The achievement, which was developed in collaboration with the first participant of a clinical research trial, builds on more than a decade of effort by UCSF neurosurgeon Edward Chang, MD, to develop a technology that allows people with paralysis to communicate even if they are unable to speak on their own. The study appears July 15 in the New England Journal of Medicine.

“To our knowledge, this is the first successful demonstration of direct decoding of full words from the brain activity of someone who is paralyzed and cannot speak,” said Chang, the Joan and Sanford Weill Chair of Neurological Surgery at UCSF, Jeanne Robertson Distinguished Professor, and senior author on the study. “It shows strong promise to restore communication by tapping into the brain’s natural speech machinery.”

Each year, thousands of people lose the ability to speak due to stroke, accident, or disease. With further development, the approach described in this study could one day enable these people to fully communicate.

Translating Brain Signals into Speech

Previously, work in the field of communication neuroprosthetics has focused on restoring communication through spelling-based approaches to type out letters one-by-one in text. Chang’s study differs from these efforts in a critical way: his team is translating signals intended to control muscles of the vocal system for speaking words, rather than signals to move the arm or hand to enable typing. Chang said this approach taps into the natural and fluid aspects of speech and promises more rapid and organic communication.

“With speech, we normally communicate information at a very high rate, up to 150 or 200 words per minute,” he said, noting that spelling-based approaches using typing, writing, and controlling a cursor are considerably slower and more laborious. “Going straight to words, as we’re doing here, has great advantages because it’s closer to how we normally speak.”

Over the past decade, Chang’s progress toward this goal was facilitated by patients at the UCSF Epilepsy Center who were undergoing neurosurgery to pinpoint the origins of their seizures using electrode arrays placed on the surface of their brains. These patients, all of whom had normal speech, volunteered to have their brain recordings analyzed for speech-related activity. Early success with these patient volunteers paved the way for the current trial in people with paralysis.

Previously, Chang and colleagues in the UCSF Weill Institute for Neurosciences mapped the cortical activity patterns associated with vocal tract movements that produce each consonant and vowel. To translate those findings into speech recognition of full words, David Moses, PhD, a postdoctoral engineer in the Chang lab and one of the lead authors of the new study, developed new methods for real-time decoding of those patterns and statistical language models to improve accuracy.

But their success in decoding speech in participants who were able to speak didn’t guarantee that the technology would work in a person whose vocal tract is paralyzed. “Our models needed to learn the mapping between complex brain activity patterns and intended speech,” said Moses. “That poses a major challenge when the participant can’t speak.”

In addition, the team didn’t know whether brain signals controlling the vocal tract would still be intact for people who haven’t been able to move their vocal muscles for many years. “The best way to find out whether this could work was to try it,” said Moses.

The First 50 Words

To investigate the potential of this technology in patients with paralysis, Chang partnered with colleague Karunesh Ganguly, MD, PhD, an associate professor of neurology, to launch a study known as “BRAVO” (Brain-Computer Interface Restoration of Arm and Voice). The first participant in the trial is a man in his late 30s who suffered a devastating brainstem stroke more than 15 years ago that severely damaged the connection between his brain and his vocal tract and limbs. Since his injury, he has had extremely limited head, neck, and limb movements, and communicates by using a pointer attached to a baseball cap to poke letters on a screen.

The participant, who asked to be referred to as BRAVO1, worked with the researchers to create a 50-word vocabulary that Chang’s team could recognize from brain activity using advanced computer algorithms. The vocabulary — which includes words such as “water,” “family,” and “good” — was sufficient to create hundreds of sentences expressing concepts applicable to BRAVO1’s daily life.

For the study, Chang surgically implanted a high-density electrode array over BRAVO1’s speech motor cortex. After the participant’s full recovery, his team recorded 22 hours of neural activity in this brain region over 48 sessions and several months. In each session, BRAVO1 attempted to say each of the 50 vocabulary words many times while the electrodes recorded brain signals from his speech cortex.

