Using machine learning, researchers are able to use data from the brain to uncover deeper insights and apply this new knowledge in clinical settings. The findings will be presented Monday, November 13, 2-3 pm EST at Neuroscience 2023, the annual meeting of the Society for Neuroscience and the world’s largest source of emerging news about brain science and health.
Machine learning is a branch of artificial intelligence (AI) that enables computers to analyze data in increasingly complex ways. Researchers use algorithms with adaptive models that can change and evolve without outside intervention. In neuroscience research, machine learning can study large datasets with information from the human brain and apply models to predict outcomes based on that information.
New findings show that:
- Researchers have successfully identified activity between the thalamus and motor-related regions as correlated with depressive symptoms, applying neuroimaging techniques to identify abnormal brain regions in psychiatric disorders. (Ayumu Yamashita, Advanced Telecommunications Research Institute International)
- Machine learning models using patient data can predict progression from mild cognitive impairment (MCI) to Alzheimer’s disease (AD); Their study focused on the use of KPCA (Kernel Principal Components Analysis). (Nikita Goel, University of Southern California)
- A new application of artificial neural networks can uncover how cells in the visual cortex interpret light signals from the retina and translate that data into vision. (Dan Butts, University of Maryland)
- The researchers applied machine learning models to deep brain recordings and combined them with traditional control theory methods to create a tool for predicting the targets of deep brain stimulation (DBS). This optimization may help the treatment for dystonia in children. (Maral Kasiri, University of California, Irvine)
““Advances in AI and machine learning are transforming brain research and clinical treatment.” said Terry Sejnowski, Francis Crick Professor at the Salk Institute for Biological Studies and distinguished professor at UC San Diego, who will moderate the press conference. “Brain recordings create huge datasets that can be analyzed with machine learning. Predictive modeling, machine-brain interfaces and neuroimaging/neuromodulation are areas of particular promise in developing new therapeutics and treatment plans for patients.”
This research was supported by national funding agencies, including the National Institutes of Health and private funding agencies. Learn more about AI and brain research BrainFacts.org.
Monday, November 13, 2023
2-3 pm EST
Walter E. Washington Convention Center, Room 202B
Summary of the press conference
- These presentations explore the wealth of data that can be mined and analyzed using machine learning. These topics range from predictive modeling that detects brain activity for disorders such as Alzheimer’s disease (AD) and depression (the first two studies) to analyzing the brain’s complex visual system and applying deep brain network models to advance therapy (the third and fourth studies). ), respectively).
Unsupervised feature selection methods strongly correlate resting state functional connectivity with major depressive disorder.
Ayumu Yamashita, [email protected]abstract PSTR099.15
- Researchers used neuroimaging techniques combined with machine learning to identify brain biomarkers. To determine how accurately and robustly these techniques identify abnormal brain regions in psychiatric disorders, the researchers used three types of feature selection methods related to major depressive disorder (MDD).
- Among the three methods, they found that the supervised-based feature selection method robustly extracted functional associations with large effect sizes in the test data. Using MRI data from 1,162 patients, including 334 with depression, the researchers extracted coordinated activity between brain regions (particularly the thalamus and motor-related regions) that correlated with depressive symptoms.
Predicting future decline from mild cognitive impairment to Alzheimer’s disease with machine learning and 3D brain MRI
Nikita Goel, [email protected]abstract NANO03.07
- Predictive models use machine learning to try to predict who will develop brain disorders like AD. The researchers used a range of models to see which models could predict mild cognitive impairment (MCI) that would progress to AD.
- The researchers used MRI scans with demographic and genetic data on more than 2,448 people. The most efficient model they developed used kPCA (kernel principal component analysis) to generate new features by combining the scan data with basic patient data alongside an AD risk gene.
Biologically constrained deep neural networks for parsing visual computations performed in primary visual cortex.
Dan Butts, [email protected]abstract PSTR149.16
- Researchers used a novel implementation of artificial neural networks called convolutional deep neural networks (CNNs) to learn how cells in the brain’s visual cortex interpret light signals captured by the retina.
- Using machine learning, researchers are beginning to understand how information about the visual world is carried across brain cells in the visual cortex.
Identifying effective targets for deep brain stimulation: system identification using a simple recurrent neural network
Maral Kasiri, [email protected]abstract PSTR154.16
- Clinicians face challenges with deep brain stimulation (DBS) in determining optimal targets within the brain. Thus, treatment can be tailored to a specific patient to better treat movement disorder symptoms.
- Based on the underlying brain signals of brain regions (eg, basal ganglia and thalamus) of children with dystonia, machine learning can be applied to develop models of deep brain networks combined with traditional control theory and statistical methods. The end result is a tool to identify the most optimal DBS target for patients.