working memory decline. Such activities promoted brain plasticity, they were associated with grey matter volume increase. Positive impacts have also been measured on working memory. This study was conducted among 132 healthy retirees from 62 to 78 years of age. One of the conditions for participation was that they had not taken any music lessons for more than six months in their lives.

Practicing music vs. listening to music

‘‘We wanted people whose brains did not yet show any traces of plasticity linked to musical learning. Indeed, even a brief learning experience in the course of one’s life can leave imprints on the brain, which would have biased our results’’, explains Damien Marie, first author of the study, a research associate at the CIBM Center for Biomedical Imaging, the Faculty of Medicine and the Interfaculty Center for Affective Sciences (CISA) of UNIGE, as well as at the Geneva School of Health Sciences.

The participants were randomly assigned to two groups, regardless of their motivation to play an instrument. The second group had active listening lessons, which focused on instrument recognition and analysis of musical properties in a wide range of musical styles. The classes lasted one hour. Participants in both groups were required to do homework for half an hour a day.

Positive effects on both groups

‘‘After six months, we found common effects for both interventions. Neuroimaging revealed an increase in grey matter in four brain regions involved in high-level cognitive functioning in all participants, including cerebellum areas involved in working memory. Their performance increased by 6% and this result was directly correlated to the plasticity of the cerebellum,’’ says Clara James, last author of the study, a privat-docent at the Faculty of Psychology and Educational Sciences of UNIGE, and full professor at the Geneva School of Health Sciences. The scientists also found that the quality of sleep, the number of lessons followed over the course of the intervention, and the daily training quantity, had a positive impact on the degree of improvement in performance.

However, the researchers also found a difference between the two groups. In the pianists, the volume of grey matter remained stable in the right primary auditory cortex - a key region for sound processing, whereas it decreased in the active listening group. ‘‘In addition, a global brain pattern of atrophy was present in all participants. Therefore, we cannot conclude that musical interventions rejuvenate the brain. They only prevent ageing in specific regions,’’ says Damien Marie.

These results show that practicing and listening to music promotes brain plasticity and cognitive reserve. The authors of the study believe that these playful and accessible interventions should become a major policy priority for healthy ageing. The next step for the team is to evaluate the potential of these interventions in people with mild cognitive impairment, an intermediate stage between normal ageing and dementia.

---

JOURNAL

NeuroImage

DOI

10.1016/j.ynirp.2023.100166 

METHOD OF RESEARCH

News article

SUBJECT OF RESEARCH

People

ARTICLE TITLE

Music interventions in 132 healthy older adults enhance cerebellar grey matter and auditory working memory, despite general brain atrophy

Two brain networks are activated while reading, study finds

Peer-Reviewed Publication

UNIVERSITY OF TEXAS HEALTH SCIENCE CENTER AT HOUSTON

When a person reads a sentence, two distinct networks in the brain are activated, working together to integrate the meanings of the individual words to obtain more complex, higher-order meaning, according to a study at UTHealth Houston.

The study, led by Oscar Woolnough, PhD, postdoctoral research fellow in the Vivian L. Smith Department of Neurosurgery with McGovern Medical School at UTHealth Houston, and Nitin Tandon, MD, professor and chair ad interim of the department in the medical school, was published today in The Proceedings of the National Academy of Sciences (PNAS).

“This study helps us better understand how distributed hubs in the brain’s language network work together and interact to allow us to understand complex sentences,” said Woolnough, first author on the study and member of the Texas Institute for Restorative Neurotechnologies (TIRN) at UTHealth Houston. “Our brains are remarkably interconnected, and for us to understand language requires a precise sequence of rapid, dynamic processes to occur in multiple sites all across our brain.”

In order to identify the specific roles and interactions of the brain areas involved in reading, the research team performed recordings from the brains of patients with electrodes surgically placed to localize epilepsy. The neural activity of these patients was measured while reading three forms of sentences: regular sentences; “Jabberwocky” sentences (based on Lewis Carroll’s “Jabberwocky” poem), which use correct grammar and syntax but contain nonsense words, making them meaningless; and lists of words or nonsense words.

From these recordings, they identified two brain networks that play a key role in the reading process. One network involves a region of the brain’s frontal lobe that sends signals to the temporal lobe, which shows progressive activation when a person is building up complex meaning along the length of a sentence.

The second network involves another region of the brain’s temporal lobe that sends signals to an area of the frontal lobe, allowing understanding of the context of a sentence to enable easier comprehension and processing of each new word that is read.

