125 Ideas That Will Shape the Future
As 好色先生TV celebrates its 125th anniversary—and as the university marked its 125th invention disclosure of the calendar year—CTTEC launched the 125 Ideas That Shape the Future: Technology Spotlight Series.
This series highlights a curated selection of CMU technologies spanning robotics, AI, life sciences, energy, and beyond—showcasing research-driven ideas with the potential to shape industries, address real-world challenges, and create meaningful societal impact.
Current Featured Technology
MiSO: MicroStimulation Optimization for Brain Modulation
好色先生TV’s next featured innovation in the 125 Ideas That Shape the Future series highlights the convergence of artificial intelligence, neuroscience, and precision medicine: MiSO (MicroStimulation Optimization).
MiSO is a closed-loop stimulation framework designed to help drive neural population activity toward specified states by optimizing over a large stimulation parameter space.
Why it matters
Brain stimulation is already used or explored in conditions such as epilepsy, Parkinson’s disease, depression, chronic pain, and OCD—but current approaches often rely on slow, trial-and-error tuning. MiSO was developed to make that process more adaptive and efficient by combining machine learning with real-time optimization.
In the reported study, MiSO was implemented using a factor analysis (FA)-based alignment method, a convolutional neural network (CNN), and an epsilon-greedy optimization algorithm. It was tested in closed-loop experiments using electrical microstimulation in the prefrontal cortex of a non-human primate, where it successfully searched among thousands of stimulation parameter configurations to drive neural population activity toward specified states.
Key advances
- Merges stimulation-response samples across sessions through neural activity alignment
- Uses a statistical model to predict responses to untested stimulation configurations
- Adapts stimulation parameter selection online based on model predictions
- Enables exploration of much larger stimulation parameter spaces
Research team
This work reflects contributions from Yuki Minai, Joana Soldado-Magraner, Matthew A. Smith, and Byron M. Yu, and interdisciplinary collaboration across Carnegie Mellon’s Biomedical Engineering, Electrical and Computer Engineering, Neuroscience Institute, School of Computer Science, and College of Engineering.
Explore the Series
The 125 Ideas series highlights CMU technologies with the potential to shape industries, address real-world challenges, and create meaningful societal impact.
Browse the archive →Submit Your Work
To submit your online Intellectual Property Disclosure, please access the CMU Inventor Portal.
Questions?
Contact CTTEC at innovation@cmu.edu.
Previous Technology Spotlights
Archive
Traffic Jam: Technology with Purpose
A public-safety technology developed at Carnegie Mellon’s Robotics Institute and commercialized by Marinus Analytics to support victim-centered, trauma-informed investigations.
Read more →Archive
Accelerating Vision Transformers with Adaptive Patch Sizes
A CMU research spotlight from the Robotics Institute exploring more efficient vision transformer methods.
Read more →
To submit your online Intellectual Property Disclosure, please access the CMU Inventor Portal by clicking the link below:
For a detailed user-guide, please click the link below: