The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Using AI to Diagnose and Treat Neurological Disorders with Archana Venkataraman - #312
Oct 28, 2019
Archana Venkataraman, a John C. Malone Assistant Professor at Johns Hopkins University, specializes in using machine learning to tackle neurological disorders. In this discussion, she shares insights on predicting clinical severity in autism using fMRI data. Venkataraman highlights innovative tools for enhancing emotional understanding in individuals with autism and touches on combining EEG and MRI data for diagnoses. She emphasizes the transformative potential of AI in clinical decision-making and improving patient outcomes.
AI Snips
Chapters
Transcript
Episode notes
Early Engineering Influence
- Archana Venkataraman's parents, both engineering professors, instilled a strong educational focus.
- This led her to target MIT from fourth grade, ultimately pursuing signal processing and its application to neuroscience.
Resting-State fMRI and Biomarkers
- Resting-state fMRI reveals functional brain systems through correlation patterns in passive brain activity.
- This data helps predict neurological disorders and discover biomarkers by identifying predictive coactivation patterns.
Predicting ASD Severity
- Venkataraman's team uses dictionary learning and predictive modeling to forecast autism spectrum disorder (ASD) severity.
- This approach allows interpretable biomarker discovery and performs on par with other state-of-the-art methods.