“Implanted electrodes in the brain provide us an unparalleled insight into the inner workings of the human mind, especially for processes that are rapid, such as reading. Our work is making it clear that most processes – say comprehension or language generation – don’t occur in a single region, but are best understood as very transient states that many separate areas of the brain achieve by very brief, yet critical, interactions,” said Tandon, the study’s senior author, who is also the Nancy, Clive and Pierce Runnels Distinguished Chair in Neuroscience of the Vivian L. Smith Center for Neurologic Research and the BCMS Distinguished Professor in Neurological Disorders and Neurosurgery with McGovern Medical School.

Understanding the science behind the highly rapid, complex process of reading will allow the researchers to learn more about how the brain functions during dyslexia. Ultimately, they hope their findings will help guide treatment options for the reading disorder, which affects approximately 15% of people living in the U.S.

The research was funded through a five-year, $4.4 million grant from the National Institutes of Health Brain Research Through Advancing Neurotechnologies (BRAIN) Initiative, which aims to accelerate the development and application of innovative technologies to produce a new dynamic picture of the human brain.

Co-authors on the paper with McGovern Medical School’s neurosurgery department included Cristian Donos, PhD, who is now with the University of Bucharest in Romania; Elliot Murphy, PhD; Patrick Rollo; and Zachary Roccaforte. Murphy, Rollo, Roccaforte, and Tandon are also members of TIRN, and Tandon is a faculty member with The University of Texas MD Anderson UTHealth Houston Graduate School of Biomedical Sciences. Also contributing was Stanislas Dehaene, PhD, with Universite Paris-Saclay in France and the College de France.

---

JOURNAL

Proceedings of the National Academy of Sciences

ARTICLE TITLE

Spatiotemporally Distributed Frontotemporal Networks for Sentence Reading

ARTICLE PUBLICATION DATE

17-Apr-2023

Texas Children’s and Baylor College researchers use innovative dual-target deep brain stimulation approach to treat patients with obsessive-compulsive disorder and Tourette Syndrome

Peer-Reviewed Publication

TEXAS CHILDREN'S HOSPITAL

Up to two-thirds of patients with Tourette syndrome (TS), a tic disorder characterized by sudden uncontrollable physical movements, also suffer from obsessive-compulsive disorder (OCD), a psychiatric condition characterized by intrusive thoughts and repetitive behaviors. Unfortunately, many of these dual-diagnosis patients are resistant to conventional treatments such as medications or behavioral therapy. While deep brain stimulation (DBS) has been approved for compassionate use by the U.S. Food and Drug Administration for OCD, this promising procedure is under investigational use for TS.

Dr. Sameer Sheth, Dr. Wayne Goodman, Dr. Steven Bellows, and their colleagues at the Jan and Dan Duncan Neurological Research Institute (Duncan NRI) at Texas Children’s Hospital and Baylor College of Medicine, have recently demonstrated favorable outcomes using a new DBS approach that targets two distinct brain regions to simultaneously treat OCD and TS. The study, published in Biological Psychiatry, not only demonstrated the feasibility and effectiveness of using dual-target DBS to simultaneously treat these disorders but also advances understanding of the functional bases of these complex neuropsychological conditions.

“It is challenging to treat severely affected dual-diagnosis patients, many of whom do not respond to current treatments. Unfortunately, several previous attempts at using standard DBS approaches focused on a single brain target have failed to show marked sustained improvement in this cohort of patients,” said Dr. Sheth, who is also the director of Cain Foundation Labs for Pediatric Neurology at Duncan NRI and professor in the department of neurosurgery at Baylor College. “For this reason, we decided to simultaneously target two distinct regions of the brain – one implicated in OCD and the other in TS.”

A multidisciplinary team of experts selected two patients as ideal candidates for testing this dual-target DBS surgery. Both patients presented with severe forms of both conditions and, despite extensive treatment histories, had not shown sustained clinical improvements. Dr. Sheth implanted a pair of DBS electrodes in the bilateral ventral capsule/ventral striatum (VC/VS) region to treat OCD and another pair in the posteroventral (motor) globus pallidus internus brain area to treat TS using an established robotic procedure for DBS.

The new dual-target DBS strategy allowed Dr. Goodman and Dr. Bellows the flexibility to program the two DBS devices independently in order to achieve optimal and sustained improvement in severe OCD and TS symptoms in both patients.

Additionally, the DBS system they used not only stimulated these brain regions but also utilized recently available technology to record neural activity at different symptom states on the DBS device. These brain recordings allowed them to get a snapshot of the neural activity in that region at specific times (i.e. when OCD or TS symptoms were better or worse). The advantage of such a system is that it can be used to study the disease-modifying effects of DBS therapy on OCD- and TS-associated neurophysiology and thereby, optimize the delivery of this therapy in the future.

