Making AI a partner in neuroscientific discovery
New paper argues that Large Language Models can reveal breakthroughs humans alone cannot
The past year has seen major advances in Large Language Models (LLMs) such as ChatGPT. The ability of these models to interpret and produce human text sources (and other sequence data) has implications for people in many areas of human activity. A new perspective paper in the journal Neuron argues that like many professionals, neuroscientists can either benefit from partnering with these powerful tools or risk being left behind.
In their previous studies, the authors showed that important preconditions are met to develop LLMs that can interpret and analyze neuroscientific data like ChatGPT interprets language. These AI models can be built for many different types of data, including neuroimaging, genetics, single-cell genomics, and even hand-written clinical reports.
In the traditional model of research, a scientist studies previous data on a topic, develops new hypotheses and tests them using experiments. Because of the massive amounts of data available, scientists often focus on a narrow field of research, such as neuroimaging or genetics. LLMs, however, can absorb more neuroscientific research than a single human ever could. In their Neuron paper, the authors argue that one day LLMs specialized in diverse areas of neuroscience could be used to communicate with one another to bridge siloed areas of neuroscience research, uncovering truths that would be impossible to find by humans alone. In the case of drug development, for example, an LLM specialized in genetics could be used along with a neuroimaging LLM to discover promising candidate molecules to stop neurodegeneration. The neuroscientist would direct these LLMs and verify their outputs.
Lead author Danilo Bzdok mentions the possibility that the scientist will, in certain cases, not always be able to fully understand the mechanism behind the biological processes discovered by these LLMs.
“We have to be open to the fact that certain things about the brain may be unknowable, or at least take a long time to understand,” he says. “Yet we might still generate insights from state-of-the-art LLMs and make clinical progress, even if we don’t fully grasp the way they reach conclusions.”
To realize the full potential of LLMs in neuroscience, Bzdok says scientists would need more infrastructure for data processing and storage than is available today at many research organizations. More importantly, it would take a cultural shift to a much more data-driven scientific approach, where studies that rely heavily on artificial intelligence and LLMs are published by leading journals and funded by public agencies. While the traditional model of strongly hypothesis-driven research remains key and is not going away, Bzdok says capitalizing on emerging LLM technologies might be important to spur the next generation of neurological treatments in cases where the old model has been less fruitful.
“To quote John Naisbitt, neuroscientists today are ‘drowning in information but starving for knowledge,’” he says. “Our ability to generate biomolecular data is eclipsing our ability to glean understanding from these systems. LLMs offer an answer to this problem. They may be able to extract, synergize and synthesize knowledge from and across neuroscience domains, a task that may or may not exceed human comprehension.”
The Neuro – The Montreal Neurological Institute-Hospital – is a bilingual, world-leading destination for brain research and advanced patient care. Since its founding in 1934 by renowned neurosurgeon Dr. Wilder Penfield, it has grown to be the largest specialized neuroscience research and clinical center in Canada, and one of the largest in the world. The seamless integration of research, patient care, and training of the world’s top minds make The Neuro uniquely positioned to have a significant impact on the understanding and treatment of nervous system disorders. It was the first academic institute in the world to fully adopt Open Science, to help accelerate the generation of knowledge and discovery of novel effective treatments for brain disorders. The Neuro is a McGill University research and teaching institute and part of the Neuroscience Mission of the McGill University Health Centre. For more information, please visit www.theneuro.ca
ELSE PRESS RELEASES FROM THIS DATE:
About The Study: This study including 202,000 veterans with COVID-19 and 202,000 matched uninfected veterans found that outpatient use increased significantly in the month after infection, then attenuated but remained greater than the uninfected cohorts’ use through 12 months, which suggests that there are sustained impacts of COVID-19 infection.
Authors: Matthew L. Maciejewski, Ph.D., of the Durham Veterans Affairs Health Care System in Durham, North Carolina, is the corresponding author.
To access the embargoed study: Visit our For The Media ...
About The Study: The findings of this study including 133 adults with severe obesity suggest that bariatric surgery was associated with health benefits two years after surgery. Bariatric surgery was associated with improved cognition and general health and changed blood vessel efficiency and cortical thickness of the temporal cortex. These results may improve treatment options for patients with obesity and dementia.