Translating Attempted Speech into Text

To translate the patterns of recorded neural activity into specific intended words, the other two lead authors of the study, Sean Metzger, MS and Jessie Liu, BS, both bioengineering doctoral students in the Chang Lab used custom neural network models, which are forms of artificial intelligence. When the participant attempted to speak, these networks distinguished subtle patterns in brain activity to detect speech attempts and identify which words he was trying to say.

To test their approach, the team first presented BRAVO1 with short sentences constructed from the 50 vocabulary words and asked him to try saying them several times. As he made his attempts, the words were decoded from his brain activity, one by one, on a screen.

Then the team switched to prompting him with questions such as “How are you today?” and “Would you like some water?” As before, BRAVO1’s attempted speech appeared on the screen. “I am very good,” and “No, I am not thirsty.”

The team found that the system was able to decode words from brain activity at rate of up to 18 words per minute with up to 93 percent accuracy (75 percent median). Contributing to the success was a language model Moses applied that implemented an “auto-correct” function, similar to what is used by consumer texting and speech recognition software.

Moses characterized the early trial results as a proof of principle. “We were thrilled to see the accurate decoding of a variety of meaningful sentences,” he said. “We’ve shown that it is actually possible to facilitate communication in this way and that it has potential for use in conversational settings.”

Looking forward, Chang and Moses said they will expand the trial to include more participants affected by severe paralysis and communication deficits. The team is currently working to increase the number of words in the available vocabulary, as well as improve the rate of speech.

Both said that while the study focused on a single participant and a limited vocabulary, those limitations don’t diminish the accomplishment. “This is an important technological milestone for a person who cannot communicate naturally,” said Moses, “and it demonstrates the potential for this approach to give a voice to people with severe paralysis and speech loss.”

Co-authors on the paper include Sean L. Metzger, MS; Jessie R. Liu; Gopala K. Anumanchipalli, PhD; Joseph G. Makin, PhD; Pengfei F. Sun, PhD; Josh Chartier, PhD; Maximilian E. Dougherty; Patricia M. Liu, MA; Gary M. Abrams, MD; and Adelyn Tu-Chan, DO, all of UCSF. Funding sources included National Institutes of Health (U01 NS098971-01), philanthropy, and a sponsored research agreement with Facebook Reality Labs (FRL), which completed in early 2021.

UCSF researchers conducted all clinical trial design, execution, data analysis and reporting. Research participant data were collected solely by UCSF, are held confidentially, and are not shared with third parties. FRL provided high-level feedback and machine learning advice.

The researchers trained 37 domestic dogs to operate a food dispenser by pressing a button, before separating the button and dispenser in separate enclosures. In the first stage, dogs were paired with two unfamiliar humans one at a time. One human partner was helpful — pressing their button to dispense food in the dog’s enclosure — and one was unhelpful. The researchers also reversed the set-up, with a button in the dog’s enclosure that operated a food dispenser in the human’s enclosure. They found no significant differences in the dogs’ tendency to press the button for helpful or unhelpful human partners, and the human’s behavior in the first stage did not affect the dog’s behavior towards them in free interaction sessions after the trials.

Previous studies have demonstrated that dogs are capable of directing helpful behaviors towards other dogs that have helped them previously — a behavior known as reciprocal altruism — and research suggests dogs are also able to distinguish between cooperative and uncooperative humans. However, the present study failed to find evidence that dogs can combine these capabilities to reciprocate help from humans. This finding may reflect a lack of ability or inclination among dogs to reciprocate, or the experimental design may not have detected it. For example, the authors suggest that the dogs may not have understood the experiment because humans are typically the food-giver in the relationship, not the receiver, or because the dogs failed to recognize the connection between the human’s helpful behavior and the reward.

The authors add: “In our study, pet dogs received food from humans but did not return the favour.”

Story Source:

Materials provided by PLOS. Note: Content may be edited for style and length.

Research professor Bruce Wilkinson and professor Linda Ivany, both from the Department of Earth and Environmental Sciences, recently co-authored a paper in the Biological Journal of the Linnean Society where they determined that forecasting the total number of species may never be possible.