“The eventual goal is to move towards a ‘closed-loop’ DBS which will function analogous to how the thermostat in our homes regulates temperature,” Dr. Sheth added. “When we set a particular temperature, the system automatically regulates the heating and cooling to ensure that the desired temperature is achieved and maintained. Similarly, as we understand the desired pattern of brain activity associated with healthy, asymptomatic states, we can train the DBS device to automatically adjust stimulation parameters in order to achieve and maintain that desired state. In order to move towards this goal, we need a better understanding of the intricate relationships between changes in neural signals and disease symptoms, which is what studies like this are helping build.”

Others involved in the study were Ricardo Andres Najera, Nicole Provenza, Huy Dang, Kalman Katlowitz, Alyssa Hertz, Sandesh Reddy, Ben Shofty, and Eric Storch. They are affiliated with one or more of the following institutions: Baylor College of Medicine and Texas Children’s Hospital. The study was funded by grants from the National Institutes of Health and supported by the McNair Foundation.

---

JOURNAL

Biological Psychiatry

DOI

10.1016/j.biopsych.2023.01.014 

ARTICLE TITLE

Dual-Target Deep Brain Stimulation for Obsessive-Compulsive Disorder and Tourette Syndrome

Distinguished scientists highlight vital role of Science and “techno-pragmatism” for the future of humanity at Inaugural HKU President’s Forum

Meeting Announcement

THE UNIVERSITY OF HONG KONG

 

IMAGE: INAUGURAL HKU PRESIDENT'S FORUM view more 

CREDIT: THE UNIVERSITY OF HONG KONG

At the inaugural HKU President’s Forum held today (April 3) at Loke Yew Hall, the University of Hong Kong (HKU), four pioneering scientists and Professor Xiang Zhang, the President and Vice-Chancellor of HKU and convenor of the Forum, stressed the crucial role science will play in safeguarding the future of humanity and tackling pertinent global issues including climate change, sustainable growth and aging populations.

The renowned speakers were Professor Dame Carol M Black, a respected clinician and UK government advisor; Professor Steven Chu, Nobel Laureate in Physics and former Secretary of Energy of the US government; Dr Jack Dangermond, an environmental scientist and creator of the Geographic Information Systems (GIS) that have been applied to almost every field of human activity; and Professor John Hennessy, a former President of Stanford University, pioneer in computer architecture and winner of the Turing Prize, the Nobel Prize equivalent in computing.

Professor Chu, a passionate advocate for clean energy, spoke on the progress made in his own laboratory and other companies in the States on reducing carbon emissions, and enhancing the use and storage of renewable energy – in alignment with the global goal to achieve carbon neutrality by 2060. He said universities can play an important role in trying to work with the government on the most efficient policies that create the least political resistance, as “scientists and engineers might not be the most astute at determining that”. He also supports helping poorer nations gain access to technology that can fuel their economic growth, noting that: “I think the richer countries should and must help the poorer countries in responding to their changing climates. They do have an obligation.”

Professor Black is adamant in using her knowledge to influence government policy-making for social good, regardless of pressure from the opposition or powerful lobbyists. She had made recommendations to the UK Government on a range of issues including mental health, drug use, workforce, young people and inadequate aftercare. Her long years of research has uncovered the causes and treatment for the potentially fatal disease of systemic sclerosis. Three decades of research work, she noted, has shown “how basic science, technology, healthcare infrastructure and effective collective effort can transform the most lethal complication of the most lethal rheumatic disease.”

Dr Dangermond exalted the transformation technological advancement has brought to the world. As a result of the Internet and other developments, our world today is complex and interlocked, resembling a kind of eco-system. Yet sustainable growth remains a huge challenge, as the threats of overpopulation, climate change, fragile ecosystem loom. He emphasized the importance of cross-discipline, cross-sector collaborations in creating a sustainable future. “The rapidly emerging Global Geospatial Infrastructure will profoundly transform how we manage our world,” he noted.

Professor Hennessy, a driving force behind artificial intelligence (AI), believes in the contributions AI can make in further improving human lives. Self-driving cars will certainly become reality, alongside other developments resulting from breakthroughs in deep learning, such as detailed image analysis.

In response to the issue about whether AI will become smarter than humans and the potential loss of jobs from its widespread adoption, Professor Hennessy remarked: “Our obligation as a society is to help train people so that they can use this new technology, take advantage of it and find useful and gainful employment,” he added. He also insisted that humans still have an important role to play in ensuring proper analysis or diagnosis, e.g. in critical medical cases. “Having critical thinking is important,” he said.

HKU President Professor Xiang Zhang echoed the importance of collaboration in tackling global issues: “Let us keep the realm of sciences and technology as open, accessible, and non-partisan as possible, despite the increasing geopolitical tensions that we cannot ignore.”

“I believe firmly that the right approach to new scientific development should be neither overwhelming pessimism nor optimism, but instead, techno-pragmatism: pragmatism when it comes to new knowledge, is absolutely vital in unlocking the multilateral, multi-stakeholder solutions to global challenges.”