Authors: Amanda J. Kiliaan, Ph.D., of the Radboud University Medical Center, Donders Institute for Brain, Cognition, ...
A technique can determine for the first time how frequently, and exactly where, a molecular event called “backtracking” occurs throughout the genetic material (genome) of any species, a new study shows.
Published online February 9 in Molecular Cell, the study results support the theory that backtracking represents a widespread form of gene regulation, which influences thousands of human genes, including many involved in basic life processes like cell division and development in the womb.
Led by researchers from NYU Grossman School of Medicine, the work revolves around genes, the stretches ...
BOSTON, MA — A new study led by researchers at the Harvard Pilgrim Health Care Institute has found that reducing out-of-pocket costs for patients with diabetes, especially those considered low income, can be one step toward improving health outcomes.
The study, “Acute Diabetes Complications After Transition to a Value-Based Medication Benefit,” was published in the February 9 edition of JAMA Health Forum.
Treatment with antidiabetic agents, antihypertensives, and lipid-lowering ...
An international research project led by the Martin Luther University Halle-Wittenberg (MLU) delves into the next generation of seasonal thermal energy storage systems. The "INTERSTORES" project is investigating optimal ways to construct new types of storage systems and integrate them into energy systems. The collaboration between science and industry will receive nearly eleven million euros as part of "Horizon Europe", with around 1.5 million euros going to MLU. The aim is to improve the ...
L'Hospitalet de Llobregat, February 8th, 2024. Once they enter the body, drugs, apart from carrying out their therapeutic function, are biochemically transformed by the action of the metabolic machinery, a process that facilitates their expulsion. This biotransformation results in a gradual disappearance of the drug, which is converted into its metabolites. These, in turn, can reach high concentrations in the body and also show a biological activity that may be different from that of the original drug. That is, the metabolites and the drug coexist ...
Hot water tanks, washing machines, kettles: limescale forms in every domestic appliance that comes into contact with (hot) water – especially in areas where the water is hard, meaning high in calcium. Often the only thing that helps is to use vinegar or a special descaler to dissolve the rock-hard deposits and restore the appliance’s functionality.
This is a nuisance in households – and an expensive problem in thermal power stations, for example those that generate electricity, where the formation of limescale is known as fouling. Heat exchangers are particularly prone to limescale, which greatly reduces the efficiency of the systems: ...
OAK BROOK, Ill. – An innovative breast imaging technique provides high sensitivity for detecting cancer while significantly reducing the likelihood of false positive results, according to a study published today in Radiology: Imaging Cancer, a journal of the Radiological Society of North America (RSNA). Researchers said the technique has the potential to offer more reliable breast cancer screening for a broader range of patients.
Mammography is an effective screening tool for early detection of breast cancer, but its sensitivity ...
Children’s Hospital Colorado (Children’s Colorado) is announcing its recent acceptance into the Pediatric Heart Network (PHN), a collective of leading hospitals working to improve outcomes and quality of life for children – and more recently adults – with heart disease. The hospital’s Heart Institute will become one of nine clinical research centers across North America selected to be a part of this national network.
“We are thrilled to be accepted as a new core site for the Pediatric Heart Network,” said Shelley Miyamoto, MD, professor and Jack Cooper Millisor Chair of Pediatric Cardiology ...
Scientists have developed a new AI tool that maps the function of proteins in a cancerous tumour, enabling clinicians to decide how to target treatment in a more precise way.
In cancers such as clear cell renal cell carcinoma (ccRCC), responses to existing treatments are different for each patient, making it difficult to identify the right drug treatment regime for each patient.
For example, cancer therapeutic Belzutifan has recently been approved to treat ccRCC, but only has a response rate of 49% in patients with the most common form of the condition.
To understand better why some patients respond better than others, researchers from the Universities of Bath and Nottingham studied ...
LAST 30 PRESS RELEASES:
[Press-News.org] Making AI a partner in neuroscientific discovery
New paper argues that Large Language Models can reveal breakthroughs humans alone cannot