When asking the question, ‘how many species?,’ it is important to note that only a fraction of existing species have been named. In order to make a prediction on a total number, researchers project the curve of new species descriptions each year into the future until eventually reaching a point when all species should have been found.

Wilkinson, a geologist, noticed parallels between the discovery curves of new species and the total reservoir size of nonrenewable resources like oil or mineral ores. Similar to the species curve, by extending the oil reservoir curve researchers thought they should be able to estimate the total global reservoir and how long it will take to get to it all. The theory of resource exploitation suggests that the number of discoveries over time follows a bell-shaped curve: The curve rises as production rate increases due to new discoveries and then decreases as production declines, despite all the effort continuing to go into finding the resource. The time of maximum discovery is known as Hubbert’s peak, after M. King Hubbert who predicted it. Following that time, the resource is being evermore depleted until it is used up.

“The problem with using that curve to predict how much is left is that you have to assume that the effort invested and the approach used to discover new oil, or species, is consistent and known,” says Wilkinson. “We used to think we’d gone over the peak for oil and gas around 1972, but then 15 or so years ago someone figured out how to do horizontal drilling and all of the sudden there was a new bump in the amount being discovered.”

Wilkinson and Ivany say that the discovery curve for new species of vertebrate animals shows a similar bump. Like the increase in the oil curve caused by horizontal drilling in the early 2000s, there was a surge in new species discovery beginning around 1950, when new funding was being dedicated to science after the World War II, more scientists were going into biology, and new molecular techniques were leading to an increase in the ability to distinguish species from one another.

In both cases, unforeseen changes in the effort and method of discovering new oil or species altered the way the discovery curves were playing out.

If researchers had estimated the total number of species based on data prior to 1950, their estimates would be much different from any estimate made today, and both would likely be wrong because those new advents cannot be predicted.

In some ways, this is a reflection of the scientific method, in which hypotheses stand until new facts are discovered, which lead to changes in the hypothesis.

“As much as we’d like to know ‘the number,’ the total species richness of the planet will remain an elusive target,” says Ivany.

Story Source:

Materials provided by Syracuse University. Original written by Dan Bernardi. Note: Content may be edited for style and length.

The study, led by investigators at St. Michael’s Hospital, a site of Unity Health Toronto, and the University of Vermont Larner College of Medicine, is available as a preprint on MedRxiv.

Heparin — a blood thinner given regularly at low dose to hospitalized patients — stops clots from forming and reduces inflammation. “This study was designed to detect a difference in the primary outcome that included ICU transfer, mechanical ventilation or death,” says Mary Cushman, M.D., M.Sc., study co-principal investigator and a professor of medicine at the University of Vermont’s Larner College of Medicine.

The open-label randomized international multi-center RAPID Trial (also known as the RAPID COVID COAG — RAPID Trial) examined the benefits of administering a therapeutic full dose of heparin versus a prophylactic low dose to moderately ill patients admitted to hospital wards with COVID-19.

The primary outcome was a composite of ICU admission, mechanical ventilation, or death up to 28 days. Safety outcomes included major bleeding. Primary outcome occurred in 37 of 228 patients (16.2%) with therapeutic full dose heparin, and 52 of 237 (21.9%) with low dose heparin (odds ratio [OR], 0.69; 95% confidence interval [CI], 0.43-1.10; p=0.12). Four patients (1.8%) with therapeutic heparin died vs. 18 (7.6%) with prophylactic heparin (OR, 0.22; 95% CI, 0.07-0.65).

“While we found that therapeutic heparin didn’t statistically significantly lower incidence of the primary composite of death, mechanical ventilation or ICU admission compared with low dose heparin, the odds of all-cause death were significantly reduced by 78 percent with therapeutic heparin,” says first author and co-principal investigator Michelle Sholzberg, M.D.C.M., M.Sc., Head of Division of Hematology-Oncology, medical director of the Coagulation Laboratory at St. Michael’s Hospital of Unity Health Toronto, and assistant professor at the University of Toronto.