Also, he stressed the key role a cosmopolitan city with global networks like Hong Kong plays in driving forward innovation and technological growth, and the key part HKU plays in relation to that.

An archive of the Forum on Youtube: https://www.youtube.com/live/fa1eMn2kf2U?feature=share
RTHK 32 (TV channel) will rerun the Forum at 2pm, April 8, Saturday.

Please click here for more photos and short videos of the Forum.

Optimization could cut the carbon footprint of AI training by up to 75%

Deep learning models that power giants like TikTok and Amazon, as well as tools like ChatGPT, could save energy without new hardware or infrastructure.

Reports and Proceedings

UNIVERSITY OF MICHIGAN

Images  //  Zeus Demo Video

A new way to optimize the training of deep learning models, a rapidly evolving tool for powering artificial intelligence, could slash AI's energy demands. 

Developed at the University of Michigan, the open-source optimization framework studies deep learning models during training, pinpointing the best tradeoff between energy consumption and the speed of the training.

"At extreme scales, training the GPT-3 model just once consumes 1,287 MWh, which is enough to supply an average U.S. household for 120 years," said Mosharaf Chowdhury, an associate professor of electrical engineering and computer science.

With Zeus, the new energy optimization framework developed by Chowdhury and his team, figures like this could be reduced by up to 75% without any new hardware—and with only minor impacts on the time it takes to train a model. It was presented at the 2023 USENIX Symposium on Networked Systems Design and Implementation (NSDI), in Boston. 

Mainstream uses for hefty deep learning models have exploded over the past three years, ranging from image-generation models and expressive chatbots to the recommender systems powering TikTok and Amazon. With cloud computing already out-emitting commercial aviation, the increased climate burden from artificial intelligence is a significant concern. 

"Existing work primarily focuses on optimizing deep learning training for faster completion, often without considering the impact on energy efficiency," said Jae-Won Chung, a doctoral student in computer science and engineering and co-first author of the study. "We discovered that the energy we're pouring into GPUs is giving diminishing returns, which allows us to reduce energy consumption significantly, with relatively little slowdown."

Deep learning is a family of techniques making use of multilayered, artificial neural networks to tackle a range of common machine learning tasks. These are also known as deep neural networks (DNNs). The models themselves are extremely complex, learning from some of the most massive data sets ever used in machine learning. Because of this, they benefit greatly from the multitasking capabilities of graphical processing units (GPUs), which burn through 70% of the power that goes into training one of these models.

Zeus uses two software knobs to reduce energy consumption. One is the GPU power limit, which lowers a GPU's power use while slowing down the model's training until the setting is adjusted again. The other is the deep learning model's batch size parameter, which controls how many samples from the training data the model works through before updating the way the model represents the relationships it finds in the data. Higher batch sizes reduce training time, but with increased energy consumption.

Zeus is able to tune each of these settings in real time, seeking the optimal tradeoff point at which energy usage is minimized with as little impact on training time as possible. In examples, the team was able to visually demonstrate this tradeoff point by showing every possible combination of these two parameters. While that level of thoroughness won't happen in practice with a particular training job, Zeus will take advantage of the repetitive nature of machine learning to come very close.

"Fortunately, companies train the same DNN over and over again on newer data, as often as every hour. We can learn about how the DNN behaves by observing across those recurrences," said Jie You, a recent doctoral graduate in computer science and engineering and co-lead author of the study.

Zeus is the first framework designed to plug into existing workflows for a variety of machine learning tasks and GPUs, reducing energy consumption without requiring any changes to a system's hardware or datacenter infrastructure. 

In addition, the team has developed complementary software that they layer on top of Zeus to reduce the carbon footprint further. This software, called Chase, privileges speed when low-carbon energy is available, and chooses efficiency at the expense of speed during peak times, which are more likely to require ramping up carbon-intensive energy generation such as coal. Chase took second place at last year's CarbonHack hackathon and is to be presented May 4 at the International Conference on Learning Representations Workshop.

"It is not always possible to readily migrate DNN training jobs to other locations due to large dataset sizes or data regulations," said Zhenning Yang, a master's student in computer science and engineering. "Deferring training jobs to greener time frames may not be an option either, since DNNs must be trained with the most up-to-date data and quickly deployed to production to achieve the highest accuracy.

"Our aim is to design and implement solutions that do not conflict with these realistic constraints, while still reducing the carbon footprint of DNN training." 

The study was supported in part by the National Science Foundation grants CNS-1909067 and CNS-2104243, VMWare and the Kwanjeong Educational Foundation, and computing credits provided by CloudLab and Chameleon Cloud.

 Study: Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training

Study: Chasing Low-Carbon Electricity for Practical and Sustainable DNN Training

Open-source software:

Zeus on GitHub

Chase on Github