Nathan Intrator
Edit My PageTitle: Adjunct Professor
Department: Brain Science
Nathan_Intrator@Brown.EDU
+1 401 863 3391, +1 401 863 2585
Biography
Professor Nathan Intrator is an international scholar in neural computation, machine learning and pattern recognition and has authored/co-authored more than 120 refereed scientific publications.
His significant contributions include model estimation, validation, selection, interpretation and discrimination techniques for high dimensional problems with a small amount of observed data. He is best known for his contribution to the Theory of Cortical Plasticity, brain imaging, decisions from multiple experts, and improving sonar system accuracy.
Professor Intrator has been supported by US, European and Israeli federal agencies on projects related to brain imaging and Brain-Machine Interface. His applied research led to several patents and applications and the founding of three companies in biomedical signal analysis, sonar imagery and homeland security.
Research Description
Current Research Focus:
Improving EEG information extraction for diagnostics and continuous brain monitoring. Improving brain machine interface using a single EEG electrode. Analyzing and modeling the computation properties of bio-sonar animals.
Research Methods:
Development of machine learning time/frequency methods for EEG localization and signal decomposition for the purpose of information extraction, detection and classification. Relying on concurrent fMRI/EEG for improving spatial and temporal brain scanning resolution. Development of novel signal processing and machine learning methods for creation of super-resolution and super-accuracy in bio-sonar, inspired by bat and dolphin sonar research.
Projects in the lab include:
1. Single EEG feature extraction for epilepsy, attention and other brain pathologies.
2. Development of neuro-feedback techniques.
3. Clinical validation of novel EEG information extracted with a single electrode.
4. Analysis of EEG networks of activity from EEG/fMRI concurrent recording.
5. Modeling of super resolution and accuracy from real and simulated bat recorded data.