Peter Jüni, M.D., co-principal investigator, director of the Applied Health Research Centre (AHRC) at St. Michael’s, and professor of medicine at the University of Toronto, says that the researchers also presented a meta-analysis of randomized evidence (including data from a large multiplatform trial of ATTACC, ACTIV-4a and REMAP-CAP), which clearly indicated that therapeutic heparin is beneficial in moderately ill hospitalized COVID-19 patients. He adds that an additional meta-analysis presented in the preprint showed that therapeutic heparin is beneficial in moderately ill hospitalized patients but not in severely ill ICU patients.

Another unique aspect of the RAPID Trial was its funding mechanism — a sort of grassroots effort in which support was gathered via Defence Research Development Canada, St. Michael’s Hospital Foundation, St. Joseph’s Healthcare Foundation, participating institutional grants, and even a GoFundMe campaign, among other sources.

“We called this trial ‘The Little Engine that Could,’ because of the sheer will of investigators around the world to conduct it,” says Cushman.

Sholzberg says, “We believe that the findings of our trial and the multiplatform trial taken together should result in a change in clinical practice for moderately ill ward patients with COVID-19.”

The study, published in Biological Conservation, evaluates potential tradeoffs between climate benefits and energy costs, especially any negative impacts on biodiversity. While the environmental consequences of some renewable energy sources, like hydropower, have been widely studied, the large-scale impacts of other renewables, like solar, aren’t well known.

“The study points to a need for the global community to understand the opposing endpoints of sustainability, which are scale-dependent,” McManamay said. “At one endpoint, efforts to mitigate climate change at global scales via large-scale energy transitions may be completely incognizant of the consequences on local biodiversity. Likewise, local conservationists may not appreciate the magnitude of energy transitions required to shift global carbon emissions. Finally, I think there is a broad misconception among much of the community that if renewable energies are good for climate, they must also be good for the local ecosystem. Energies like solar have the potential to be deployed haphazardly with that mindset.”

The challenges of climate mitigation in relation to different climate policy scenarios are projected in five Shared Socio-economic Pathways (SSPs) — qualitative descriptions of plausible alternative socio-economic development in the next century. The SSPs include alternative projections in deployment of electricity generations by technology. Considering scenarios under the SSPs, ecological footprinting was used to evaluate the potential land and biodiversity tradeoffs of 10 different energy sources: solar photovoltaic, concentrated solar power, wind, hydropower, coal, conventional oil, conventional gas, unconventional oil, unconventional gas, and biomass.

“Ecological footprinting took into account land-use efficiencies of each technology as well as estimates of the degree of habitat alteration arising from technology deployment,” McManamay said. “This provides a standardized way to compare the biodiversity consequences of large-scale deployment of alternative energy technologies.”

Researchers estimated a biodiversity footprint for each of the 10 energy sources by overlaying energy densities and habitat alteration probabilities with biodiversity patterns. They then used spatial modeling to examine regional variations in future energy deployment and potential biodiversity impacts at a high-resolution. Different biodiversity footprints were scored based on their impact and a cumulative biodiversity score was determined for each of the 10 energy sources.

The cumulative impact scores among the SSPs showed significant and consistent differences — the fossil-fueled development pathway (SSP 5) had the highest impacts while the regional rivalry scenario (SSP3) had the lowest. The sustainability-focused scenario (SSP1) represented a moderate impact score by comparison. Unexpectedly, the variation among SSPs didn’t show a clear tradeoff between global climate mitigation and cumulative biodiversity impact.

“It was surprising to see the lack of a clear tradeoffs among sustainability endpoints,” McManamay said. “This elicited us to take a deeper look into differences among the SSPs. Although SSP5 is termed ‘fossil-fueled’ development, the pathway includes significant technological advances in both advanced fossil and renewable technologies to meet highly consumptive, energy-luxurious lifestyles. In other words, the biodiversity impacts are more related to total energy deployment than fossil versus renewable technologies. Although SSP1 is also characterized by significant renewable energy deployment, overall energy demand decreases due to increases in energy efficiencies. So, our work suggests that climate mitigation may not necessarily have to be at odds with biodiversity conservation.”

Additionally, land constraints accounted for the most variation in biodiversity impact, particularly with regard to protected land use. Downscaled electricity generation scenarios were constrained by alternative land conservation and energy development policies.

The results offer an approximation of land and biodiversity impacts of future energy strategies outlined in the SSPs. While there were differences in the SSPs, the impact scores suggest that land protection measures and energy diversification could have greater implications for biodiversity challenges than the national-level global energy pathways outlined in the SSPs. Future planning and objectives for climate mitigation will require both broad and local consideration of biodiversity challenges.

Story Source:

Materials provided by Baylor University. Note: Content may be edited for style and length.

Significantly, the results from the research with human-derived neurons validated previous and new experiments that found the same major decrease in neurotransmitter release and synaptic signaling in genetically engineered human neurons with the same genetic variant — the deletion of neurexin 1 (NRXN1). NRXN1 is a protein-coding gene at the synapse, a cellular junction that connects two nerve cells to communicate efficiently.

Both the research with human-derived and engineered human neurons also found an increase in the levels of CASK, an NRXN1-binding protein, which were associated with changes in gene expression.

“Losing one copy of this neurexin 1 gene somehow contributes to the etiology or the disease mechanism in these schizophrenia patients,” says molecular neuroscientist ChangHui Pak, assistant professor of biochemistry and molecular biology at the University of Massachusetts Amherst and lead author of the research published in the Proceedings of the National Academy of Sciences. “It causes a deficit in neural communication.”

Pak is quick to add that although this single-gene mutation puts people at risk for schizophrenia, autism, Tourette syndrome and other neuropsychiatric disorders, “at the end of the day, we don’t know what causes schizophrenia. This variant gives us insight into what cellular pathways would be perturbed among people with schizophrenia and a lead to study this biology.”

When she conducted most of the research, Pak was working in the Stanford University lab of Thomas Südhof, a neuroscientist who shared the 2013 Nobel Prize in Physiology or Medicine for helping to lay the molecular basis for brain chemistry, including neurotransmitter release.

The research team obtained cell specimens from schizophrenia patients with an NRXN1 deletion who donated samples to a national biorepository for genetic studies of psychiatric disorders. Pak and colleagues converted the participants’ specimens into stem cells and then turned them into functional neurons to study. “We’re rewinding these cells back, almost like a time machine — what did these patients’ brains look like early on,” Pak explains.

Labs at Stanford, Rutgers University and FUJIFILM Cellular Dynamics were independently involved in the generation and analysis of neurons. For comparison with the human-derived neurons, Pak and team also created human neurons from embryonic stem cells, engineering them to have one less copy of the NRXN1 gene. With engineered human neurons, they had previously noted the neurotransmitter impairment and were interested in whether they would have the same findings with patient-derived neurons.

“It was good to see the consistent biological finding that indeed the neurexin 1 deletion in these patients actually does mess up their neuronal synaptic communication, and secondly that this is reproducible across different sites whoever does the experiment,” Pak says.

Notably, the researchers did not see the same decrease in neurotransmitter release and other effects in engineered mouse neurons with analogous NRXN1 deletion. “What this suggests is there is a human-specific component to this phenotype. The human neurons are particularly vulnerable to this genetic insult, compared to other organisms, adding to the value of studying human mutations in human cellular systems,” Pak says.

Being able to reproduce the results is key to the development of drugs that can better treat schizophrenia. “Everything was done blindly and at different sites. We wanted to not only learn about the biology but also be at the top of our game to ensure rigor and reproducibility of these findings,” Pak says. “We showed the field how this can be done.”

Pak and her team are now continuing the research in the Pak Lab, supported by a five-year, $2.25 million grant from the National Institute of Mental Health. The scientists are using the latest stem cell and neuroscience methodologies to explore the molecular basis of synaptic dysfunction in schizophrenia and other neuropsychiatric disorders.